Method for obtaining recombinant human brain-derived neurotrophic factor

Abstract

The present invention relates to human proBDNF muteins and methods for recombinantly producing biologically active human brain-derived neurotrophic factor (BDNF).

Description

[Technical Field]

[0001] The present invention relates to human proBDNF muteins and methods for recombinantly producing human brain-derived neurotrophic factor. [Background technology]

[0002] Neurotrophins are a family of growth factors that are crucial for regulating the maintenance, differentiation, and survival of neurons. In particular, neurotrophin signaling regulates a diverse array of neural processes, including differentiation, neurite outgrowth, axonal pruning, apoptosis, and cell survival during development, adulthood, and after nervous system injury.

[0003] Neurotrophins mediate these actions by binding to various receptors: tropomyosin-related kinase (Trk) receptor tyrosine kinases, p75 neurotrophin receptors, and members of the sortilin family. The human neurotrophin family consists of nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3), and neurotrophin-4 (NT-4). Each of these structurally related neurotrophins is synthesized as a precursor and then proteolytically cleaved to yield the mature neurotrophin (Huang EJ et al., Annu. Rev. Biochem. 2003;72:609-642).

[0004] Mature human BDNF contains 119 amino acid residues (SEQ ID NO: 1; Gray K. et al., FEBS Letters. 2008; 582(6):907-10), which represent the functional domains of the protein and are separated by three disulfide bonds (Robinson RC et al., Protein Science. 1999; 8(12):2589-97). The BDNF protein gene is located on the short arm of human chromosome 11 (Tian F. et al., Amino acids. 2010; 38(4):1067-74).

[0005] BDNF binds to at least two cell surface receptors, TrkB and p75, that can respond to this growth factor. This signaling molecule has well-documented abilities to regulate neuronal plasticity, cell growth, proliferation, cell survival, and long-term memory. In particular, it has been shown to act on certain neurons in the central and peripheral nervous systems, helping to support the survival of existing neurons and promoting the growth and differentiation of new neurons and synapses. In the brain, it is active in the hippocampus, cortex, and basal forebrain—areas essential for learning, memory, and higher-order thinking. BDNF is also expressed in the retina, kidney, prostate, motor neurons, and skeletal muscle, and is found in saliva.

[0006] Currently, BDNF administered as a therapeutic molecule is generally recombinantly produced. To be clinically viable, recombinant human BDNF (rhBDNF) must meet current regulations for the creation of clinically acceptable therapeutic formulations. Specifically, it is necessary to produce recombinant human BDNF characterized by a sufficiently high degree of purity and a consistent composition and activity profile.

[0007] Therefore, an optimal method for the production of recombinant human BDNF for therapeutic use must ensure the production of a molecule that meets these requirements and is suitable for large-scale implementation.

[0008] Although several methods have been developed to obtain and / or produce rhBDNF, they are characterized by several drawbacks related to complex operational flows and do not yield proteins with characteristics suitable for therapeutic use. Summary of the Invention [Problem to be solved by the invention]

[0009] Therefore, there is a felt need to develop an improved and simple method for producing biologically active recombinant human BDNF that can ensure a high degree of purity of the final product and sufficient and reproducible biological activity. [Means for solving the problem]

[0010] The present invention relates to a human proBDNF mutein having the amino acid sequence of SEQ ID NO: 3, in which X1, X2 and X3 are selected from non-basic amino acids and histidine, and X4 is selected from arginine and lysine.

[0011] Another object of the invention is a nucleic acid encoding a human proBDNF mutein as defined above. A further object of the present invention is a method for producing human BDNF, comprising the steps of: (i) providing a solution of a human proBDNF mutein as defined above; (ii) purifying the human proBDNF mutein from the solution provided in step (i), thereby obtaining a solution of purified human proBDNF mutein; and (iii) cleaving the human proBDNF mutant protein in the solution to obtain human BDNF. The method includes:

[0012] A further object of the present invention is human BDNF obtained or obtainable by the method described above. [Brief explanation of the drawings]

[0013] [Figure 1] FIG. 1 is a graph showing the concentration-response (0.23-9 μg / mL) growth curve of C6 cells incubated with BDNF for 48 hours. [Figure 2A]Panel (A) shows the electrophoretic analysis results of SeeBlue Plus2 Prestained Standard (protein molecular weight marker, MK), rhBDNF (standard, RS), TO load (TA1), fraction 1 (TA2), fraction 2 (TA3), fraction 3 (TA4), fraction 4 (TA5), fraction 5 (TA6), flow-through (TA7), and SeeBlue Plus2 Prestained Standard (protein molecular weight marker, MK) according to Example 9. Panel (B) shows the electrophoretic analysis results of SeeBlue Plus2 Prestained Standard (protein molecular weight marker, MK), rhBDNF (standard, RS), fraction 6 (TA8), fraction 7 (TA9), fraction 8 (TA10), fraction 9 (TA11), fraction 10 diluted 1:50 (TA12), fraction 10 diluted 1:20 (TA13), empty lane (EL), SeeBlue Plus2 Prestained Standard FIG. 1 shows the results of electrophoretic analysis of standards (protein molecular weight markers, MK). [Figure 2B] Continued from Figure 2. DETAILED DESCRIPTION OF THE INVENTION

[0014] definition The term "human wild-type BDNF" or "human BDNF" refers to the human form of the protein BDNF having the amino acid sequence of SEQ ID NO: 1, and optionally, said human BDNF may be "recombinant human BDNF" (rhBDNF), i.e., said "human BDNF" is recombinantly produced.

[0015] The term "human wild-type proBDNF" or "human proBDNF" refers to the precursor of human BDNF having the amino acid sequence of SEQ ID NO: 2, and optionally, said human proBDNF may be "recombinant human proBDNF" (rh-proBDNF), i.e., said "human proBDNF" is recombinantly produced.

[0016] The term "human proBDNF mutein" or "human proBDNF variant" refers to a human proBDNF containing one or more amino acid substitutions, and optionally, said human proBDNF variant may be a "recombinant human proBDNF variant" (rh-proBDNF variant), i.e., said "human proBDNF variant" is recombinantly produced.

[0017] As used herein, the term "correctly folded" means - References to human wild-type BDNF / human BDNF refer to human wild-type BDNF having a three-dimensional structure that corresponds to the three-dimensional structure of biologically active native human BDNF. - when reference is made to a human proBDNF mutein, it means a human proBDNF mutein having a three-dimensional structure that results in the formation of correctly folded human BDNF upon cleavage of the prosequence by a serine protease. The term "non-basic amino acid" refers to any amino acid other than a basic amino acid.

[0018] Detailed Description of the Invention As explained in more detail in the experimental section, the present inventors have developed a method for recombinantly producing human brain-derived neurotrophic factor (rhBDNF) that is advantageous over methods disclosed in the prior art because it is simple and allows for obtaining biologically active rhBDNF in high purity and high yield.

[0019] This is accomplished by expressing the human proBDNF mutein described below as a precursor to BDNF and using specific process parameters and materials discussed below.

[0020] Therefore, a first object of the present invention is to provide a method for the production of a nucleic acid having the amino acid sequence of SEQ ID NO: 3: MAPMKEANIRGQGGLAYPGVRTHGTLESVNGPKAGSRGLTSLADTFEHVIEELLDEDQKVRPNEENNKDADLYTSRVMLSSQVPLEPPLLFLLEEYKNYLDAANMSMX1X2X3X4HSDPARRGELSVCDSISEWVTAADKKTAVDMSGGTVTVLEKVPVSKGQLKQYFYETKCNPMGYTKEGCRGIDKRHWNSQCRTTQSYVRALTMDSKKRIGWRFIRIDTSCVCTLTIKRGR (wherein X1 at position 108, X2 at position 109, and X3 at position 110 are selected from a non-basic amino acid and histidine, and X4 at position 111 is selected from arginine and lysine). It is a human proBDNF mutant protein having the following structure:

[0021] Preferably, X1, X2 and X3 are independently selected from the group consisting of alanine, asparagine, aspartic acid, cysteine, glutamine, glutamic acid, glycine, isoleucine, leucine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine, valine, and histidine (SEQ ID NO: 5).

[0022] More preferably, X1, X2 and X3 are independently selected from the group consisting of valine, alanine, glycine, serine, threonine, tyrosine, asparagine, aspartic acid, glutamine, glutamic acid, methionine, and histidine (SEQ ID NO: 6).

[0023] In one preferred embodiment, X1 and X3 are independently selected from alanine and valine (SEQ ID NO: 7), in another preferred embodiment, X2 is selected from valine and serine (SEQ ID NO: 8), and in a further preferred embodiment, X4 is arginine (SEQ ID NO: 9).

[0024] According to a preferred embodiment, X1 and X3 are independently selected from alanine and valine, X2 is selected from valine and serine, and X4 is arginine (SEQ ID NO: 10).

[0025] A particularly preferred proBDNF mutein according to the invention is the mutein of SEQ ID NO: 4, which corresponds to the mutein having the amino acid sequence of SEQ ID NO: 3, wherein X1 is valine, X2 is serine, X3 is alanine and X4 is arginine.

[0026] A further object of the present invention is a nucleic acid encoding the human proBDNF muteins described herein. A further object of the present invention is a method for recombinantly producing human BDNF, comprising: (i) providing a solution of a human proBDNF mutein according to the first object of the present invention; (ii) purifying the human proBDNF mutein from the solution provided in step (i), thereby obtaining a solution of purified human proBDNF mutein; and (iii) cleaving the human proBDNF mutant protein in the solution to obtain human BDNF, more particularly recombinant human BDNF. The method includes:

[0027] Preferably, said method for recombinantly producing human BDNF comprises: (i) providing a solution of a human proBDNF mutein according to the first object of the present invention, said human proBDNF mutein having been correctly folded by expressing the human proBDNF mutein according to the first object of the present invention in host cells cultured in a liquid culture medium and isolating the human proBDNF mutein from the host cells; (ii) purifying the human proBDNF mutein from the solution provided in step (i) via chromatographic purification, preferably using an elution solution comprising urea, more preferably an elution solution comprising citric acid and urea, thereby obtaining a solution of purified human proBDNF mutein; and (iii) cleaving the prosequence of the human proBDNF mutein in said solution to obtain biologically active human BDNF, more particularly recombinant human BDNF, preferably by adjusting the pH of the solution obtained in step (ii) to a pH value comprised between 5.8 and 9 before said cleavage. Includes:

[0028] Preferably, step (i) of said method for recombinantly producing human BDNF comprises the steps of: a) providing a nucleic acid encoding a human proBDNF mutein according to the first object of the present invention, b) introducing said nucleic acid into an expression vector; c) introducing the expression vector into a host cell; d) growing said host cells in a suitable liquid culture medium, wherein said host cells express the human proBDNF mutein according to the first object of the present invention in the form of inclusion bodies, e) lysing the host cells and isolating the human proBDNF mutein inclusion bodies produced by the host cells in step d); f) dissolving the inclusion bodies in a denaturing solution, thereby obtaining a solution of denatured human proBDNF mutein; g) diluting the solution of denatured human proBDNF mutein into a refolding solution, so that the denatured human proBDNF mutein assumes a correctly folded conformation, thereby obtaining a solution of correctly folded human proBDNF mutein. Includes:

[0029] Preferably, the nucleic acid provided in step a) is DNA. Preferably, the nucleic acid provided in step a) has a codon-optimized sequence, more preferably a codon-optimized DNA sequence, for efficient expression in a host cell, preferably E. coli.

[0030] Preferably, the expression vector in step b) is selected from pET28a, pET28b, pET28c, pBR322, pMAL, pUC19, and all derivatives, more preferably, the expression vector in step b) is pET28a or pET28b or pET28c, however, any suitable expression vector known in the art can be used for this purpose.

[0031] Preferably, said host cell in step c) is a prokaryotic host cell. Suitable prokaryotic host cells that can be used in accordance with the present invention are well known in the art and include, but are not limited to, prokaryotic cells such as bacteria, e.g., E. coli, Bacillus sp., and Salmonella, that can be transformed with plasmid DNA, recombinant bacteriophage DNA, or cosmid DNA expression vectors containing the nucleic acids of the invention; kits for such expression systems are commercially available.

[0032] According to a preferred embodiment, the host cell of step c) is preferably an E. coli selected from E. coli BL21(DE3), JM108 / 109(K12), JM106, JM83, and TBI. Any other E. coli strain suitable for the expression of recombinant proteins may also be used.

[0033] In step c), the step of introducing the expression vector into the host cell can be carried out by any method known in the art. Preferably, according to the present invention, the step of introducing the expression vector into the host cell comprises transforming the expression vector into the host cell, preferably E. coli, by temperature shock.

[0034] Preferably, in step d), said suitable liquid culture medium in which the host cells are grown is Terrific Broth (TB). Preferably, in step d), said liquid culture medium contains kanamycin, preferably at a concentration of 30 μg / ml.

[0035] Preferably, in step d), the liquid culture medium contains a compound that induces expression of the recombinant protein, preferably isopropyl β-d-1-thiogalactopyranoside (IPTG).

[0036] Preferably, in step d), the compound inducing the expression of the recombinant protein, preferably isopropyl β-d-1-thiogalactopyranoside (IPTG), is used to induce an optical density (OD) of 1 measured at 600 nm. 600 =1).

[0037] Preferably, the inclusion bodies obtained in step d) may also contain proteins other than the human proBDNF mutein of the invention. Preferably, the inclusion bodies contain at least 60 wt.%, at least 70 wt.%, at least 80 wt.%, or at least 90 wt.% of the human proBDNF mutein according to the invention, based on the total amount of protein present in the inclusion bodies.

[0038] In step e), lysis of the host cells can be carried out by conventional methods, such as high-pressure homogenization, sonication, or lysozyme (Rudolph, R. et al. (1997); Folding proteins. in Creighton, TE (ed.): Protein Function: A Practical Approach. Oxford University Press, pp. 57-99).

[0039] Preferably, in step e), lysing the host cells is carried out by high pressure homogenization in a lysis buffer, preferably comprising, preferably consisting of, 0.1 M Tris buffer, pH 7, containing 0.025 M EDTA.

[0040] Preferably, in step e), the step of isolating the human proBDNF mutein inclusion bodies is carried out by serial centrifugation of the host cell lysate. Preferably, in step f), the denaturing solution comprises the following components: i. 1 to 8 M, preferably 3 to 6 M, more preferably 4 M guanidinium-HCl, 1 to 100 mM, preferably 5 mM cysteine, ii. 0.001 to 1 M, preferably 0.1 M Tris; iii. 1 to 50 mM, preferably 10 mM EDTA and has a pH between 7 and 10, preferably 8.

[0041] After solubilization of the human proBDNF mutein, the protein is refolded to obtain a correctly folded human proBDNF mutein. For the refolding process, it is important to minimize competing reactions of misfolding and aggregation. Because protein aggregation becomes dominant at high protein concentrations, refolding is performed at a very low protein concentration (approximately 250 μg per ml) to prevent aggregation.

[0042] Preferably, in step g), the refolding solution comprises: i. a chaperone, preferably arginine, at a concentration of preferably 0.5 to 1.0 M, more preferably 0.75 M; ii. a metal chelating agent, preferably EDTA, at a concentration of preferably 1-10 mM, more preferably 5 mM; iii. A redox shuffling system, preferably selected from a combination of L-cystine and L-cysteine, and a combination of oxidized glutathione and reduced glutathione, more preferably selected from a combination of 1 mM L-cystine and 5 mM L-cysteine, and a combination of 1 mM GSSG (oxidized glutathione) and 5 mM GSH (reduced glutathione). and has a pH between 8 and 11.

[0043] Alternative redox shuffling systems such as cystamine / cysteamine may also be used. Preferably, 360 ml of the solution of denatured human proBDNF mutein is diluted into 6 L of the refolding solution disclosed above. Preferably, the concentration of guanidine in the final solution obtained after dilution into the refolding solution is at most 0.3 M.

[0044] According to the present invention, chaperones are compounds that promote protein folding. Such compounds are known to those skilled in the art. They can assist folding in various ways. Arginine is a preferred chaperone according to the present invention. Arginine destabilizes misfolded intermediates, resulting in them being at least partially unfolded (from thermodynamic dead ends), so that they can be folded correctly again.

[0045] Preferably, step g) further comprises adjusting the pH of the final solution in which the human proBDNF mutein is refolded to a value between 8.5 and 9.5, more preferably to a value between 8.8 and 9.2, even more preferably to a value of 9.1.

[0046] By carrying out the method according to the invention, which involves denaturation and subsequent refolding, an aqueous solution of correctly folded human proBDNF mutein is obtained.

[0047] Preferably, the solution of purified human proBDNF mutein obtained in step (ii) of said method for recombinantly producing human BDNF comprises urea, more preferably comprises citric acid and urea.

[0048] Preferably, in the solution, urea has a concentration of less than 2 M, more preferably a concentration of 1 M. Preferably, in the solution, citric acid has a concentration of between 5 mM and 100 mM, more preferably a concentration of 50 mM. Preferably, the solution has a pH in the range of 3 and 5.5, preferably a pH of 4.

[0049] Preferably, step (ii) comprises purifying the human proBDNF mutein from the solution provided in step (i) via chromatographic purification.

[0050] More preferably, the chromatographic purification uses an elution solution comprising urea, more preferably citric acid and urea. According to this embodiment, the eluate obtained in the chromatographic purification can be used directly in step (iii).

[0051] Preferably, in the elution solution, urea has a concentration of less than 2 M, more preferably a concentration of 1 M. Preferably, in the elution solution, citric acid has a concentration of between 5 mM and 100 mM, more preferably a concentration of 50 mM. Preferably, the elution solution has a pH in the range of 3 and 5.5, preferably a pH of 4.

[0052] Preferably, said chromatographic purification is carried out by mixed mode chromatography. In certain embodiments, the pH of the solution provided in step (i) is adjusted to 8 before loading onto the chromatography column.

[0053] In an alternative embodiment, the pH of the solution provided in step (i) is not altered before loading onto the chromatography column, and therefore the pH is comprised between 8 and 11, preferably the pH is comprised between 8.5 and 9.5, more preferably the pH is comprised between 8.8 and 9.2, and even more preferably the pH is 9.1.

[0054] The most preferred column used in the chromatographic purification of step ii) is a column with a synthetic affinity ligand, preferably 4-mercaptoethylpyridine (MEP Hypercell, Sartorius). The advantages of this medium are that binding is independent of ionic strength, no salt stacking is required, and higher flow rates are possible for immobilization processes. Furthermore, elution is achieved by pH shift.

[0055] Other mixed-mode material columns are known and may be used. For example, but not limited to, MEP (Sartorius, affinity ligand is 4-mercaptoethylpyridine), HEA (Sartorius, affinity ligand: hexylamino), PPA (Sartorius, affinity ligand: phenylpropylamino), MBI (Sartorius, affinity ligand: 2-mercapto-5 benzamidazole sulfonate), Capto MMC (GEHC), Capto adhere (GEHC, affinity ligand: N-benzyl-N-methylethanolamine), CHT hydroxyapatite (BioRad), CHT fluoroapatite. MEP, HEA, PPA, and MBI columns have hydrophobic bonds; Capto MMC is a cation exchanger with mixed-mode functionality; and Capto adhere is an anion exchanger with mixed-mode functionality. BioRad columns are ion exchange columns with hydrophobic components.

[0056] Preferably, in step (iii) of the method for recombinantly producing human BDNF, the cleavage comprises adjusting the pH of the solution of purified human proBDNF mutein obtained in step (ii) to a pH value between 5.8 and 9, more preferably between 5.8 and 8, more preferably between 5.8 and 7, more preferably between 5.8 and 6.5, even more preferably to a pH value of 6, and cleaving the human proBDNF mutein in the solution to obtain biologically active human BDNF, more particularly recombinant human BDNF.

[0057] Preferably, said cleaving of the human proBDNF mutein is carried out by a trypsin-like protease, more preferably by trypsin. Proteases with trypsin-like substrate specificity cleave proteins without digesting the active portion of the protein molecule. Trypsin-like proteases cleave peptide bonds after positively charged amino acids such as arginine or lysine. Preferably, trypsin is used to cleave the prosequence, but other proteases may be used instead. Note that cleavage is not limited to trypsin itself, but may also involve other proteases with trypsin-like substrates. Generally, the ratio of human proBDNF mutein to trypsin (or other proteases) is appropriately adjusted so that correctly folded mature human BDNF (more specifically, recombinant human BDNF) is not cleaved by this protease. In contrast, denatured proteins and folding intermediates expose sequences that are susceptible to attack by proteases.

[0058] Preferably, for cleavage of human proBDNF mutein to human BDNF (more particularly recombinant human BDNF), the ratio of trypsin-like protease, preferably trypsin, to human proBDNF mutein is between 1:20,000 and 1:30,000 w / w, more preferably between 1:23,000 and 1:27,000 w / w, with a ratio of 1:25,000 w / w being most preferred. In a preferred embodiment, cleavage is carried out at room temperature for at least 12 hours, preferably more than 13 hours, more preferably between 12 and 22 hours, even more preferably between 13 and 17 hours, most preferably 15 or 16 hours.

[0059] Preferably, step (iii) further comprises the step of inhibiting the activity of trypsin-like proteases. Preferably, the inhibiting step is carried out by adjusting the pH of the solution of human BDNF obtained in step (iii) to a value between 4.5 and 5.5, preferably 5.

[0060] Preferably, the inhibiting step is carried out by diluting the solution obtained in step (iii) with a solution of arginine, citric acid and urea to obtain a final solution having a pH value between 4.5 and 5.5 and an arginine concentration between 0.1 M and 0.2 M, more preferably a concentration of 0.125 M. Preferably, the solution comprises 1 M arginine, 50 mM citric acid and 1 M urea, which is added to the solution of step (iii) with a dilution ratio of 1 / 8 to achieve a reduction in pH to approximately 5.2 and a final arginine concentration of 0.125 M. Under these conditions, arginine i) naturally lowers the pH; ii) acts as a chaperone, preventing protein aggregation; iii) inhibits trypsin activity as a substrate competitor; and iv) is compatible with the chromatography step (iv) of the purification of said human BDNF, if performed as follows. This has the advantage that:

[0061] According to a preferred embodiment, the method of the present invention further comprises a step (iv) of purifying the human BDNF obtained in step (iii). Preferably, said purifying step in step (iv) comprises separating trypsin and product-related impurities of trypsin digestion from BDNF.

[0062] Preferably, the purifying step also reduces host cell proteins (HCPs), endotoxins, and DNA impurities. Any method known in the art for protein purification can be used in step (iv).

[0063] Preferably, said purifying step in step (iv) comprises one or more chromatographic purification steps. Preferably, a Sepharose column, more preferably an SP Sepharose High Performance, Phenyl Sepharose 6 Fast Flow or Q Sepharose Fast Flow column, is used in said one or more chromatographic purification steps.

[0064] Preferably, step (iv) involves chromatographic purification of human BDNF on a SP Sepharose High Performance column, more preferably followed by further purification on a Phenyl Sepharose Fast Flow low substitution column.

[0065] The final product BDNF made from human proBDNF muteins can be analyzed for purity by SDS-PAGE, RP-HPLC, SE-HPLC, IEX-HPLC, RP-UPLC, SE-UPLC, and IEX-UPLC.

[0066] Preferably, the human BDNF (more particularly recombinant human BDNF) obtained by the method according to the present invention has a purity of at least 94%, preferably at least 95%, more preferably at least 96%, more preferably at least 97%, more preferably at least 98%, and even more preferably at least 99%, as measured by RP-UPLC.

[0067] This high level of purity is achievable due to specific features of the method according to the present invention that have been identified by the inventors. The present inventors have found that cleavage of the pro-sequence from the human proBDNF mutant protein according to the present invention with a trypsin-like protease to produce human BDNF (more specifically, recombinant human BDNF) has higher specificity than cleavage of the pro-sequence from wild-type proBDNF. Thus, this approach allows for the efficient production of correctly folded full-length BDNF (more specifically, recombinant human BDNF) with reduced generation of erroneously cleaved by-products and efficient and easier protein purification.

[0068] In this regard, Example 9 highlights the disadvantages and difficulties of cleaving wild-type proBDNF with trypsin-like proteases, which in fact results in very low yields and quality of correctly digested mature full-length BDNF (more specifically, recombinant human BDNF), along with unexpectedly high amounts of improperly digested and over-digested BDNF forms.

[0069] Furthermore, as also explained in more detail in the experimental section below, the inventors have found that the use of a solution containing urea as a cleavage buffer is advantageous because under these conditions the protein is cleaved correctly and no precipitation is observed.

[0070] Furthermore, the inventors have found that when the purification step (ii) is carried out by chromatographic purification using a solution containing urea and citric acid as the eluent, the cleavage step (iii) of the human proBDNF mutein and the purification step (iv) of the resulting BDNF can all be carried out in the same solution eluted from the chromatographic column in step (ii) by simply adjusting the pH to the value required for the particular process step by adding acid or base to the solution containing citric acid and urea. This is a significant advantage because it simplifies the operational flow, requires the process to use fewer materials, facilitates further purification steps, and allows for the achievement of a highly pure final product. In particular, the use of an eluent containing citric acid and urea in step (ii) avoids the need for buffer exchange by tangential flow filtration at the end of the purification step and before the cleavage step. Furthermore, the association of citric acid and urea not only allows the pH of the solution to be easily changed, but also prevents protein aggregation and, surprisingly, does not interfere with trypsin activity.

[0071] A further object of the present invention is the use of a human proBDNF mutein according to any one of the above-described embodiments of the first object of the invention in a method for producing human BDNF, more particularly recombinant human BDNF.

[0072] A further object of the present invention is human BDNF, more particularly recombinant human BDNF, obtainable or obtained by the method according to the invention. A further object of the present invention is human BDNF (more particularly recombinant human BDNF) having a purity of at least 94%, preferably at least 95%, more preferably at least 96%, more preferably at least 97%, more preferably at least 98%, and even more preferably at least 99%, as measured by RP-UPLC.

[0073] A further object of the present invention is to provide a method for producing a compound of the sequence: RGELSVCDSISEWVTAADKKTAVDMSGGTVTVLEKVPVSKGQLKQYFYETKCNPMGYTKEGCRGIDKRHWNSQCRTTQSYVRALTMDSKKRIGWRFIRIDTSCVCTLTIKRGR (SEQ ID NO: 23) or GELSVCDSISEWVTAADKKTAVDMSGGTVTVLEKVPVSKGQLKQYFYETKCNPMGYTKEGCRGIDKRHWNSQCRTTQSYVRALTMDSKKRIGWRFIRIDTSCVCTLTIKRGR (SEQ ID NO: 22) The present invention relates to a human BDNF (more particularly, recombinant human BDNF) composition containing a peptide impurity having the formula:

[0074] A further object of the present invention is a pharmaceutical composition comprising the above-described human BDNF (more particularly recombinant human BDNF) and at least one pharmaceutically acceptable ingredient. In certain embodiments, the pharmaceutically acceptable ingredients may be diluents, carriers, fillers, salts, buffers, stabilizers, penetration enhancers. Techniques for formulating the pharmaceutical compositions of the present invention may be found in standard textbooks such as the latest edition of "Remington's Pharmaceutical Sciences," Mack Publishing Co., Easton, PA.

[0075] The present invention is further described in the following examples, which do not limit the scope of the invention, which is defined in the claims. [Example]

[0076] Example 1 - Cloning of DNA encoding proBDNF muteins into an expression vector and insertion into E. coli A synthetic gene corresponding to the human proBDNF mutein of SEQ ID NO: 4 was prepared, triplet codon-optimized for expression in E. coli. For cloning purposes, an NcoI restriction site was engineered at the initiating methionine residue (CCATGG), and an XhoI restriction site was inserted after the final stop codon.

[0077] After digestion with NcoI and XhoI restriction enzymes, the corresponding DNA fragment encoding the proBDNF ORF was introduced into the pET28a expression vector (Novagen) cut with the same enzymes.

[0078] The obtained expression vector was transformed into the E. coli strain BL21(DE3) by temperature shock according to the supplier's instructions (BL21(DE3) Chemi Competent cells ref 156-3003, BioRad).

[0079] Example 2 - Fermentation and induction of protein expression A single colony of the recombinant E. coli strain was inoculated into 30 ml of Terrific Broth (TB) medium (Sigma Aldrich) containing 30 μg / ml kanamycin, which was then incubated overnight at 37° C. under agitation. The next day, 300 ml of fresh TB containing 30 μg / ml kanamycin was inoculated with 3 ml of the overnight culture, and the OD 600nm When the ρ reached 1, 1 mM isopropyl β- d -1-thiogalactopyranoside (IPTG) was added to the medium to induce the expression of the recombinant protein, and the culture was continued for another 3 h.

[0080] To monitor the overexpression of the target protein during the fermentation process, samples were analyzed by SDS-PAGE before and after induction. Because the human proBDNF mutant protein has three intrachain disulfide bridges, the protein accumulated in the cytoplasm of cells in the form of inclusion bodies (iBs).

[0081] Example 3 - Isolation and solubilization of inclusion bodies The cells were harvested by centrifugation, resuspended in lysis buffer (0.1 M Tris, 0.025 M EDTA, pH 7.0) at 1.5 wt / vol and homogenized using an NS1001L2K Niro Soavi high-pressure homogenizer at 800 bar for four cycles. The resulting homogenate was diluted 1:0.5 with a preparation of Brij 35 (60 gr / L) and stirred for 30 minutes.

[0082] The pellet corresponding to the inclusion bodies was isolated by centrifugation (8500 rpm, Sorval GS3 rotor) and washed twice with 0.1 M Tris, pH 7.0, 0.4-1 wt / vol. The mixture was centrifuged (7500 rpm, Sorval GS3 rotor), and the pellet was dissolved in guanidine solution to obtain a final concentration of 4 M. Specifically, the following guanidine solutions were used: i. 4 M guanidinium-HCl, 5 mM cysteine; ii. 0.1M Tris, iii. 10 mM EDTA; iv. pH 8.0.

[0083] Example 4 - Refolding of rh-proBDNF muteins To prepare correctly folded human proBDNF muteins according to the present invention (i.e., rh-proBDNF muteins), 360 mL of solubilized material obtained according to Example 3 was diluted into 6 L of a refolding solution having the following composition: 0.75 M arginine-HCl, 5 mM EDTA, 5 mM cysteine-HCl, and 1 mM cystine, and maintained at a temperature between 0 and 10°C, at which point the human proBDNF muteins adopt a correctly folded conformation. The concentration of guanidine after dilution into the refolding solution was up to 0.3 M. The pH of the refolding solution was adjusted to 9.1 using sodium hydroxide.

[0084] The performance of the refolding reaction was analyzed by RP-HPLC and estimated to be 55% of the initially solubilized protein. Example 5 - Purification of human proBDNF muteins from refolding solution To purify the human proBDNF muteins from the refolded solution, hydrophobic-induced charge chromatography was performed using a column with 4-mercaptoethylpyridine (MEP) as a synthetic affinity ligand. Before loading the refolded protein solution, the column was equilibrated with 5 column volumes of 0.75 M arginine-HCl, 5 mM EDTA, pH 9.1.

[0085] Two different experimental setups, designated 5a and 5b, were tested for elution of the protein from the column and are summarized below. Setup 5a: In this setup, refolded human proBDNF muteins were loaded onto an MEP column after filtration, the column was washed with 0.1 M Tris, pH 8.0, and the protein was eluted in 0.75 M arginine-HCl, 0.05 M acetate buffer at pH 4.0. The eluted protein-containing fraction was then buffer-exchanged by tangential flow filtration on a cassette with a 10 kDa cutoff, followed by six successive diafiltrations with 0.025 M sodium phosphate, pH 6.5, to remove arginine and favor the subsequent enzymatic cleavage of the human proBDNF muteins, as discussed in the following examples.

[0086] Setup 5b: In this setup, refolded human proBDNF muteins were loaded directly in the refolding solution at pH 9.1, the column was washed with Tris buffer at pH 8.0, and the protein was eluted with a solution containing 50 mM citric acid and 1 M urea at pH 4.0.

[0087] Without being bound by theory or principle, the inventors have found that, for reasons explained in the Examples below, setup 5b, in which human proBDNF muteins are eluted with a solution containing citric acid and urea, is more advantageous than setup 5a, in which human proBDNF muteins are eluted in an arginine buffer which is then exchanged into a phosphate buffer.

[0088] Example 6 - Cleavage of human proBDNF muteins to obtain mature rhBDNF After elution from the MEP column, the purified human proBDNF muteins were cleaved to form mature BDNF (i.e., rhBDNF). Trypsin is the preferred enzyme for cleaving human proBDNF muteins to obtain mature BDNF because it cleaves after arginine (R) or lysine (K) and is commercially available in GMP grade. Therefore, the purified protein solution obtained according to Example 5 was incubated with trypsin. The pH of the protein solution was adjusted to 6.5 or 6.0 (see Table 1 below) to favor enzymatic cleavage of human proBDNF muteins by trypsin.

[0089] We tested several experimental setups for the cleavage step, designated 6a to 6g, whose conditions are summarized in Table 1 below and discussed later.

[0090] [Table 1]

[0091] To purify and then cleave the human proBDNF mutein, we first attempted purification setup 5a described in Example 5, followed by each of the cleavage setups 6a to 6e described in Table 1.

[0092] When cleavage setting 6a was performed, the formation of a large amount of precipitate was observed as the digestion progressed during incubation. Without being bound by theory or principle, we hypothesize that this precipitation is due to peptides generated by trypsin activity, which aggregate as digestion progresses. Due to carryover, as the solubilized material aggregates and precipitates, it carries with it the soluble rhBDNF that is formed.

[0093] Since trypsin cleaves after lysine (K) and arginine (R) residues, taking into account the proprotein sequence, all of the following peptides can be generated during digestion: 1. MAPMK (SEQ ID NO: 11) 2. EANIR (SEQ ID NO: 12) 3. GQGGLAYPGVR (SEQ ID NO: 13) 4. THGTLESVNGPK (SEQ ID NO: 14) 5. AGSR (SEQ ID NO: 15) 6. GLTSLADTFEHVIEELLDEDQK (SEQ ID NO: 16) 7. VR 8. PNEENNK (SEQ ID NO: 17) 9. DADLYTSR (SEQ ID NO: 18) 10. VMLSSQVPLEPPLLFLLEEYK (SEQ ID NO: 19) 11. NYLDAANMSMVSAR (SEQ ID NO: 20) Also, a further hexapeptide corresponding to the first six amino acids of BDNF with the following sequence: HSDPAR (SEQ ID NO: 24) can be generated during digestion. Indeed, this peptide with one arginine at the carboxy terminus is accessible during trypsin digestion, and its cleavage results in the following sequence: RGELSVCDSISEWVTAADKKTAVDMSGGTVTVLEKVPVSKGQLKQYFYETKCNPMGYTKEGCRGIDKRHWNSQCRTTQSYVRALTMDSKKRIGWRFIRIDTSCVCTLTIKRGR (SEQ ID NO: 23) This produces a truncated form of BDNF, hereafter referred to as "overdigested", having the following structure:

[0094] It can be assumed that this truncated BDNF form is inactive because it has been demonstrated that the first seven amino acids at the N-terminus of BDNF (HSDPARR, SEQ ID NO: 21) are responsible for the primary interaction between BDNF and its receptor, TrkB. Therefore, during trypsin cleavage, not only the precipitation of the product but also the generation of over-digested BDNF forms must be controlled, which must be removed by successive chromatography along with the peptides corresponding to the digestion of the pro-part.

[0095] To attempt to keep the protein in solution, purification setup 5a described in Example 5 was performed, followed by each of cleavage setups 6b through 6e reported in Table 1 above, in place of cleavage setup 6a. These setups involved adding various materials to the protein solution obtained after procedure 5a to attempt to keep the protein in solution while the digestion proceeded. However, even in the presence of these additives, the protein precipitated as the enzymatic digestion proceeded.

[0096] We then decided to add urea (1 M) to the cleavage buffer. Without being bound by theory or principle, the inventors hypothesize that urea may act as a disaggregating agent and prevent the protein from precipitating from solution. However, the use of urea may also be detrimental to producing the desired BDNF because it may denature the proBDNF mutant protein, making new cleavage sites accessible that were previously buried within the refolded protein.

[0097] Surprisingly, we found that when urea was added to the cleavage buffer, the protein was correctly cleaved in just 2 hours and no precipitation was observed. Based on these results, it was determined that the preferred cleavage settings involved adding 1 M urea to the cleavage buffer.

[0098] To further simplify the process, we then tested a different purification setting, namely setting 5b described in Example 5 above, followed by cleavage setting 6f described in Table 1 above.

[0099] Without being bound by theory or principle, the inventors have found that the combination of conditions of purification setting 5b and cleavage setting 6f is advantageous as it results in a significant simplification of the operational flow of the method according to the invention.

[0100] Indeed, considering the following, setup 5b disclosed in Example 5, in which human proBDNF muteins are eluted with a solution containing citric acid and urea, is more advantageous than setup 5a, in which human proBDNF muteins are eluted in an arginine buffer and then exchanged into a phosphate buffer.

[0101] When setting 5b is used instead of setting 5a, the eluate containing purified human proBDNF mutein eluted from the MEP column can be used directly as a buffer solution for subsequent trypsin cleavage of the human proBDNF mutein into mature BDNF and subsequent ion exchange chromatography to purify mature BDNF by adjusting the pH to the value required for the specific process step by adding acid or base to the citric acid-containing solution. This is possible because citric acid has three pKa values ​​(5.21, 4.28, and 2.92) at 25°C and can therefore cover the entire required buffer range in this process, from pH 4.0 for MEP elution to pH 6.0 or 6.5 for enzymatic digestion.

[0102] The presence of urea in the elution solution of setup 5b is also advantageous not only because it prevents aggregation and precipitation during the subsequent cleavage step of the human proBDNF mutein, but also because it avoids the need to use arginine during the purification step of the human proBDNF mutein, thus avoiding the need to perform tangential flow filtration to formulate the protein solution and digest the arginine with trypsin, as in setup 5a. Indeed, arginine is well known to help keep proteins in solution and thus counteract the natural tendency of proBDNF to aggregate. However, it is also a substrate for trypsin and must therefore be removed before cleaving the human proBDNF mutein into BDNF. Its removal complicates the operational flow of the process, a drawback that is resolved by using the elution solution according to the present invention, which does not involve the use of arginine. The elimination of arginine in the purification step of the human proBDNF mutein also allows for increased efficiency in the subsequent cleavage of the human proBDNF mutein with trypsin. As a result, less trypsin needs to be used, which significantly simplifies the final purification of the obtained mature BDNF.

[0103] Based on all of the above, the association of citric acid with urea is a preferred solution for eluting the MEP column, performing the enzymatic digestion, and loading the digested material onto a further cation exchange chromatography (CEX) column, all in one compatible buffer.

[0104] To allow for some in-process control, the parameters of cleavage setting 6f were slightly modified at the end to increase the digestion time. Therefore, cleavage setting 6g was tested. In this particular setting, the pH of the solution was adjusted to 6.0, a value at which trypsin activity is lower than that at 6.5, and the trypsin / proBDNF mutein ratio was changed from 1 / 7500 to 1 / 25000. The temperature was maintained at 22°C throughout the entire reaction time.

[0105] This trypsin / proBDNF mutant protein ratio is surprisingly very low. Because trypsin needs to be removed from the final product, having a smaller amount of trypsin is advantageous, simplifying the purification process. In fact, the isoelectric point of trypsin is 10.5, which means that after cleavage, trypsin binds to the SP Sepharose column used to purify BDNF from the fragments generated during digestion (see Example 7 below). Therefore, the less trypsin used during the cleavage step, the better, in terms of achieving both higher cleavage specificity and easier purification of the mature BDNF formed.

[0106] For the cleavage of human proBDNF mutant proteins, different digestion times were tested: 13, 14, 15, 16, and 17 hours. These digestion times were long enough to allow for some degree of in-process control. The results are reported in Table 2 below.

[0107] [Table 2]

[0108] The temperature of the reaction was maintained at 22°C for all digestion times used. As can be seen, all digestion times tested (i.e., 13 hours or longer) resulted in BDNF from human proBDNF mutein with sufficient yield and little formation of overdigested forms. Notably, the human proBDNF mutein was almost completely eliminated at 15 hours (2.56% of the initial human proBDNF mutein remained), and the overdigested form was approximately 38%, an amount that could be removed by continuous chromatography. These parameters were also valid for 16 hours, leaving ample time for processing the solution.

[0109] Based on these results, purification set 5b followed by cleavage set 6g were selected as the preferred conditions for the purification of refolded human proBDNF muteins and their subsequent cleavage to mature BDNF, respectively.

[0110] Example 7 - Purification of active rhBDNF After trypsin digestion was completed, the pH of the cleavage solution containing BDNF (i.e., rhBDNF) was adjusted to 5.0 by adding a solution of 1 M arginine, 50 mM citric acid, and 1 M urea in a 1 / 8 ratio to inhibit trypsin activity, and then the solution was loaded onto an SP Sepharose high Performance column to remove trypsin, cleavage by-products, and further impurities. To protect the column, the solution was filtered before loading.

[0111] The column was pre-equilibrated with 50 mM citric acid, 1 M urea, pH 5.0, then washed with 25 mM phosphate buffer, pH 6.5, and the protein was then eluted with a step gradient of increasing salt concentration in the same buffer.

[0112] During elution, different peaks fractionated, with fractions 9 and 10 corresponding to the main peak. All fractions were then analyzed by RP-UPLC and the results are reported in Table 3 below.

[0113] [Table 3]

[0114] The results show that the BDNF protein elutes in a major peak and the percentage of the overdigested form is reduced from 33% present in the loaded material to approximately 10%. Fractions 9 and 10 were pooled and the resulting solution was adjusted to a 2M NaCl concentration.

[0115] At this point, the solution was filtered and loaded onto a Phenyl Sepharose Fast Flow low displacement column, which was pre-equilibrated with phosphate buffer, pH 6.5, 2 M NaCl, and the protein was eluted by decreasing the salt concentration.

[0116] The protein was eluted in a single fraction equivalent to 5 column volumes. Analysis performed on this fraction confirmed that the protein was approximately 94% pure w / w, but this fraction also contained product-related impurities, primarily approximately 1% oxidized rhBDNF form w / w and approximately 5% over-digested BDNF form described above (SEQ ID NO: 23).

[0117] Example 8 - Testing the biological activity of rhBDNF The biological activity of recombinant human BDNF obtained according to the method disclosed above was tested by using rat C6 glioma cells expressing high levels of the BDNF TrkB receptor (Xiong J. et al., Oncology Letters 2015;10:223-227). It was demonstrated that, via this receptor, mature BDNF promotes the growth and survival of C6 cells in a dose-dependent manner (Xiong J. et al., Oncology Reports 2013;30(6):2719-2724). C6 cells were plated in a 96-well microplate at a density of 5000 cells / well and treated with rhBDNF concentrations ranging from 0.23 to 9 μg / mL. After 48 hours of incubation, the cells were cultured in CellTiter 96® AQ. ueous One Solution Cell proliferation assay (MTS assay) was performed to determine the biological activity of rhBDNF from the number of viable C6 cells.

[0118] EC corresponds to the concentration of rhBDNF required to induce 50% cell proliferation. 50was determined for rhBDNF produced according to the method disclosed above by using GraphPad statistical software, and an average value of approximately 3 μg / mL (0.11 μM) was obtained (FIG. 1).

[0119] Example 9 - Analysis of full-length rhBDNF yield and purity obtained from wild-type human proBDNF by methods according to the invention. The inventors also analyzed the yield and purity of correctly folded mature full-length rhBDNF obtained by the method of the present invention from human wild-type proBDNF but not from the human proBDNF muteins according to the present invention.

[0120] Specifically, a synthetic gene corresponding to human wild-type proBDNF, SEQ ID NO: 2, was prepared that was triplet codon optimized for expression in E. coli. Cloning of the DNA encoding wild-type proBDNF into an expression vector and subsequent insertion into E. coli was performed according to the protocol reported in Example 1.

[0121] Fermentation and induction steps for protein expression, isolation and solubilization of inclusion bodies, and refolding steps of human wild-type proBDNF were performed according to the protocols reported in Examples 2, 3, and 4, respectively.

[0122] Purification of human wild-type proBDNF was carried out according to setting 5b of the protocol reported in Example 5. The purified human wild-type proBDNF thus obtained was cleaved with trypsin to obtain human mature full-length BDNF.

[0123] Specifically, 735 ml of purified human wild-type proBDNF, corresponding to approximately 400 mg of proBDNF (and a proBDNF concentration of approximately 550 μg / mL), was digested with trypsin (trypsin ratio of 1 / 25,000) at pH 6.0 and 20°C.

[0124] After 16 hours of digestion, the digested material was analyzed by RP-UPLC and the results are reported in Table 4 below.

[0125] [Table 4]

[0126] As shown in Table 4 reported above, the overdigested form represents approximately 14% of the correctly digested mature rhBDNF. In addition, the BDNF purity was less than 50% (i.e., 48.78%).

[0127] Digestion was stopped by adding a solution of 1 M arginine, 50 mM citric acid, and 1 M urea to the digested material, followed by incubation at room temperature for 1 hour, followed by filtration.

[0128] The digested solution so obtained contained a total of approximately 200 mg of rhBDNF, a mixture of apparently correctly digested mature full-length rhBDNF forms, over-digested rhBDNF forms, and incompletely digested rhBDNF forms, and this solution was loaded onto an SP Sepharose High Performance column equilibrated with 50 mM citric acid, 1 M urea, pH 5.0.

[0129] After loading the digested solution, the column was washed with 25 mM phosphate buffer (pH 6.5), and then the rhBDNF forms were eluted in the same buffer with a step gradient of increasing salt concentration.

[0130] During elution, the different peaks were fractionated and the fractions so obtained were then analyzed by RP-UPLC. The RP-UPLC results are reported in Table 5 below.

[0131] [Table 5]

[0132] The results showed that only less than 25 mg (i.e., 24.48 mg) of rhBDNF was recovered in fractions 5 to 8, which corresponds to approximately 13% of the rhBDNF (200 mg) contained in the digested solution loaded onto the SP column.

[0133] Closer examination of fractions 6 and 7, which were expected to contain primarily properly digested mature full-length rhBDNF, revealed that only approximately 12 mg of rhBDNF was detected, and in addition, an unexpectedly high percentage of over-digested BDNF forms were detected.

[0134] SDS-polyacrylamide (SDS-PAGE) gel analysis of the above fractions 1 to 10 was then carried out to achieve high-resolution separation of the rhBDNF forms contained therein. SDS-PAGE gels were prepared according to the following protocol.

[0135] SDS-PAGE analysis was performed under reducing conditions. Samples were mixed with NuPAGE 4x LDS sample buffer (Invitrogen) and NuPAGE 10x reducing agent (Invitrogen) and stepwise denatured at 100°C for 10 min before loading onto the gel. Approximately 2.33 μg of rhBDNF standard (in-house, Dompe) was loaded onto the gel, while the amounts loaded for analytical test samples (fractions 1–10) depended on their respective measured concentrations. NuPAGE Novex 4–12% Bis-Tris polyacrylamide gels (Invitrogen) were assembled in an XCell Surelock Mini Cell apparatus (Invitrogen). The inner chamber was filled with 200 mL of 1x running buffer supplemented with antioxidant (NuPAGE antioxidant (Invitrogen)), and the outer chamber was filled with approximately 300 mL of 1x running buffer. 1x running buffer was prepared from NuPAGE 20x MES SDS running buffer stock solution (Invitrogen). Gels were run at a constant voltage of 200 V for 35 minutes, after which the gels were fixed and stained using a NOVEX Colloidal Blue staining kit (Invitrogen).

[0136] Figures 2A and 2B report the results of the electrophoretic analysis according to the following gel lanes:

[0137] [Table A]

[0138] As expected, fractions 1 to 5 contained primarily incompletely digested rhBDNF forms (which are higher molecular weight species relative to mature full-length rhBDNF), and fractions 8 and 9 contained primarily over-digested rhBDNF forms (which are lower molecular weight species relative to mature full-length rhBDNF).

[0139] As expected, mature full-length rhBDNF was observed in fractions 6 and 7, but these bands were heavily contaminated with overdigested rhBDNF forms with nearly equal intensity of mature full-length rhBDNF.

[0140] The material eluted in fraction 10 at higher salt concentrations corresponds to abundant protein, as this band shows a strong signal even when diluted 50-fold, and we observed that this band was also contaminated with overdigested forms of rhBDNF.

[0141] To further characterize the eluted material, fractions 6, 7, and 10 described above were analyzed by mass spectrometry according to the following protocol. All samples were injected “neat” into an HPLC-MS Orbitrap Fusion system (ThermoFisher) and analyzed.

[0142] The chromatographic conditions used for the analysis are reported below: - Analytical column: Acquity Premier Protein C4 300Å 100×2.1mm, 1.7μm (Waters), - Eluent (mobile phase): A=H2O 0.05%TFA, B=CH3CN 0.05%TFA, - Flow rate: 0.3mL / min, - Injection volume: 2 / 20μL, - Analysis time: 13 minutes, - Column temperature: 60°C, - Sample temperature: 5℃, - Detection: MS (full scan), Typical retention time: rhBDNF = 6.8 min, - Gradient:

[0143] [Table B]

[0144] The mass spectrometry conditions used for sample analysis are reported below:

[0145] [Table C]

[0146] The mass spectrometry results for fraction 6 are reported in Table 6 below.

[0147] [Table 6]

[0148] The results showed that, as expected, fraction 6 contained correctly digested mature full-length rhBDNF (27.84%), but that it was significantly contaminated with an over-digested rhBDNF form (31.87%) and trace amounts of a form with an extra arginine residue at the NH2-terminus of the protein (rhR-BDNF, SEQ ID NO: 25).

[0149] The mass spectrometry results for fraction 7 are reported in Table 7 below.

[0150] [Table 7]

[0151] As observed in fraction 6, the results for fraction 7 also showed correctly digested mature full-length rhBDNF (30.22%) as expected, but this fraction was significantly contaminated with an over-digested rhBDNF form (34.06%). Trace amounts of a form with an extra arginine residue at the NH2-terminus of the protein (rhR-BDNF, SEQ ID NO: 25) were also observed in fraction 7.

[0152] The mass spectrometry results for fraction 10 are reported in Table 8 below.

[0153] [Table 8]

[0154] The results showed that less than 9% of the digested material corresponded to correctly digested full-length rhBDNF in fraction 10. Indeed, this fraction contained primarily an rhBDNF form with three extra amino acids at the NH2-terminus of the protein (VRR-BDNF, SEQ ID NO: 26, 25.82%) and an rhBDNF form with one extra arginine (R-BDNF, SEQ ID NO: 25, 26.72%); moreover, fraction 10 contained 13% over-digested rhBDNF forms.

[0155] Considering the above, the results showed that performing the method according to the invention from wild-type human proBDNF rather than the proBDNF muteins according to the invention resulted in very low yields of correctly digested mature full-length rhBDNF, as well as unexpectedly high amounts of impurities, namely improperly digested and over-digested BDNF forms (SEQ ID NO: 23).

[0156] Due to the unexpectedly high amount of impurities, it was nearly impossible to purify the correct full-length mature recombinant human BDNF from wild-type human pro-BDNF by carrying out the method according to the present invention.

Claims

1. The following amino acid sequence: MAPMKEANIRGQGGLAYPGVRTHGTLESVNGPKAGSRGLTSLADTFEHVIEELLDEDQKVRPNEENNKDADLYTSRVMLSSQVPLEPPLLFLLEYKNYLDAANMSMX1X2X3X4HSDPARRGELSVCDSISEWVTAADKKTAVDMSGGTVTVLEKVPVSKGQLKQYFYETKCNPMGYTKEGCRGIDKRHWNSQCRTTQSYVRALTMDSKKRIGWRFIRIDTSCVCTLTIKRGR (SEQ ID NO: 3) wherein X1, X2 and X3 are selected from non-basic amino acids and histidine, and X4 is selected from arginine and lysine. A human proBDNF mutant protein having the following structure:

2. 2. The human proBDNF mutant protein according to claim 1, wherein X1, X2 and X3 are independently selected from the group consisting of alanine, asparagine, aspartic acid, cysteine, glutamine, glutamic acid, glycine, isoleucine, leucine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine, valine and histidine (SEQ ID NO: 5), preferably independently selected from the group consisting of valine, alanine, glycine, serine, threonine, tyrosine, asparagine, aspartic acid, glutamine, glutamic acid, methionine and histidine (SEQ ID NO: 6).

3. 3. A human proBDNF mutein according to claim 1 or 2, wherein X1 and X3 are independently selected from alanine and valine, X2 is selected from valine and serine, and X4 is arginine (SEQ ID NO: 10).

4. 4. A human proBDNF mutein according to any one of claims 1 to 3, wherein X1 is valine, X2 is serine, X3 is alanine and X4 is arginine (SEQ ID NO: 4).

5. A nucleic acid encoding a human proBDNF mutein according to any one of claims 1 to 4.

6. 1. A method for producing human BDNF, comprising: (i) providing a solution of a human proBDNF mutein as defined in any one of claims 1 to 4, (ii) purifying the human proBDNF mutein from the solution provided in step (i), thereby obtaining a solution of purified human proBDNF mutein; and (iii) cleaving the human proBDNF mutant protein in the solution to obtain human BDNF. A method comprising:

7. Step (i) comprises the steps of: a) providing a nucleic acid encoding a human proBDNF mutein as defined in any one of claims 1 to 4, b) introducing said nucleic acid into an expression vector; c) introducing the expression vector into a host cell; d) growing the host cells in a suitable liquid culture medium, wherein the host cells express the human proBDNF mutein in the form of inclusion bodies; e) lysing the host cells and isolating the human proBDNF mutein inclusion bodies produced by the host cells in step d); f) dissolving the inclusion bodies in a denaturing solution, thereby obtaining a solution of denatured human proBDNF mutein; g) diluting the solution of denatured human proBDNF mutein in a refolding solution, so that the denatured human proBDNF mutein adopts a correctly folded conformation, thereby obtaining a solution of correctly folded human proBDNF mutein. The method of claim 6, comprising:

8. In step f), the denaturing solution comprises the following components: i. 1-8 M, preferably 3-6 M, more preferably 4 M guanidinium-HCl, 1-100 mM, preferably 5 mM cysteine; ii. 0.001-1 M, preferably 0.1 M Tris; iii. 1 to 50 mM, preferably 10 mM EDTA and having a pH between 7 and 10, preferably 8.

9. In step g), the refolding solution comprises: i. a chaperone, preferably arginine, at a concentration of preferably 0.5-1.0 M, more preferably 0.75 M; ii. a metal chelator, preferably EDTA, preferably at a concentration of 1-10 mM, more preferably 5 mM; iii. A redox shuffling system preferably selected from a combination of L-cystine and L-cysteine, and a combination of oxidized glutathione and reduced glutathione, more preferably a combination of 1 mM L-cystine and 5 mM L-cysteine, and a combination of 1 mM GSSG (oxidized glutathione) and 5 mM GSH (reduced glutathione). and having a pH between 8 and 11.

10. 10. The method according to any one of claims 6 to 9, wherein the solution of purified human proBDNF mutein obtained in step (ii) comprises urea, preferably citric acid and urea.

11. 11. The method according to any one of claims 6 to 10, wherein step (ii) comprises purifying the human proBDNF mutein from the solution provided in step (i) via chromatographic purification.

12. 12. The method according to claim 11, wherein in the chromatographic purification an elution solution comprising urea, preferably comprising citric acid and urea, is used.

13. 13. The method according to any one of claims 6 to 12, wherein in step (iii), the cleavage comprises adjusting the pH of a solution of the purified human proBDNF mutant protein obtained in step (ii) to a pH value between 5.8 and 9, and cleaving the human proBDNF mutant protein in the solution to obtain human BDNF.

14. 14. The method of any one of claims 6 to 13, wherein in step (iii), the cleavage is carried out by a trypsin-like protease.

15. 15. The method according to claim 14, wherein the ratio of trypsin-like protease to human proBDNF mutein is between 1:20000 and 1:30000 w / w, preferably between 1:23000 and 1:27000 w / w, more preferably the ratio is 1:25000 w / w.

16. 16. The method according to any one of claims 6 to 15, further comprising a step (iv) of purifying the human BDNF obtained in step (iii).

Images