JPH06508036A - Production and recovery of recombinant neurotrophins - Google Patents

Abstract

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

Description

[Detailed description of the invention] Production and recovery of recombinant neurotrophins This application was filed June 12, 1991. This is a continuation in part of the pending application number 07/715.185, which is hereby incorporated by reference in its entirety. The present invention, which is incorporated herein by reference, relates to neurobiology, molecular biology and protein biology. There are also new developments in the field of biochemistry, especially regarding the 22-bit trophin and its production and recovery methods. Lotrophins are a group of proteins involved in the functionality of the nervous system. they are nerve cells Stimulates the growth of vesicles and supports their survival during embryonic development. So far, research have identified four types of neurotrophins: nerve growth factor (NGF); (Ullrich et al., 1983. Nature 303:821 -825), brain-induced neurotrophic factor (BDNF) (Leibrock et a l,, 1989. Nature 341:149-152), neurotrophic factors Child 3 (NT-3) (Maisonpierre et al, 199 0.5science247:1446-1451) and neurotrophins- 4 (Hallbook, etal, 1991. Neuron 6:84 5-858), nerve growth factors affect the peripheral and sympathetic nervous systems and the brain's cholinergic Essential for neuron development and survival. Currently, this is Alzheimer's high. It is currently being tested for application to Mer's disease. BDNF plays a role in the production of sensory neurons in the central nervous system. It promotes survival and is seen as a promising treatment for Parkinson's disease. NT-3 and NT-4 are They have been recently discovered and their biological role is being studied. Ru. Similar to NGF, NT-3 acts on sensory and sympathetic neurons. It seems that

Natural human NGF has three subunits: α, β and γ. βsa Only the buunit exhibits neurotrophic activity. β-NCF is a pre-pro protein. It is secreted from cells and processed into its mature form. play plow The peptide contains 187 amino acid residues. During processing, play and the bro sequence are removed, and the two C-terminal amino acid residues are also removed. This is it A mature protein consisting of 118 amino acid residues is produced. The mature protein is It has six cysteine residues that form three disulfide bonds. Furthermore, this The protein contains many basic amino acids that provide a positive charge at neutral pH. .

Structurally, neurotrophins clearly evolved from a common ancestral gene. (Hyman et al., 1991゜Wo 91) 103568). For example, the amino acid of the β polypeptide of human NGF (hNGF) The acid sequence is 90% homologous to mouse and bovine NGF. 3 types of human news Lotrophin, hNGF, hBDNF and hNT-3 are 60% at the amino acid level. share % homology.

Six cysteine residues are also conserved in all known neurotrophins. This suggests that disulfide bonds are important for function.

Natural sources of neurotrophins are limited. The submandibular gland of mice is rich in NGF. However, it is difficult to isolate and the amount recovered is small (Hyman et al. al., 1991.　Wo　91103568). del IaValle (EP O 333 574, 1989) developed human embryos by chromatographic fractionation. reported that a small amount of hNGF was isolated from the disc.

Thus, researchers have identified readily available sources of biologically active neurotrophins. In order to secure a supply source, he began to pursue molecular genetics. NGF, BDNF and All DNA sequences encoding NT-3 have been isolated (Ullrich et al. tal, Nature 303:821-825; Hyman et al., 1991. WO91103568; Hohn et al. WO91103569; and Kaisho et al, FEBS Le tters 266:187-191). Researchers have discovered that neurotrophins These sequences were used to transform animal and non-animal hosts to achieve this goal.

Kanaya et al. (1989, Gene 83:65-74) reported that mature hNGF The encoding DNA sequence is a sequence encoding the yeast α-mating factor pre-pro sequence. fused into. When this is expressed, the culture supernatant is recognized by α-hNGF antibody. Contains protein that can be absorbed. However, partially purified proteins have low It showed neurotrophic activity. Moreover, the expression level was also low. Kanaya et al. These results suggest that the yeast system removes two extra C-terminal amino acid residues during the maturation process. Either it couldn't be done or there was a problem with the folding of hNGF. It is said that this is probably for the sake of

Chan et al. (EP O370171, 1990) developed mature hNG in insect cells. produced F. They transferred the gene for Prepro hNGF to the baculovirus DNA. When fused to the polyhedrin promoter and leader sequence of A, 6 μg/m reported that 1 hNGF was produced. However, the physical characterization of this product is It wasn't done.

Researchers have also demonstrated human NGF, BDNF and NT in mammalian expression systems. -3 was expressed. Bruce and Helnrich (1989゜Neurob iology of Aging 10:89-94) in COS cells. A DNA sequence encoding the complete precursor of hNGF is expressed and in conditioned medium. A dimer of hNGF was detected. However, they found that playpro hNGF The efficiency with which it was converted to mature hNGF was not investigated. Kakinuma et al. (IEP) 0414151.1991) expressed active hNGF in CHO cells.

Hyman et al. (WO91103568, 1991) reported that hBD in CHO cells NF was expressed. Nakahama et al. (EPo 386752, 1990) and Hohn et al. (WO91103569, 1991) showed hN in CO8 cells. ``r-3 was expressed.

Mammalian systems provide the most natural environment for the production of mammalian proteins . However, producing large amounts of protein using these systems is extremely expensive. . Therefore, there is a need to develop systems that are less expensive and more productive. be. One such system is Escherichia coli (E. coli).

Researchers attempted to express the 22-bit trophin in E. coli. However, it was not very successful. Gray and Ullrich (EP 01213 38, 1984) constructed an expression vector encoding N-methionyl-hNGF. and this gene was expressed in E. coli.

They reported that hNGF was identified by immunodetection in Western blots; The protein has not been isolated and no biological activity has been demonstrated.

Hu and Neet (1988, Gene: 57-65) An attempt was made to express SNGF. They have a DNA sequence encoding adult mouse NGF. The N-terminal amino acid of the natural protein, serine, is cloned. Replaced with thionine. They converted this DNA sequence into a temperature-inducible lambda PL promoter. The vector was inserted into a plasmid containing the vector. This system allows other foreign proteins to be extracted whole-cell. It is expressed at a rate of 10-25% of the cell proteins. they express this gene, and by ammonium sulfate precipitation followed by dialysis against acetate buffer. NGF was isolated. However, when tested in a bioassay, this system showed only Only 0,0005-0.1% NGF was produced. The authors predicted that In contrast, very low yields may be due to NGF toxicity to cells, mRNA instability, or Poor translation efficiency or disulfide bond of regenerated oxidized protein I guessed that it was due to a mismatch.

Iwai et al. (1986, Chem, Pharm, Bull, 34:472 4-4730) is E.

reported the synthesis of an hNGF-encoding gene with codons preferred by E. coli.

They expressed this gene directly as N-methionyl-hNGF or in human growth proteins. It was expressed as a fusion protein with Lumon. Direct expression is fusion protein expression It was only one-fourth as efficient. They analyzed the protein by 5DS-PAGE. quality but did not isolate them by any other method.

Dicou et al. (1989, J, Neurosci, Res, 22:13- 19) used fragments of mouse pre-pro-NGF and hNGF in β- When expressed as a fusion protein with galactosidase, it inhibited proteases. It was found that the addition of harmful agents improved the yield.

In summary, previous attempts to express neurotrophins in E. coli The result is cell death, low-level protein accumulation, and virtually non-bioactive proteins. This resulted in the development of dark texture.

3. Summary of the invention The present invention provides biologically active recombinant neurotrophins from non-animal host cells. provide a method for producing and recovering How to produce recombinant neurotrophins , a protein operably linked to a DNA sequence encoding a neurotrophin. From culturing host cells transformed with recombinant DNA molecules containing current regulatory sequences. It has become. When neurotrophins are expressed directly in the cytoplasm, host cells Protease-deficient mutants, according to the best methods we know, are heat-shocked. A nodal gene mutant strain is preferred. In bacterial hosts, neurotrophin genetics Place the offspring under the control of a regulatable promoter and the cells reach late logarithmic phase It is preferable that expression is not induced or suppressed until Also, neurotrophins Inserting a DNA sequence encoding a signal peptide upstream of the encoding DNA sequence Neurotrophins can then be secreted from cells during production.

of biologically active recombinant neurotrophins produced by host cell cultures. The recovery method consists of neurotrophication in a solution containing strong denaturing agents but essentially free of reducing agents. It consists of solubilizing lophine. According to one aspect of the invention, this solution is an effective amount of protein to prevent the degradation of neurotrophins by proteases. It further contains a rotease inhibitor. Other aspects of the invention further include the following steps: : exchanging a strong denaturing agent for a weak denaturing agent; neurotrophs in a non-denaturing environment Contains basic amino acids or equivalents at concentrations effective to maintain the solubility of the adjusting the solution so that the neurotrophin is isolated from other molecules in the solution; purification; removal of weak denaturants; and completion of purification in the absence of denaturants. consists of some or all of;

In yet another aspect of the invention, the bioactive form expressed in E. coli and The neurotrophin molecule recovered in this state is neurotrophin-4. E, In a preferred embodiment of the present invention directed to the expression of bioactive NT-4 in E. coli NT-4 is substantially as shown for hNT-4 in Figure 11 (SEQ ID NO: 22). encoded by a nucleic acid molecule that contains a sequence of and sequences that are about 70% homologous.

4. Explanation of drawings FIG. 1 is a schematic diagram of plasmid pRPN133. Solid line is derived from pBR322 represents the DNA sequence containing the origin of replication (ORI) and the β-lactamase gene (amphi Syrin resistance') (Ap) is shown. A unique feature of this plasmid is the box The regions are indicated with arrows indicating the direction of transcription or replication. Pi, is Lambda PL promo Indicates the data.

rbsl is phage T7φ1. l wild-type promoter and liposome binding site It is. The hNGF gene encodes the mature polypeptide, with the second amino acid residue Serine is replaced with threonine. cI857 is a heat-inactivable lambda replenisher. This shows the sir gene.

Figure 2 shows (A) E8 chick embryo dorsal ganglion (DRG) explants and (B) dissection. E by isolated E8 DRG, against recombinant human NGF produced in coli A dose-response curve is shown.

Figure 3 shows wild-type LamB (Figure 3A, SEQ ID NO: 1) and synthetic LamB, That is, LamB1 (FIG. 3B, SEQ ID NO: 2), LamB2 (FIG. 30, SEQ ID NO: 2) 3), LamB5 (Figure 3D, SEQ ID NO: 4) and LamB4 (Figure 3E, The nucleotide and amino acid sequences of the signal sequence of SEQ ID NO: 5) are shown.

Figure 4 shows the signals of modified LamB signal sequences, LamB1 and LamB5. Figure 3 illustrates null sequence processing kinetics. We have the corresponding LamB-hBDNF plus The BL21/DH3 strain containing Mido was cultured. In this strain, T7 RNA polymer The gene encoding the enzyme has been inserted into the chromosome, and the transcription of the lac promoter under control (Studier and Moffatt, J. Mo1. Biol, 189:113-30). Apply IPTG to proliferating cells for 10 minutes After a while, T7 RNA polymerase is synthesized. Then rifampicin When added to 0.2 mg/ml, E. coli RNA polymerase transcription by T7 RNA polymerase occurs. the result , late T7 φ1 of rbs2. Immediately near the promoter and liposome binding site Selective transcription of the downstream LamB-hBDNF gene occurs. 35 in cells Pulse 3-methionine for 30 seconds, then check with excess cold methionine. I did it. Samples of cultures were taken at the indicated times after the chase and labeled with labeled proteins. Proteins were analyzed by 5DS-15% PAGE and fluorography. Processin The analysis was carried out by densitometric scanning of the precursor and mature forms of LamB-hBDNF. was measured.

FIG. 5 is a schematic diagram of pRPN121. The solid line is DNA derived from pBR322 represents the sequence and indicates the origin of replication (ORI) and β-lactamase gene (Ap). Ru. Characteristics unique to this plasmid are indicated by boxed regions and arrows indicate transcription or duplication. Indicates the direction of manufacture (ORI).

T with some nucleotide substitutions from the wild type, designed to make 7φ1. 1 promoter and T7φ1. 1 Liposome binding site is shown. L amB2 is a signal sequence. hBDNFmyc is the structural gene.

Figure 6 shows hBDNFmyc Fast S-Sepha0-S1 (Fast S- 3EP) IARO8E■l) Shows the protein profile of fractionation. DEAE After chromatography (7) W31101qF-/RPN121 extract, 1.6 cm x 6.5 cm of Fast S-Sepharose as described. ■ Column with 7M urea, 50mM histidine, pH 5° and 0.1mM EDTA. fractionated. Both fractions shown in the figure were analyzed by 5DS-15% PAGE, and proteins were detected. Visualized with Coomassie staining. Fractions 26-3o were pooled.

Figure 7 is a summary of the purification method. Propagation of W31101'P-/pRPN121 strain , induced, and extracted as above. The pooled column fractions were diluted with 15% Analyzed by 5DS-PAGE on Rylamide gel and stained with Coomassie blue. Ta. Lane 1. DEAE, L/-n2. First S-Sepharose o1; 3. Fast S-Sepharose ■2; lane 4. C4 reverse phase HPLC, molecular weight It shows the standard.

Figure 8 shows DRG neurite formation in E8 chicken embryos by purified hBDNFmyc. Figure 3 shows dose-response curves for long-term stimulation. China's activity of hBDNFmyc compared to recombinant hBDNF purified from a human hamster ovary cell line. Both Tampa The protein was purified to more than 95% by C4 reverse phase HPLC.

Figure 9 shows E8 chicken embryos treated with recombinant hBDNF purified from E. coli. Figure 3 shows a dose-response curve for stimulation of DRG neurite outgrowth. hBDNF is C 4. Purified to 95% or higher by reverse phase HPLC.

FIG. 10 is a schematic diagram of pRPN149. The solid line is the DN derived from pBR322 A sequence, indicating the origin of replication (ORI) and the β-lactamase gene (Ap). be. Characteristics unique to this plasmid are indicated by boxed regions and arrows indicate transcription or Indicates the direction of replication (ORI).

1acUV5 is the promoter. rbs2 is decrypted to create convenient restriction sites. engineered T7φ1.1 protein with some nucleotide substitutions from the wild type. Motor and T7φ1.　L liposome binding site is shown. LamB1 is Signa This is a file array. hBDNF is the structural gene.

Figure 11 shows isolated human genomic phage clones encoding human NT-4. This is a partial DNA sequence of 7-2 (SEQ ID NO: 22; ATCC accession number Near 50). 70). Putative hNT-4 encoded by this genomic clone 7-2 Proteins are represented by one-letter amino acid symbols (SEQ ID NO: 23). box The boxed region represents the putative cleavage site of hNT-4 preprotein. The arrow is Residues conserved in the sequence are shown. The underlined area (N-R-3) is N- Represents a consensus sequence for glycosylation. The circled area represents the starting methionine . A splice acceptor site exists at base pairs 461-462 (AG), which Represents the 3' end of tron.

FIG. 12 is a schematic diagram of pRG91. The solid line indicates the DNA sequence derived from pBR322. The origin of replication (ORI) and the β-lactamase gene are shown. This base Characteristics specific to the vector are indicated by boxed areas. 1acUV5 is a promoter be. rbs2 is the phage T7φ1.1 promoter and T7φ1.1 lipo It is a somatic binding site. LamB2 is a signal sequence.

Plasmid pRG91 (Regeneron Pharmaceutical s) Expression of recombinant proteins and Escherichia coli ), based on pBR322, designed for its secretion into the periplasmic space of vector. This vector was constructed using the EcoRI and NruI systems of pBR322. The strong regulated 1acUV5 promoter inserted between the restriction sites and then contains the phage T7φ1.1 promoter and liposome binding site.

These regulatory elements govern the expression of the LamB2 signal sequence, and this signal sequence The recombinant protein gene sequence can be fused to the recombinant protein. The only Nru I By deleting the DNA sequence between the Pvu II restriction site and the The number of copies of the code increases. This plasmid confers ampicillin (Ap) resistance. Ru.

Figure 13 shows the dose-response curve of hNT-4 expressed in E. coli. vinegar. Soluble proteins from induced cultures of strain RFJ26 containing plasmid pRG173 Protein extracts were assayed on E8 dorsal root plate meridian explants. Back of this test method Ground activity is 0. It was 2 units.

Figure 14 shows the effect of hNT-4 on CAT activity. Plasmid pRG17 Partially purified extract from induced culture of RFJ26 strain containing 3 and enriched in motor neurons. untreated control (C-NT) and buffer control (C-buffer). ), the CAT activity after 48 hours was increased by 3.6 times (at 1:20 dilution). Ta. E.

E. coli extracts were prepared as described below prior to treatment of motor neuron-enriched cultures. The mixture was applied to a Sepharose-S column as follows.

(Margins below this page) As used herein, the term "neurotrophin" refers to Refers to any natural substance belonging to the trophin family.

It shares amino acid sequence homology with known neurotrophins and has six characteristic Contains naturally occurring proteins that conserve characteristic cysteine residues. (These proteins Some of these substances may structurally belong to the neurotrophin family, but they have more than just neurotrophic activity. It may also exhibit other biological activities. ) The term “neurotrophin” is or derived from natural neurotrophins or produced using them as templates. It also refers to an artificial neurotrophin with a modified amino acid sequence. As an example obtained from two different trophins (e.g., NGF and NT-3). different amino acid sequences or the same neurotrophin from different species ( For example, a kit consisting of amino acid sequences obtained from hBNDF and porcine BDNF). Melanurotrophin, whose gene sequence has point substitutions, additions or deletions Una neurotrophin, a neurotrophin derived from the Una neurotrophin sequence. [For example, the stop cod of the natural gene (“full-length neurotrophin”) The two carboxyl-terminal amino acid residues encoded by the codon immediately preceding the [neurotrophins], as well as neurotrophins that exhibit neurotrophic activity. Fragments (e.g., those with changes or deletions in the first six amino acids) Ru. These examples are provided for illustrative purposes and are not intended to limit the scope of the definition. It should not be interpreted as

The present invention provides methods for producing and recovering biologically active recombinant neurotrophins. provide a method for The present invention also relates to purified Also provided are homogeneous recombinant neurotrophins. As used herein “Recombinant protein J refers to host cells that have been transformed with a recombinant DNA molecule. refers to a protein expressed from said recombinant DNA molecule. “Recombinant D "NA molecules" are DNA sequences obtained from different sources linked together. It is a hybrid molecule consisting of Sambrook et al. (1989) , Molecular Cloning + A Laborat.

ry Manual, Co1d Spring) labor atory Press, Cold Spring Harbor) is a recombinant It describes many commonly used techniques related to DNA engineering.

To produce recombinant neurotrophins according to the invention, neurotrophins has an expression control sequence operably linked to the DNA sequence encoding the protein. It is necessary to culture non-animal host cells transformed with recombinant DNA molecules. It is essential. Conditions in which an expression control sequence can function in a DNA sequence encoding a polypeptide This means that the expression control sequence is linked to the transcription and translation of the DNA sequence. means to direct and promote. For culturing transformed host cells The cells are incubated under culture conditions suitable for cell growth and expression of the DNA sequence. It is necessary to incubate.

DNA sequences encoding 22bittrophin can be obtained from several sources. Can be done. In the literature, there are four known human neurotrophins [i.e., hN G F (Gray et al., EP 0121338.1984), h B D NF (Hyman et al., Wo 91103568), h NT-3 (Ho hn et al.

Wo 91103569) and African frog N T-4 (Hal 1 books etc.

1991, Neuron 6:845-858)] and many non-human - The DNA sequence for lotrophin has been disclosed. Suitable for PCR amplification In order to obtain guidelines for the production of oligonucleotide primers, these It is preferable to refer to the sequence of . Genomic libraries are suitable for PCR templates. It is a hit NA source. It is also known to express neurotrophin mRNA. Also useful are cDNA libraries of cells that have been developed. For example, human fetal brain mR The NA cDNA library contains sequences for hBDNF. Also, cDNA and genomic libraries using any method known in the art. DNAs encoding neurotrophins were identified by screening A sequences can also be identified and isolated. Alternatively, normal DNA automatic synthesis Synthetic or semi-synthetic genes can also be created using synthetic equipment.

Research Sonic discovers DNA sequence encoding new neurotrophin There is no doubt that the method of the present invention produces such trophins and It will also be useful for recovery.

More specifically, generally speaking, the present invention also relates to NT-4 or derivatives thereof. or a recombinant bacterium containing a nucleic acid encoding a fragment of NT-4 or its fragment. The derivative or fragment is grown under conditions that allow expression of the derivative or fragment; NT by collecting the produced NT-4 or its derivative or fragment. -4 or a derivative or fragment thereof. Preferred implementation In this case, NT-4 is human NT-4. In another embodiment, the chimera An NT-4 derivative that is a protein or fusion protein is produced. most suitable In certain embodiments, the produced NT-4 or a derivative or fragment thereof has biological activity. That is, it is known in the art or described herein. Any method described herein (e.g., growing on E8 DRG explants) abilities that can be promoted, or purified exercise news as described in Section 9 below. (In vitro test method for the ability to stimulate CAT activity in culture) may exhibit one or more of the known functional activities of NT-4 when measured by .

In certain embodiments of the invention, the sequence encoding human NT-4 is expressed in an expression system and by use of purification methods as disclosed below. Useful amounts of human NT-4 are produced. Coding human NT-4 expressed in this way The nucleic acid to be coded is nucleic acid pRG173 (ATCC accession number 75131) or HG 7-2 (ATCC Accession No. 75070) or shown in Figure 11. (SEQ ID NO: 22). In addition, such nucleic acids are known in the art. can be isolated by any method known in the art or by the procedures described below. can. That is, known NT-4 sequences or known neurotrophin-derived 5° and 3′ oligos representing all possible codons corresponding to conserved amino acid sequences A mixture of nucleotides is used as a primer in the polymerase chain reaction (PCR). used. Human (or other mammalian) cDNA or genome sequence PCR products are isolated as a result of the primary and secondary PCR amplification reactions of the library. However, by using this PCR product as a 32p labeled probe, NT- It is possible to isolate a full-length cDNA or genomic clone encoding 4. Wear. As used herein, the term "human neurotrophin-4" The word is African frog NT-40 (Allbook et al., 1991. Neuron 6:845-858) [e.g. species but homologous (e.g., having at least about 70% homology) neurotrophin molecules].

There are several methods in the literature that are useful for expressing DNA sequences in transformed non-animal hosts. A variety of expression control sequences have been disclosed. As an example, in the case of bacteria, l ae system, trp system, TAC system, TRC system, λPL promoter, T7 late promoter , and the regulatory regions of the fd coat protein. Also, yeast In this case, phosphoglycerate kinase promoter, Ga14 promoter, H is promoter, alcohol dehydrogenase promoter, alkaline phosph atase promoter and alpha mating factor promoter. controllable Expression control sequences are preferred, especially temperature-inducible sequences for expression in E. coli. capable λP, Pa motor, 1acUV5 promoter and T7φ1, 1 promotor Motors are most preferred.

The literature also describes various expression vectors/hosts suitable for bacterial and fungal hosts. systems, such as plasmids, bacteriophages, cosmids and their derivations. The body is revealed. Examples of these systems include colFi for bacteria. pCRl, pBR322, pMB9, RP9, λ phage, M13 and fila Mention-like viruses are mentioned, and 2μ is mentioned for yeast. to bacteria Preferred plasmids are stable in the host cell and at a concentration of 20% per cell. It exists in a medium copy number of ~200 copies. The inventors have p Plasmids derived from BR322, such as Panayot atos, 1987, t! ngineering an efficient Expression System, in: Plasmids-A Pr actual Approach, ed, Herdy, K,, IRL Press, 0xford/Washington, D.C.) I used what was described. Among them, Ligieneron Pharmacy Cals Inc. (Regeneron Pharmaceuticals, Inc.) RPN-based plasmids as developed by and described in more detail below. is suitable.

The art also recognizes a number of non-animal vectors useful for expressing heterologous proteins. Hosts are known, including Escherichia coli, Bacillus sp., Streptomyces sp. Includes Tucharomyces and Pichia pastoris. In the present invention, Escherichia coli is suitable.

Expression of the 22-mouth trophin is achieved by culturing the transformed host cells. . The present inventors have demonstrated that neurotrophins expressed in bacteria act as proteins in cells. enzymes [particularly those induced as part of the “heat shock” response to foreign proteins] (Gaff et al., 1984. Proc, Natl, Aced, Sci, USA 81:6647-6651)]. I found it. Therefore, using protease-deficient mutants as host cells is preferred.

Expression of heat shock genes is controlled by heat shock regulated genes (e.g., http of E. coli). R gene). Mutants in the htpR gene are susceptible to heat shock. There is a lack of expression of the rotease gene, as well as its contribution to the heat shock response. Expression of other genes is also lacking. The present inventors have demonstrated that heat shock-regulated gene mutations Recombinant 22-bit trophin in mutants (particularly htpR-gene mutants) We have found that the Cl37 strain is preferable because we have found that the yield can be significantly improved if it is expressed. child The strain was developed by Alfred Go of the University of Barbados. ldberg) available from Prof. Goldbe Other suitable strains are also described in U.S. Pat. No. 4,758,512, such as It is listed. Another suitable E. coli strain is RFJ26.

Neurotrophins are toxic to the bacterial cells that express them. Therefore , to maximize yield, a high density bacterial culture (e.g. late phase It is preferred to induce neurotrophin expression only in cultures that are in several stages. Yes. We have demonstrated that the λPL promoter and the c1857 temperature-sensitive replenisher A sir-based inducible system was used. The production of neurotrophins is typically The incubation temperature was shifted to 42°C for 30 minutes, then for 3-20 hours. induced by continued incubation at 38-42°C. In addition, live Inducing expression for more than 16 hours in initially proliferating cells may result in station causes cell death.

If E. coli cannot accumulate neurotrophins, neurotrophin genetic This may be due to one or more properties of the child or protein. neuroto The structure of lofin mRNA, especially the structure close to the translation start site, facilitates efficient translation. It can be a hindrance. Alternatively, a neurotrophin may react directly with its mRNA. It may interfere with its own synthesis by reacting to E. coli, and may also interfere with E. May react directly or indirectly with some component in the production/transcription/translation machinery .

According to another embodiment of the invention, the neurotrophin is expressed intracellularly. Then, it is secreted into the periplasmic space of E. coli. This is a mature neuroto This is accomplished by fusing a signal sequence to the lofin. neurotrophy By fusing a signal sequence gene to the 5' end of the transgene, efficient translation can be achieved. a more advantageous nucleotide sequence is placed close to the translation start site, thereby Can result in high levels of neurotrophin accumulation. Moreover, the cell periphery Isolation of neurotrophins within the cavity allows them to be used for protein synthesis. Interference with any side sill components is prevented. It is also a neuroto It also prevents lophine from being affected by proteases in the side sills. Furthermore In addition, if mature neurotrophins are secreted into the pericellular space, protein It can also provide a more favorable environment for proper folding.

To provide a signal sequence, a DNA sequence encoding the 22-bit trophin is added. is within the same framework as the DNA sequence for neurotrophins and is relevant to the host cell. A fusion gene encoding a suitable signal or leader sequence is inserted 3′ from the 5° side. What is necessary is to produce a recombinant DNA molecule containing it toward the side. In the literature, Several signal sequences useful in the production of such recombinant DNA molecules have been described. There is. For example, LamB, 0mpA, and PhoA are useful in E. coli. (Denefle et al., 1985. Gene 85:499-510; Wong et al., Gene 68:193-203). In the present invention, LamB are preferred, and in particular modified LamB sequences that increase the translation efficiency of LamB mRNA. A GNAR arrangement is preferred. The present inventors have discovered that the remains associated with the modified LamB signal sequence. The gene was produced as follows.

That is, by replacing G or C with A or T, some of LamB A degenerate substitution was made at the third nucleotide of the codon. These substitutions are based on the LamB system. does not change the amino acid sequence of the peptide, but allows for possible changes in any secondary structure. reduce the number of possible hydrogen bonds. This region in LamB mRNA reducing the stability of any associated possible secondary structures. The inventors also found that E. coli codon usage model to better approximate the most frequently used codons. introduced codon changes based on

The inventors also demonstrated that in converting the LamB precursor protein to the mature protein, Modifications were made to the LamB signal sequence to increase processing efficiency. natural LamB has a hydrophobic core consisting of 10 amino acid residues. several large The hydrophobic core region, as suggested by mutational analysis of the S. enterica signal sequence. The length of the region can have a strong influence on signal sequence activity. Inventor et al. increases the length of the hydrophobic region by adding up to 10 hydrophobic amino acid residues By increasing the processing efficiency of the L a m B fusion precursor polypeptide, I found that it can be improved.

In this case less than 6 is preferred, and 4 is most preferred.

The type of hydrophobic amino acids added is not important, nor is it important to place them in the hydrophobic core region. The exact location of the addition is also not important. However, in the present invention, the hydrophobic region The tetrapeptide Leu-Ala-Val- It is preferable to add Leu (rLAVL") (SEQ ID NO: 6). In addition, Specific genes for modified amB signal sequences are described in Example 7. .

According to the invention, the host cells are harvested, lysed, the lysate centrifuged, and then Recombinant neurotoxins are extracted from host cell cultures by harvesting lysate pellets. Lofin is liberated. Typically, cell harvesting is performed by centrifugation. cell yield in 50mM-EDTA to prevent neurotrophin degradation after harvest. It is preferable to resuspend the cells. The cells are then used immediately or or frozen for later use. In the industry, it is used to lyse cells. A number of means are known for this purpose, including enzymatic means (e.g. lysozyme). ), chemical means (e.g. alkali or 5DS) and mechanical means (e.g. French presses and hydrodynamic shears). Furthermore, neurotrophins Mechanical means are preferred because they are less likely to cause damage. In the present invention French press or stance dead the cells under 10,000 psi pressure. STANSTED ■) It is preferable to lyse the cells by passing them through a cell disruption device. Delicious. Harvesting the lysate pellet is done by centrifugation at approximately 15,000×g. be exposed.

The invention also provides for recycling recombinant neurotrophins produced in cell culture. It also provides a method for collecting information. Such methods are applied to neurotrophins. This method solves the problems associated with conventional recombinant protein recovery methods. .

Natural neurotrophins are soluble in neutral buffers. However, E. coli? The recombinant neurotrophin obtained from this method behaves as an insoluble protein.

Recombinant 22-mouth trophin was detected in the side sill fraction and in the cell periphery. Osmotic shock, sphnyloplastization, or freeze-thaw if discharged into the marginal space It has been recovered using standard techniques such as the method [Bochner et al., U.S. Pat. No. 4,680,262]. However, recombinant neurotrophins can only be isolated as a small fraction of the total neurotrophins in the cell.

Many mammalian proteins expressed in bacteria exhibit insolubility. alien io When produced in a green environment, they do not fold properly and The normally hidden hydrophobic regions are thus exposed. As a result, such proteins forms aggregates and precipitates as inclusion bodies. In the literature, these proteins This suggests that the cysteine residues of the protein are at least partially reduced or mispaired. It has been suggested that (TsuJi et al., 1987. Biochea+1str y 26:3129-3134).

Standard methods for purifying recombinant proteins from inclusion bodies are described in the literature. [For example, U.S. Pat. No. 4,620 to Bui 1der et al. No. 948, US Pat. No. 49619 to Hershendon et al. No. 69, and Hung et al., U.S. Pat. No. 4,734,362. Please refer] These methods involve It involves an operation to dissolve the protein. After replacing strong denaturant with weak denaturant , the correct dilution is achieved by renaturing the protein in a neutral solution and then oxidizing it. A sulfide bond is generated.

Most of the recombinant neurotrophins obtained by these methods are biologically Inert. This is partly due to the complexity of the tertiary structure of the 22-bit trophin. It is thought that After denaturing the protein and reducing the sulfhydryl groups, , such polypeptides refold into the proper structure and form the correct disulfide bonds. It does not take the proper steric configuration. Furthermore, according to the results of the present inventors, Suggests that aggregates formed by recombinant neurotrophins are not typical inclusion bodies be done.

The present inventors have demonstrated that biologically active recombinant 22-bit trophs are derived from the insoluble fraction of host cells. I discovered a way to recover the coins. Essentially this method avoids the use of reducing agents By maintaining the correct oxidation state of the protein, It is solubilized in strong denaturing agents. Disulfide bonds are removed during purification. Reducing destroys the activity of purified neurotrophin polypeptides . Literature (e.g. Tsuji et al., 1987. Biochemistry 26:3129-3134), the insoluble recombinant proteins in the inclusion bodies are It is stated that it should have a disulfide bond generated or reduced in This is an unexpected result given that

To recover neurotrophins, strong denaturants are also replaced with weaker denaturants. available. However, the present inventors have discovered that basic amino acids such as histidine or The addition of these equivalents will enhance the neurotrophic effect during the renaturation process associated with the removal of weak denaturants. We found that lophines can be easily maintained in a soluble state (Prior et al., 1999). 90. WO90106764).

These findings suggest a new model for the behavior of recombinant 22-bit trophin are doing. Recombinant sneurotrophin polypeptide produced by host cells most of which fold properly and produce the correct disulfide bonds it is conceivable that.

Furthermore, neurotrophins have a high pH and are positively charged at neutral pH. Because they are electrically charged, they contain negatively charged molecules (e.g. DNA) in cells. , RNA, and other proteins). The result As a result, they become insoluble at neutral pH. Basic amino acids dissociate them It helps.

More specifically, recombinant neurotrophs produced by host cell cultures The method of the present invention for recovering fins contains strong denaturing agents but essentially eliminates reducing agents. Denature it by dissolving the neurotrophin in a solution that does not contain It includes the process of As used herein, the term "modifier" means Tertiary structure and soluble in aqueous solution by at least partial loss of secondary structure. It refers to a compound that reversibly develops a protein. As a strong denaturing solution, 4~ Guanidine salts (e.g. guanidine hydrochloride) and alkali metal titanium at a concentration of 9M ocyanate (e.g., sodium thiocyanate), as well as concentrations of 7 to 9M. Examples include urea. A preferred denaturing solution is 7-9M guanidine hydrochloride. preferred article The conditions are pH 7.0 to 9.0 and room temperature. In addition, the most preferable ones are p) Is, 8M guanidine hydrochloride.

In addition, during this process, the correct oxidation state of the sulfur atom in the cysteine residue is maintained. It is also necessary to hold Otherwise, any biologically active neurotroph There is a risk that the ability to recover the fins will be lost. Therefore, the solution essentially contains reducing agent. It must not contain any A solution essentially free of reducing agent is a This is a solution that maintains the disulfide bonds in proteins. When implementing the present invention, Sulfide reducing agents (e.g. β-mercaptoethanol or dithiothreates) Addition of even a small amount of silica (silver) can destroy the activity of purified neurotrophins. That will happen.

The susceptibility of recombinant neurotrophins to degradation by metalloproteinases it is obvious. Therefore, in solutions containing metalloproteinase inhibitors, Preferably, the lotrophin is recovered. For this purpose, heavy metals such as EDTA are required. chelating agents are preferred. The concentration of EDTA is from a minimum of at least 1mM. It may be within the range up to a maximum value of about 200mM. A concentration of 5-80mM is preferred. Yes, and 50mM is most preferred. Is the chelated heavy metal ion in solution? After dialysis, EDTA can be reduced or eliminated. In addition, Langre (Langley et al.

1990, EP 0398753) is useful as a metalloproteinase inhibitor. The following peptides are listed.

The recovery method of the present invention may also include one or more of the following steps. D After solubilizing the neurotrophin, the strong denaturant in the solution is replaced with a weak denaturant. Ru. A weak denaturing solution is 4 to 4 at pH 7.0 to 9.0 and room temperature. 9M, preferably 6-8M urea. The most preferred weak denaturing solution is 7M urea, p containing 50mM Tris-HCl, lomM-NaCl and 5mM-EDTA It is a solution of H8.0. The preferred exchange method is dialysis. Strong denaturant is 730-9 ．． If 0M urea, the weaker denaturant for this step may be lower concentrations of urea. is preferable.

To restore the neurotrophins, weak denaturants are removed from the solution. good Suitable removal methods are dialysis or diafiltration against non-denaturing solutions. In addition, θ~IM - It is preferred to use NaCl as a non-denaturing solution.

Another step in the recovery method is to remove recombinants from other contaminants present in the solution. The goal is to purify neurotrophins.

To this end, some of the typical protein isolation techniques known in the art include Any of the above can be used. In addition, S-Sepharose (S-3EPH Cation exchange on DEAE-Sepharose (Dt!AE) Anion exchange on -3EPI (ARO8E■) and reversed phase high pressure liquid chromatography A three-step isolation method involving chromatography (HPLC) is preferred. Such isolation process The treatment can begin at any stage after solubilization of the neurotrophin, It can also be continued through the denaturant exchange and removal steps. In addition, the urea phase It is preferable to start the purification at . Because this drug is an ion-exchange chlorine It does not interfere with the chromatography and partial purification can be done by diluting into a non-denaturing solution. This is because it facilitates the dissolution of eurotrophin.

The inventors have shown that unless trophins are purified to a significant degree, they are weak. We found that neurotrophins precipitate from solution when the denaturing agent is removed. . Solubility is a function of neurotrophin purity and other contaminants in solution. It depends on the nature. For example, negatively charged molecules like DNA are very It also prevents the dissolution of pure neurotrophins. Therefore, basic amino acids or (e.g., indoleacetic acid). maintains the solubility of neurotrophins in non-denaturing solutions. this The concentration of the substance must be effective to maintain the neurotrophin in solution. Must be. As a solution of basic amino acids, His concentration higher than 10mM Mention may be made of tidine, lysine and arginine and their salts. Among them, 20~ Histidine at a concentration of 500mM is useful, and also histidine at a concentration of 20-100mM. Stidine is most preferred. Note that the neurotrophin in the denaturing solution is absorbed by the denaturing agent. This step can be omitted if it is sufficiently purified before removal.

The results show that the method of the invention consistently produces homogeneous nuclei with the same N-terminal amino acid. - indicating that the lotrophin molecule is produced. In contrast, mammalian C When neurotrophins are expressed in HO cell lines, various N-terminal access points are expressed. It was found that a mixture of neurotrophin molecules with amino acids was produced.

Therefore, the recombinant neurotrophins produced by the method of the present invention are unique. This seems to be the case.

The method of the invention also uses neurotrophins that have biochemical properties similar to neurotrophins. It is also useful for recovering other proteins.

In other words, the method of the present invention allows bacteria to be incubated for one month to properly fold and maintain proper disulfurization. produces fido bonds, but is not suitable for denaturation and reduction in conventional recovery techniques. Proteins that are very difficult to restore to have positive disulfide bonds Useful for recovering quality. Some such proteins have higher than about 9.0 includes a protein that has a pH and contains at least two disulfide bonds. It will be done. [Candidate molecules include secreted leukocyte protease inhibitor (Mill er et al., 1989. J, BaCteriOlOg>’ 171:2166 -2172) O, J: and full-length recombinant CD 4 (Fisher et al., 1 989. WO89101940). ] As described in the Examples below, according to the present invention, biologically active human nuclei - Expression of lotrophin-4 is disclosed. The human NT-4 DNA sequence was converted into a DNA plaza. When subcloned into the Sumid vector pRG91, pRG173 was obtained. It was. E. coli strain RFJ 26 was transformed using this hNT-4-containing plasmid. and then remove the biologically active material from the culture system using the methods described herein. NT-4 was recovered. However, the invention is limited only to these specific embodiments. should not be done. For example, the region of HG7-2 encoding human NT-4 and Throughout this specification, any substantially homologous nucleic acid sequence may be used. Any number of DNA plasmid expression methods described in or known to those skilled in the art. Vectors can be created. In addition, these expression vectors can be used to express any a number of E. coli strains, thereby transforming a number of E. coli strains into a useful amount of biologically active NT-4. can be produced.

In order that the invention may be more clearly understood, examples are presented below. this These examples are for illustrative purposes only and do not in any way fall within the scope of the invention. should not be construed as limiting the

6. Example 2 Production and recovery of recombinant hNGF by E. coli Known human NGF sequences Column [Nancy manufactured by Regeneron Pharmaceuticals] Carrying the mature human NGF (hNGF) gene while using Ip] The DNA sequence was amplified from human genomic DNA by PCR amplification. In that case , oligodeoxynucleotide ply? -PAN-20 (5°-AAGCGGT CGA CATCTOATCCAATCTTCCAT AGAGGTGAAT TCTCAGTA-3”) (sequence number near) and EVD-3 (5°-GGCA GGCGGCCGTCATCTCA CAGCCTTCCT GCTG-3°) (SEQ ID NO: 28) was used. These primers are 5′ of the human NGF gene. Incorporating a 5all restriction site at the end and an Eagle restriction site at the 3' end. designed to. To incorporate the 5alI site at the 5° end using PAN-20, , serine the second amino acid from the amino-terminal sequence of the mature hNGF protein. (a structurally similar amino acid) to threonine (a structurally similar amino acid). there were. For the 8 N-terminal amino acids, the sequences of 3 known neurotrophins are used. Given the limited identity, such conservative amino acid changes may not affect activity. It should have no effect.

Furthermore, the identity of the amino acid residue at this position is similar to that obtained from other species. not conserved in GF (Ibanez et al., 1990, EMBOJ, 9:1477-1483). In addition, PAN-20 is a protein obtained The G+C content of this region of the NGF gene should be avoided without affecting the quality sequence. Designed to reduce the amount.

Digest the amplified DNA fragment with restriction endonucleases 5ail and EagI. and then pRPN50 (manufactured by Regeneron Pharmaceuticals). Plasmid pRPN102 was obtained by ligation between the 5alI and Eag1 sites.

pRPN50 has the λPL promoter inserted into the promoter insertion site. Expression plasmid pNKS97 (Panayotatos, 1 987. Engineering an Efficient Express sion System, in: Plasmids-A Practica l Approach, ed, Herdy, K,, IRL Press , 0xford/Washington.

D, C,).

Facilitates inhibition at 30°C and releases inhibition by heat shock at 42°C cI857λPt, repressor gene in pRPNl02 to enable Incorporated a child. Primer EVD-2 generates NruI restriction sites at each end. 6 (5'-CCATTATCGCGACCAGAGGT-3°) (SEQ ID NO: 9) and EVD-27 (5°-TCTTGCTCGCGAGTTATCAG of this DNA sequence using CTATGCG-3°) (SEQ ID NO: lO). PCR amplification was performed. This 821 bp fragment contains the PRM constitutive promoter. It extends 70 bp upstream from the c1857 coding sequence within the region that It extends asbp beyond the cI857 stop codon. PCR candidate By screening genes that have the same transcriptional direction as the hNGF gene, When searching for a plasmid containing the c1857 insert, plasmid pRPN133 was found. (Figure 1) was obtained.

Plasmid pRPN133 has been deposited with the ATCC and has accession number 75029. Granted.

Using RPN133 to transform the htpR1'/Rtslon-mutant strain of E. coli. By transformation, pRPN133/LC137 was obtained.

In a 2 liter flask, L with 100 mg/L ampicillin added. Inoculate 400 mL of B medium pRPN 133/LCI 37 and do not shake. The cells were incubated at 28°C.

In the late logarithmic phase (under these conditions OD 6*. = 1. 0 to 2.0), ■ 40 LB medium prewarmed to 55°C for cultures incubated overnight. Add 0 mL and then incubate for 5 hours at 42°C with shaking. hNGF synthesis was induced by continued treatment.

Cells were harvested by centrifugation and cell pellets were stored at -20°C. cell Thaw the pellet (1.0-2.5 g), add 2°, OmL Buffer A (100mM Resuspend in Tris-HCl, 50mM-EDTA, pH 8,0) and Passed through a Stansted cell disruptor at 0 psi and incubated at 4°C for 15 min. Centrifugation was performed at 23,000×g. The pellet thus obtained was treated with 2M guanidine hydrochloride. and a solution containing 5 m M-E D T A (pH 8,0) 10, OmL Washed twice.

To solubilize the recombinant hNGF, using a Potter homogenizer, 40 mL of a solution containing 8M guanidine hydrochloride and 5mM-EDTA (pH 8, O) The pellet was resuspended in the medium. The suspension was then heated at 23,000×g for 15 minutes. It was then centrifuged.

2 liters of buffer B (7M urea, 100mM histidine, 0.1mM-ED The supernatant was dialyzed against TA, pHe, O) for 20 hours at 25°C. So During this period, the buffer was changed twice. The obtained dialysate was heated at 23,000×g at 4°C. Centrifuged for 15 minutes.

Equilibrated with Buffer B and using the same buffer to reach the reference absorbance at 280 nm. Sepharose column (diameter 2.5 cm x height 6.0 cm) m) The supernatant liquid was placed on top. Then 0,0- in 300 mL of buffer B Proteins were eluted using a 1.0 M NaCl gradient and Collected both amounts.

100-fold excess of buffer C (7M urea, 50mM Tris-HCL, 0.1mM - EDTA, pH 8,5) for 20 hours at 25°C. Dialyzed.

2 per minute. DEAE-Sepharose column equilibrated with Buffer C at a flow rate of OmL (2.5 cm in diameter x 7.5 cm in height) A dialysate was placed on top. Contains hNGF Effluent fractions were collected and added containing 100mM histidine and 0.1mM-EDTA. Dialysis was performed twice against 40 volumes of a solution (pH 6,0) at 4° C. over one night. profit The dialysate was centrifuged at 23,000×g for 15 minutes at 4°C. next A solution containing 100mM histidine and 0.1mM EDTA (pH 6,0 ) equilibrated with TOYOPEARL [F] CM6505 (weak Place the dialysate on the cation exchange resin column (1°0 cm x 5.0 cm) , and eluted using a 0.0-1.0M NaCl gradient. Contains hNGF Collect both fractions and filter with a Millipore GV filter. and then stored at 4°C. Amino-terminal sequences of purified proteins were determined by direct sequence analysis. I confirmed.

Explanted and dissociated E8 was tested for neurite outgrowth in medullary ganglia. The biological activity of the protein was tested (Figures 2A and 2B) (Lindsay et al., 1985. Dev, Biol, 112:319-328). This way Based on the above criteria, recombinant hNG purified from E. coli by this method F was found to have an activity equivalent to NGF purified from mouse salivary glands. .

To facilitate both the synthesis and secretion of 22-mouth trophin in E. coli, A signal sequence based on LamB was created. LamB signal sequence (Fy!J3 A1 SEQ ID NO: 1) was developed by Regeneron Pharmaceuticals, Inc. To create a series of secretion vectors based on the developed pRPN series expression vectors. This is the natural E. coli signal sequence selected for this purpose. The secretion vector is a LamB system. Synthetic RNA fragment encoding the signal sequence was inserted into pRPNO9 or pRPNI6. was created by cloning into.

These plasmids were stored in the expression vector pNKS97 (Panayot atos, 1987. Engineering an Efficient Expression System, in: PIasmids-A Pr actual Approach, ed, Herdy, K,, IRL 1ac in Press, 0xford/Washington, D, C,) It was induced by inserting the UV5 promoter. LamB is , LamB expression is IacUV5 or T7φ1. Under the control of l promoter It was inserted into the insertion site of the structural gene so that the

This synthetic NA fragment LamB1 (FIG. 3B, SEQ ID NO: 2) is derived from wild-type La It encodes the same 25 amino acids as the mB signal sequence. However, the inventor et al. for the purpose of creating unique restriction enzyme sites and maximizing translation efficiency. Therefore, the nucleotide sequence was changed relative to the wild-type nucleotide sequence. Lam B1 also contains seven degenerate nucleotide positions in which C or G are replaced by A or T. has an exchange. This reduces the stability of possible secondary structures of the mRNA. L amB1 also has several new restriction sites.

The present inventors also investigated LamB2 (Figure 3C1 SEQ ID NO: = 3) was produced. To create LamB2, LamB1 was amplified by PCR amplification. Nucleotide changes were made at the 3' end. These changes result in blunt-ended DNA An EagI restriction site was introduced to facilitate cloning of the fragment.

If mature hBDNF is fused to LamB1, the fusion protein can be produced in E. coli. High quality and efficient synthesis is achieved. The authenticity of the synthetic product is determined by DNA-dependent cell-free Selective synthesis in transcription-translation protein synthesis system, T7 RNA port in E. coli Selective synthesis of products using a limerase expression system and monoclonal myc specificity Synthesis of products with a C-terminal myc tag that allows identification of chimeras by null antibodies confirmed by. This is evidenced by the mobility on 5DS-PAGE. Notably, both of these fusion proteins synthesized in E. coli Processed to DNF. The expression obtained using LamB1 or LamB2 Current levels cause hBDNF to accumulate in amounts of approximately 1-10% of total cellular protein.

We modified LamB1 to increase processing efficiency. Sunawa Finally, four amino acids (Leu-Ala-VaI- LamB5 (Figure 3D, SEQ ID NO. No.: 4) was obtained. Also, by applying the same insertion to LamB2, LamB4 (Fig. 3E, SEQ ID NO: 5) was obtained. Such insertions include LamB1 and LamB1 The hydrophobic core region of B2 is extended from 10 to 14 amino acids. So As a result, the processing in converting the hBDNF fusion protein to mature hBDNF This results in a five-fold increase in programming efficiency (Figure 4).

LamB5- and LamB4-hBDNF molecules excreted into the periplasmic space are Processed into mature hBDNF more rapidly than the native LamB fusion. deer However, since fewer molecules are excreted, the net mature hB in this case The amount of DNF does not increase. In any case, the changes brought about by LamB5 and LamB4 Increased efficiency should improve yields of other proteins.

Two complementary synthetic oligonucleotides [LamB] designed to 80. 5'-ATGATCACACTGC to TAAGCTTCCGCCTAG CT GTAGCAGTAG CAGCAGGTGTAATGTCTGCA C AGGCCATGG CCCGGGATCC-: l' (SEQ ID NO: ll) and Lam888.5'-CTAGGGATCCCGGGCCCATGGCCTGTG CAGA (SEQ ID NO: 12)]. This DNA fragment is called pRPN1 6 (manufactured by Regeneron Pharmaceuticals) Spe I and 5ph I restriction site to obtain plasmid pRPN52.

This plasmid was subsequently modified to create additional restriction sites. sand That is, complementary oligonucleotide L amB 2A [5°-CATGGCCA GT CGGCCGAG-3' (SEQ ID NO: 13)] and Lam82B [ 5'-GATCCTCGGCCGACTGG(,-3'(SEQ ID NO: 1.4 )], a synthetic NA fragment consisting of the product obtained by annealing of pRPN Unique NcoI and BamHI in the LamB1 signal sequence in 52 inserted between restriction sites. The thus obtained plasmid pRPN88 is a signal LamB2 signal with a unique restriction site at the location of the lupeptidase recognition sequence the signal sequence and any other DNA sequence. facilitate integration with The LamB2 signal sequence in pRPN88 is I It is transcriptionally regulated by the acUV5 or T7φ1.1 promoter, and T Translationally regulated by the 7φ1.1 liposome binding site.

7.2 Production and recovery of recombinant hBDNFmyc Human BDNFmyc has an N-terminal A tag consisting of mature hBDNF fused to an antigenic “tag” from the end to the C-terminus. It is protein.

Such a tag has the amino acid sequence EQKLI 5EEDL (SEQ ID NO: 15). It is a peptide that These 10 amino acid residues represent the human c-myc protoma It is derived from the tumor gene. The antibody that recognizes this tag is [Evan et al., Mo1. Ce11. Bio L, 5:3610-3616, and also [Test system for detecting neurotrophic activity j US Patent Application Box 071532128 entitled Squinto et al. 3 (incorporated herein by reference).

A fusion protein consisting of LamB2 signal sequence and hBDNFmyc protein. To prepare the oligonucleotide primer N8-hBDNF [5° -CCCACTCTGA CCCTGCCCGCCGAGGG-3' (sequence No.: 16)] and C2-hBDNFmyc [5°-GCTATGCGGC Using CGCTACAGAT CCTCCTC-3' (SEQ ID NO: 17)] pCDM8-hBDNFmyc (Regeneron Pharmaceuticals, Inc.) The DNA sequence of hBDNFmyc was extracted from the DNA sequence encoding hBDNFmyc from The sequence also consists of a DNA sequence encoding a myc tag consisting of 10 amino acids. It can also be obtained by fusing DNA sequences encoding mature hBDNF. Wear. The DNA sequence encoding mature hBDNF was obtained by PCR amplification as described. It was isolated 6 (Hyman et al., 1991. WO091101568).

The double-stranded DNA sequence encoding the myc peptide tag consisting of 10 amino acids is Can be chemically synthesized. For the purposes of this example, such hBDNF Ba1l and Ba1l at the 5° and 3′ ends of the myc hybrid DNA sequence, respectively. An EagI restriction site is provided.

It should be noted that low (New England Biolabs) o In pRPN98, the LamB2 signal sequence is the hBDNFmyc protein. Since it is fused to the mature part of the protein, the position of the signal peptidase recognition sequence cleavage at the His position +1 relative to the proprotein processing site. It should generate hBDNFmyc protein starting from a tidine residue (Le iblock, 1989. Nature 341:149-152).

Unique Nru in pRPN98! and the DNA sequence between the Pvu11 sites. sequence that controls the copy number of the plasmid by deleting the sequence (Twigg and 5herratt, 1980. Nature 283:216-2 18) was removed. The thus obtained plasmid pRPN121 (Fig. 5) is the parent It has a copy number approximately three times greater than that of the plasmid.

Plasmid pRPN121 has been deposited with the ATCC and has accession number 75028. Granted.

Transforming E. coli (7) W3110IqF- strain using pRPNl 21. W3110I'F-/pRPNI21 was obtained.

In a 2 liter flask, add W31101QF- to 500 mL of LB medium. /pRPN121 and grown to late logarithmic phase (OD6) at 37°C with shaking. 9° = approx. 1. It was grown until it reached 0). Then the final concentration was 1% (W/V ) and incubate the culture overnight at 37°C. I felt it. Cells were harvested by centrifugation and treated with 0.2M) Lis-HCl (pH 8°0 ) and then the cell pellet was frozen at -70°C.

Thaw the cell pellet (approximately 5 g) and add 0.2 M Tris-HCl (pH 8, 0), 1 mM Ca C12 and 25 units of Micrococcus nuclease [ Manufactured by Boehringer Mannheim resuspended in 20 mL of a solution containing This cell suspension was heated under a pressure of 8000 psi. Pass through a French press and then at 4°C using an SA600 rotor. Centrifugation was performed at 15,000 rpm for 15 minutes. 0.2M) Lis-HCI (pH 8,0), 10mM-EDTA and 2% Triton X-io Resuspend in 30 mL of solution containing o and rock gently for 1 hour at room temperature. Then, at 15,000 rpm at 4°C using a 5A600 rotor. Centrifuged for 5 minutes.

The pellet was washed twice with 20 mL of 2M guanidine hydrochloride. potter homogeny 8M guanidine hydrochloride, 10mM) Lys-○CI (pH 8 ,5) in 10 mL of buffer containing 10 mM-NaCl and ImM-EDTA. The pellet was resuspended. This extract was adjusted to 20 mL using the same buffer.

7M urea, 50mM) Lis-HCl (pH 8,5), 10mM-NaCl and The extract was perfused overnight at room temperature against 100 volumes of a solution containing 1 mM and 1 mM EDTA. analyzed.

Contains 7M urea, 50mM) Lis-HC1 (pH 8,5) and 1mM-EDTA. Equilibrated by passing the same buffer at a flow rate of 3 mL/min and using the same buffer. DEAE zeta blebs (DEARZ) washed to basal levels. IETA-PRBP) disk [Cuno, Inc., Meride dialysate was placed on top of the tube.

The effluent fraction was adjusted to 50mM histidine (pH 5,0), then 7M urea, 50mM For 100 volumes of a solution containing M histidine (pH 5,0) and 1mM-EDTA and dialyzed overnight at 4°C.

A mild solution containing 7M urea, 50mM histidine (pH 5.0) and 1mM-EDTA. Equilibrated by passing the buffer solution at a flow rate of 1 mL/min and using the same buffer. 1.6 cm x 6.5 cm of S-Sepharose washed to basal level The dialysate was placed on the column. 0, 0-1. in 200 ml of solution. OM's N The protein was eluted using an aC1 gradient, and both proteins containing hBDNFmyc were (Figure 6).

50mM histidine (pH 5,0), 50mM-NaCl and 1mM-ED Both volumes containing hBDNFmyC were dialyzed against 200 volumes of the solution containing TA.

50mM histidine (pH 5,0), 50mM-NaCl and lmM - Equilibrate by passing a buffer containing E D T A at a flow rate of 1 mL/min. 1.6 cm The dialysate was placed on a 5 cm S-Sepharose column. In 200 ml of solution 0, 0-1. Elute the protein using an OM NaCl gradient and Fractions containing hBDNFmyc were collected.

The hBDNFmyc (purity 85%) in this sample was collected in a 0.45cm x 5. Ocm directly onto a C4 reverse phase column [VYDAC■], and in 80 mL of 0.1% trifluoroacetic acid delivered at a flow rate of 0.75 mL. Elution was performed using a 0-67% acetonitrile gradient. Peak fraction is over 95% (Figure 7).

Analysis of the N-terminal amino acid sequence revealed that the purified protein has a uniform N-terminus. It was confirmed that it was a genuine hBDNFrrryc (Leibrock, 1 989. Nature 341:149-152).

Purified hBDNFmyc was isolated from the medullary ganglion (DRG) derived from E8 chicken embryos. ) promotes neurite outgrowth from explants and inferior ganglion explants, as well as promoting E8 chicken in promoting survival and dendritic outgrowth of dissociated neurons from avian DRG. It had biological activity (Figure 8). This activity was tested using DRG explants. is equivalent to recombinant human BDNF purified from mammalian cell extracts. Ta. The tested substance corresponds to lane 4 in FIG.

7.3 Production and recovery of recombinant hBDNF Production and recovery of hBDNFmVc Production of full-length mature recombinant hBDNF using methods described in connection with Upon production and recovery, a protein with similar biological activity was obtained ( Figure 9). Plasmid expressing LamB1-hBDNF fusion protein I) RP The preparation of N149 (FIG. 10) is similar to that of pRPN121. The above The synthetic NA fragment of LamB1 was then transformed into pRPNO9 (Regeneron Pharmash). The protein was cloned into the sph I and Spe I restriction sites of Lasmid pRPN31 was obtained. In addition, oligonucleotide primer N1-hB DNF　[5°　-ACTCTGACCCTGCCCGCCGA　GGGGAG CTG-3' (SEQ ID NO: l8)] and CI-hBDNF [5' -GCGCGGATCCCTATCTTCCCCTTTTAATGG TCAA pCDM8-hBDNF ( DPN6 of mature hBDNF from Regeneron Pharmaceuticals, Inc. I got 6. of the pBR322 sequence resulting in a higher copy number (as described above). Plasmid pRPNI49 was created by performing NruI-Pvull deletion in pRPN66. I got it. In this plasmid, LamB1-hBDNF expression is 1 acU Regulated by the V5 and T7φ1.1 promoters and liposome binding sites. Let's go. Plasmid pRPNl 49 has been deposited with the ATCC and has accession number 75. 027 was given.

pRPN149 was used to transform E. coli strain 1qF-W3110 (7). , recombinant hBDNF was produced and recovered according to the procedure described in Example 7. .

8. Example: Production and recovery of recombinant hNT-3 Regarding hNGF and hBDNF Human NT-3 was produced and recovered by a method similar to that described above. It will be done.

The DNA sequence encoding full-length mature hNT-3 is derived from human brain cells. cDNA library [Hohn et al., 1991. WO91103569 ( (incorporated herein by reference)]. In that case, Oligonucleotide primer EVD-45 [5°-CCTATGCAGA G CATAAGAGTCACCGAGGA-3° (SEQ ID NO: 20) and E VD-7[5’-GTAAGGGCGG CCGAAGTTTA ATAAA TAAAG GTC-3° (SEQ ID NO: 21)] is used. these The primers were derived from the DNA sequence for rat NT-3 (Maisonpierre et al., 5science 247:1446-1451). Ru. The sense primer has almost the same sequence as human NT-3. anti sense sp The primer is a hybridization site approximately 100 base pairs downstream from the stop codon of the human gene. Is it?

This DNA fragment was inserted into the Ba1l and Eag1 sites of pRPN88 to generate hB It has a C-terminal EagI restriction site suitable for replacing the DNA sequence of DNF. ing. E. coli strain IQF-W3110 was constructed using the plasmid thus obtained. be transformed.

Recombinant hNT-3 was then produced and recovered in the same manner as in Example 7. Ru. Recombinant hNT-3 purified from Escherichia coli was purified by Hohn et al. 91. WO091103569) i: Similar news to the one described -Expected to exhibit lotrophin activity.

Oligonucleotide N1-NT4 [5'-CCGGGGTCTCTGAAA CTGCACCAGCGAGTCG-3' (SEQ ID NO: 23)] and C1- NT4 [5°-GGTGCAGTTTCAGAGACCCCATACGCC GGCTGCGGTTGGC-3° (SEQ ID NO: 24)] as a primer encodes the putative mature region of the human NT-4 (hNT-4) gene. The DNA sequence was amplified from NT-4 HO2-2 DNA by PCRi.

Oligonucleotide C1-NT4 has an Eagl restriction site on the 3' side of the NT-4 gene. This is to give a certain rank. The fragment generated by PCH was digested with Eagl, and cloned into pRG91 digested with Mscl-Eagl. instructor The resulting plasmid pRG173 contains 1acUV5 and T7φ1, transcriptional regulation by T7φ1. Translation regulation by liposome binding site The mature region of hNT-4 (amino acid residues in the sequence translated from HO2-2) Glycine at position 81) is fused to the LamB signal sequence. It was.

This plasmid also has a ropl deletion that increases plasmid copy number. there was. Such structures were determined by restriction enzyme analysis of purified plasmid DNA, purified DNA sequence analysis of plasmid, presumed to be encoded by pRG173 In vitro synthesis of proteins of approximate size, coded by pRG173. neurite outgrowth stimulation activity (as described below). a suitable N-terminus [GVSET APAE (SEQ ID NO: 25)] by in vivo synthesis of a protein having It was confirmed.

pRG17 using LB medium supplemented with 100 mg/L ampicillin. 5 ml of a culture of E. coli strain RFJ26 transformed with 3 was incubated at 37°C under aeration. The cells were grown overnight. Cultures incubated overnight were ml of LB medium and cultured at OD5. . Proliferated until it reached =5.3 .

At this point, add lactose to 1% (W/V). to induce hNT-4 expression, and cultures were grown overnight under aeration. Ta. Cells were then collected by centrifugation and frozen at -70°C. cell pellet Melt (7.2 grams) and add 200 mM Tris-HCl (pH 8.0) and Resuspend in 73ml of solution containing 50mM-EDTA and Lysed by three consecutive passes through a cell disruptor. 26000X lysate After centrifugation for 10 minutes at Obtained. This soluble fraction was added to 50mM) Lis-MCI (pH 8,5) at 4% Dialyzed at ℃ for 5 hours and diluted 10 times with 20mM-MES (pH 6,0). and high-speed S-Sepharose scaffold equilibrated with 20mM-MES (pH 6,0). 20mM-MES (pHe, O) containing LM-NaCI It was eluted by Human recombinant NT-4 protein stimulates the growth of H8 DRG. The material was recovered in a wash with LM-NaCl.

8M guanidine hydrochloride, 50mM) Lis-HCl (pH 8,5), 10mM-N The insoluble fraction was resuspended and homogenized in 83 ml of a solution containing aC1 and 1 mM EDTA. It has become a quality. A solution containing 7M urea and 50mM Tris-HCl (1)H8,5) This insoluble fraction was dialyzed against 7M urea and 50mM Tris-HC. Fast DEAE-Sepharose column equilibrated with a solution containing I (pH 8,5) I put it on top. Collect the effluent fractions, 7M urea, 100mM histidine (pH 5,5). and 7M urea, 100mM Equilibrated with a solution containing M histidine (pH 5,5) and 1mM-EDTA. After loading onto a fast S-Sepharose column, 0 to IM N in the same buffer hNT-4 was eluted using an aC1 gradient. Collect fractions containing hNT-4, and a solution containing 100mM histidine (pH 5,5) and 1mM-EDTA. Dialyzed against. Approximately 95% of hNT-4 protein was fractionated with insoluble matter. .

9.3. Preparation of motor neuron enrichment cultures For these experiments, use Spragg- Dolly rats [H2O or Zivic-Miller ) was used. Carbon dioxide asphyxiation in pregnant rats Therefore, after killing, the embryos are quickly removed and immersed in a water-cooled medium before further treatment. I did it. The pulp was aseptically removed from rat embryos on day 14 of pregnancy. The marrow of this ( The first is cut caudal to the medulla oblongata (at the location of the medullary ganglia), and the sensory ganglia and The attached meninges were removed.

The pith was then divided into a ventral part and an intermediate dorsal part and cultured separately. Ventral side Cut the pulp tissue into small pieces, and add 0.1% trypsin (GIBCO) to PBS. ) and deoxyribonuclease type l [manufactured by Sigma] Incubate in lysed solution for 20 minutes at 37°C. tripe Remove the syn solution, wash and add 45% Eagle's Minimal Essential Medium (MEM), 45% Ham Nutrient Mix F12 (F12), 5% heat-inactivated fetal bovine serum (manufactured by GIBCO), 5% heat-inactivated horse serum (manufactured by GIBCO), glutamine (2 ), penicillin G (0,5 U/ml) and streptomycin (0,5 μ g/ml). Then, gently pass through a Pasteur pipette. After mechanically dissociating the tissue by grinding, the supernatant was collected and Nylon fiber [Nftex manufactured by Tetko, 40 μm). Schnard and Schaff for filtered cell suspensions. Schnaar and 5chaffner, 1981, J, N eurosci, 1:204-217). provided. All steps were performed at 4°C. F12:Metho in MEM (1:l) medium Dissolve lizamide and 18% metrizamide cushion (0,5 ml), 3 m l of 17% metrizamide, 3 ml of 12% metrizamide, and 3 ml of 8% metrizamide. A discontinuous gradient consisting of trizamide was formed. Filtered obtained as above After overlaying the ventral side of the pulp cell suspension (2.5 ml) on the graded gradient, swing out. Using a trotor [Sorvall HB 4], separate the tubes into two Centrifuged at 500g for 15 minutes. As a result of centrifugation, three cell layers, namely ( Fraction I (at 0-8% interface), Fraction I■ (at 8-12% interface) and ( Fraction III (12-17% at the interface) was obtained. Remove a small amount of cells from each interface (approximately 1 m1) and consisted of 50% F12 and 50% MEM. Glutamine (2mM), insulin (5μg/ml), transferrin (10 0 μg/ml), progesterone (20 nM), putrescine (100 μM) and and serum-free defined medium (Botte) supplemented with sodium selenite (30 nM). nstein and 5ato, 1979. PNAS 76:514-5 17) twice. Live in the presence of trypan blue while using a hemocytometer. Cell numbers were calculated. Next, poly-■,-ornithine (manufactured by Sigma, 10 μg/ ml) and laminin (manufactured by GIBCO, 10 μg/ml) in advance. 100,000 fractionated ventral pith cells (rich in motor neurons) in a well /cm”. On the day of plating, the cells were treated with NT-4. Ta. Do not use a 95% air 15% CO2 atmosphere with approximately 100% relative humidity. However, the culture was continued at 37°C in serum-free defined medium. 2nd day (48 hours later) ) and measurement of choline acetyltransferase (CAT). (Fonnum, 1975. J, Neuroche+n, 24 :407-409).

9.4. result Cultures incubated overnight after induction of pRGl73/RFJ26 The soluble fraction obtained from the purification procedure was tested on E8 DRG explants. It was done. Figure 13 shows that neurite outgrowth from E8 DRG explants is dose-dependent. It shows that he was furious.

The insoluble fraction was also tested on E8 medullary ganglion (DRG) explants, but the No scientific activity was observed.

Recombinant human NT-4 protein recovered from washing of insoluble fraction with IM-NaC1 The activity of the protein was determined by the dissociation kinetic reaction prepared as described in Section 9.3 above. Tested in neuron cultures and other test systems. At a dilution of l:20 When recombinant human NT-4 was added, motor neuron enrichment medium 48 hours after treatment Choline acetyltransferase activity in the diet increased 3.6 times.

Such a 3.6-fold increase was observed in the untreated control (C-NT) and buffer control (C-Buffer). (Fig. 14).

Bacteriophage is located downstream of the liposome binding site as well as the LamB signal sequence. A plasmid was created by subcloning the NT-4 coding region of HG7-2. pRG173 was obtained. Transforming E. coli strain RFJ26 with pRG173 Then, when large-scale culture of pRGl73/RFJ26 was induced, biological A sexually active form of recombinant human NT-4 was expressed. in vitro prokaryotes The ability to express a biologically active form of human NT-4 in an expression system shows that recombinant Production of human NT-4, peptides or their derivatives for therapeutic and diagnostic purposes This makes it substantially easier to scale up for specific purposes.

In this way, in this example, human NT-4 was cocooned in such a way that human NT-4 was expressed. The present invention also shows that a DNA sequence encoding a DNA sequence can be transcribed and translated in a prokaryotic system. to purify it into its biologically active form in such a way that its activity remains after purification. This indicates that the operation can be performed.

10. Deposit of microorganisms Under the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purposes of Patent Procedure, Recombinant DNA molecules pRPN133 (hNGF), pRPNl 21 (hBD NFmyc) and pRPN149 (bBDNF) on June 7, 1991, p RGl 73 (NT-4) was developed on October 28, 1991 as a recombinant bacteriophore. The HG7-2 was released by American Type Culture on August 22, 1991. The Collection (12301 Burklawn Drive, Rockville, MD) 20852) and was assigned the following accession number:

Recombinant DNA molecule ATCC accession number pRPNl 33 (hNGF) 75 029pRPNl 21 (hBDNFmyc) 75028pRPN149 (hBDNF) 75027pRG173 (hNT-4) 75131HG7 -2 (human NT-4 genome clone) 75070 The scope of the present invention microorganisms or as limited by the specific embodiments described herein. It's not possible. Indeed, based on the above description and accompanying drawings, the Various modifications of the invention in addition to the embodiments described herein will be apparent to those skilled in the art. . Such modifications are intended to be included within the scope of the following claims.

Various publications are cited herein, all of which are incorporated by reference in their entirety. Incorporated herein.

Although some embodiments of the present invention have been described above, the basic embodiment Other embodiments of using the methods and compositions of the invention by modifying the It is clear that the benefits of Therefore, the scope of the present invention extends beyond the foregoing specification and beyond. The present invention encompasses all modifications encoded by the claims. Not to be limited by the specific embodiments illustrated herein Needless to say.

FIG.1 FIG, 2A ng/ml hNGF FIG, 2B eight eight FIG.5 J FIG.7 FIG.9 FIG. 10 <　ha−←　c　to υ← tpo Φ L) Ll Be darkness 0 ward < (!l % OLI L5 LJ FIG. 12 Capacity (uL) FfG, 13 FIG. 14 Continuation of front page (51) Int, C1,5 identification symbol Internal office serial number // (C12P 211 02 C12R1:19) (81) Designated countries EP (AT, BE, CH, DE.

DK, ES, FR, GB, GR, IT, LU, MC, NL, SE), AU, CA, C3, FI, HU, JP, KR, NO, RU FI (72) Inventor: Fandle, James Bee.

United States of America 12520 New York State Grangeville, Oheia - Drive (no address)

Claims (86)

[Claims] 1. Neurotrophins in a solution containing strong denaturing agents and essentially no reducing agents recombinantly produced in a non-animal host cell culture, including the step of solubilizing the How to recover neurotrophins. 2. Strong modifiers include 4M to 9M guanidine salts, 4M to 9M alkali metal thiosyl 2. The method of claim 1, wherein the method is anate or 7M to 9M urea. 3. 3. The method of claim 2, wherein the strong denaturant is 7M to 9M guanidine hydrochloride. 4. 4. The method of claim 3, wherein the strong denaturant is 8M guanidine hydrochloride. 5. The method of claim 1, further comprising replacing the strong denaturant with a weak denaturant. . 6. 6. The method of claim 5, wherein the weak denaturant is 4M to 9M urea. 7. The strong denaturant is 7M to 9M guanidine hydrochloride, and the weak denaturant is 6M to 8M urine. 7. The method according to claim 6, wherein 8. The strong denaturant is 8M guanidine hydrochloride and the weak denaturant is 7M urea. The method described in claim 7. 9. The solution is effective in preventing the degradation of neurotrophins by proteases. 2. The method of claim 1, further comprising an effective amount of a metalloproteinase inhibitor. 10. The solution prevents the degradation of neurotrophins by proteases. 3. The method of claim 2, further comprising an effective amount of a metalloproteinase inhibitor. 11. 11. The solution further comprises 1mM to 200mM EDTA. the method of. 12. 4. The method of claim 3, wherein the solution further comprises 5mM to 80mM EDTA. Law. 13. 5. The method of claim 4, wherein the solution further comprises 50 mM EDTA. 14. 8. The method of claim 7, wherein the solution further comprises 5mM to 80mM EDTA. Law. 15. 9. The method of claim 8, wherein the solution further comprises 50 mM EDTA. 16. (1) Effective in maintaining the solubility of neurotrophins in a non-denaturing environment. Adjust the solution to contain the basic amino acid or indole acetic acid at the desired concentration. and (2) removing the weak denaturant. 5. The method described in 5. 17. (1) Effective in maintaining the solubility of neurotrophins in a non-denaturing environment. Adjust the solution to contain the basic amino acid or indole acetic acid at the desired concentration. and (2) removing the weak denaturant. The method described in 6. 18. (1) Adjust the solution to contain 20mM to 500mM histidine. and (2) removing the weak denaturant. Method described. 19. (1) Adjust the solution to contain 20mM to 100mM histidine. and (2) removing the weak denaturant. Method described. 20. (1) Adjust the solution to contain 20mM to 100mM histidine. and (2) removing the weak denaturant. Method described. 21. (1) Adjust the solution to contain 20mM to 100mM histidine. and (2) removing the weak denaturant. The method described in 4. 22. (1) Adjust the solution to contain 20mM to 100mM histidine. and (2) removing the weak denaturant. 5. The method described in 5. 23. Claims for recovering recombinant hNGF, hBDNF or hNT-3 3, 4, 7, 8, 12, 13, 14, 15, 19, 20, 21 or 22 Method. 24. Recombinant with threonine instead of serine as the second amino acid residue Claim 23 for recovering hNGF, full length hBDNF and full length hNT-3. Method described. 25. operably linked to a DNA sequence encoding a neurotrophin Animals other than those transformed with recombinant DNA molecules containing regulatable expression control sequences culturing a host cell, the host cell being a heat shock regulated gene mutant; A method for producing recombinant neurotrophins. 26. The host cell is E. The claim is an E. coli HtpRtmlon-mutant strain. 25. The method described in 25. 27. The host cell is E. The claim is an E.coli HtpRts lon-mutant strain. 26. The method described in 26. 28. The host cell is E. 28. The method according to claim 27, wherein the strain is E. coli LC137 strain. 29. 27. The method of claim 26, wherein the expression control sequence consists of the λPL promoter. 30. 28. The method of claim 27, wherein the expression control sequence consists of the λPL promoter. 31. 31. The method of claim 30 for producing hNGF. 32. hNG with threonine instead of serine as the second amino acid residue 32. The method of claim 31 for producing F. 33. E. coli transformed with pRPN133(hNGF). coli LC137 26. The method of claim 25, comprising culturing the strain. 34. Claims 26-3 for producing hNGF, hBDNF or hNT-3. 30. The method according to any one of 30. 35. Shift the signal sequence from 5' to 3' in the same reading frame as neurotrophins. was once operably linked to the DNA sequence consisting of the encoding fusion gene. Cultivating non-animal host cells transformed with recombinant DNA molecules containing expression control sequences A method for producing recombinant neurotrophins, comprising the step of culturing. 36. The host cell is E. coli, and the signal sequence is LamB or modified La 36. The method of claim 35, wherein mB. 37. Expression control sequence binds to lacUV5 promoter and T7φ1.1 ribosome 37. The method of claim 36, comprising a site. 38. The signal sequences are LamB1 (SEQ ID NO: 2) and LamB2 (SEQ ID NO: 3). ), LamB3 (SEQ ID NO: 4) or LamB4 (SEQ ID NO: 5). The method according to claim 37. 39. 39. The method of claim 38 for producing full-length hBDNF. 40. The method according to claim 39, wherein the signal sequence is LamB1 (SEQ ID NO: 2). Law. 41. 39. The method of claim 38 for producing hBDNFmyc. 42. The signal sequence is LamB2 (SEQ ID NO: 3) or LamB4 (SEQ ID NO: 3). :5). The method according to claim 41. 43. E. coli transformed with pRPN149(hBDNF). coliIqF-W 36. The method according to claim 35, comprising the step of culturing the 1330 strain. 44. For producing hNGF, hBDNF or hNT-3, claims 35- 39. The method according to any one of 38. 45. Recombinant proteins in solutions containing strong denaturing agents and essentially no reducing agents the protein has a pH of at least 9.0 and a low Produced by a non-animal host cell culture containing at least two disulfide bonds Method for recovering the produced recombinant protein. 46. Plasmid pRPN133 (hNGF). 47. Plasmid pRPN121 (hBDNFmyc). 48. Plasmid pRPN149 (hBDNF). 49. Purified homogeneous neurotrophin recovered by the method of claim 1. . 50. Purified homogeneous neurotrophin recovered by the method of claim 5. . 51. Purified homogeneous neurotrophin recovered by the method of claim 9. . 52. Purified homogeneous neurotrophy recovered by the method of claim 10. hmm. 53. Purified homogeneous neurotrophy recovered by the method of claim 16. hmm. 54. Purified homogeneous neurotrophs recovered by the method of claim 17. hmm. 55. Any of claims 49 to 54, which is hNGF, hBDNF or hNT-3. A purified homogeneous neurotrophin according to any of the above. 56. hNG with threonine instead of serine as the second amino acid residue 56. The purified homogeneous neurotrophin of claim 55, which is F. 57. 56. The purified homogeneous protein of claim 55, which is a full-length neurotrophin. - Rotrophin. 58. has at least one degenerate substitution with A or T in place of G or C A modified LamB sequence consisting of a DNA sequence encoding a LamB signal sequence. Genes related to GNAR sequence genes. 59. The gene of claim 58, having 7 degenerate substitutions. 60. LamB1 (SEQ ID NO: 2) or LamB2 (SEQ ID NO: 3), The gene according to claim 59. 61. LamB signal sequence with hydrophobic core region of 11-20 amino acid residues The genetic code for the modified LamB signal sequence gene, which consists of a DNA sequence encoding Denji. 62. The gene of claim 61, comprising a core region of 14 amino acid residues. 63. 63. The core region of claim 62, wherein the core region comprises the peptide LAVL (SEQ ID NO: 6). gene. 64. LamB3 (SEQ ID NO: 4) or LamB4 (SEQ ID NO: 5), The gene according to claim 63. 65. 4. The method of claim 3 for recovering recombinant neurotrophin-4. 66. 66. The method of claim 65 for recovering human neurotrophin-4. 67. Bacteriophore deposited with ATCC and designated accession number 75070 The DNA sequence contained in diHG7-2 and shown in Figure 11 (SEQ ID NO: 22) 67. for recovering human neurotrophin-4 encoded by How to put it on. 68. DNA plasmid deposited with ATCC and designated accession number 75131 The human protein contained in pRG173 and encoded by the DNA sequence shown in FIG. 67. The method of claim 66 for recovering toneurotrophin-4. 69. 26. The method of claim 25 for producing neurotrophin-4. 70. 70. The method of claim 69 for producing human neurotrophin-4. 71. Human neurotrophin-4 has been deposited with the ATCC under accession number 7507. contained in bacteriophage HG7-2 designated 0 and shown in FIG. 11 (SEQ ID NO: 71. The method according to claim 70, wherein the method is encoded by the DNA sequence shown in :22). 72. Human neurotrophin-4 has been deposited with the ATCC under accession number 7513. E.1 was specified. contained in the E. coli plasmid vector pRG173 and 71. The method of claim 70, wherein the method is encoded by the DNA sequence shown in 12. 73. Deposited with the ATCC with accession number 7513 having the sequence shown in FIG. B.1 was specified. transformed with coli plasmid vector pRG173. E. 26. The method of claim 25, comprising culturing E. coli RFJ26 strain. 74. 36. The method of claim 35 for producing neurotrophin-4. 75. 75. The method of claim 74 for producing human neurotrophin-4. 76. Human neurotrophin-4 has been deposited with the ATCC under accession number 7507. Contained in bacteriophage HG7-2 designated 0 and shown in FIG. 76. The method of claim 75, wherein the method is encoded by a DNA sequence. 77. Human neurotrophin-4 has been deposited with the ATCC under accession number 7513. E.1 was specified. contained in the E. coli plasmid vector pRG173 and 76. The method of claim 75, wherein the method is encoded by the DNA sequence shown in 12. 78. Deposited with the ATCC with accession number 7513 having the sequence shown in FIG. E.1 was specified. transformed with coli plasmid vector pRG173. E. 36. The method of claim 35, comprising culturing E. coli RFJ26 strain. 79. DNA plasmid vector-pRPN91. 80. Deposited with the ATCC with accession number 7513 having the sequence shown in FIG. DNA plasmid pRG173 designated 1. 81. Purified homogeneous neurotrophy recovered by the method of claim 49. N-4. 82. 82. The purified homogeneous protein of claim 81, which is human neurotrophin-4. Neurotrophin-4. 83. Bacteriophore deposited with ATCC and designated accession number 75070 The DNA sequence contained in diHG7-2 and shown in Figure 11 (SEQ ID NO: 22) 83. The purified homogeneous neurotrophin-4 of claim 82, encoded by . 84. E.C., deposited with the ATCC and designated accession number 75131. coli plastic According to the DNA sequence contained in the Sumid vector pRG173 and shown in FIG. 83. The purified homogeneous neurotrophin-4 of claim 82, wherein the purified homogeneous neurotrophin-4 is encoded by: 85. Coding human NT-4 under conditions such that human NT-4 is expressed by bacteria. Grow bacteria containing the recombinant nucleic acid and collect the produced human NT-4. A method for producing human NT-4, comprising: 86. 86. The method of claim 85, wherein the human NT-4 is biologically active.