Recombinant human kininogenase mutant and application thereof

Abstract

The invention discloses a recombinant human pancreatic kininogenase mutant subjected to engineering modification, a composition containing the kininogenase, and application of the mutant enzyme in treatment of related diseases caused by cerebral apoplexy, stroke and hyperglycemia. The modified mutant protein has high activity, high stability and more uniform glycosylation modification characteristics than wild protein, and when the mutant protein is used for treatment, biological potential safety hazards caused by a traditional biological extraction production process can be avoided, the consistency of products is improved, and more stable and effective drug protein is obtained.

Description

[0001] Specification Technical Field

[0002] The present invention relates to methods and combinations for treating diseases or conditions such as stroke and thrombolysis. More specifically, the present invention relates to recombinant mutant human kallidinogenase and methods of using the same, as well as the use of compositions containing such kallidinogenase in the field of first aid and rehabilitation for stroke. Technical Background

[0003] Human tissue kallidinogenase-1 (Kallikrein 1,), abbreviated as KLK1 or hK1, belongs to the peptidase S1 family of serine proteases, or the kallidinogenase subfamily. In humans or animals, it mainly acts on kallidinogen and hydrolyzes to release bradykinin, playing a regulatory role in the body's blood pressure, electrolyte balance, inflammatory response, and cell proliferation.

[0004] KLK1 is highly abundant in human urine. It has been isolated and purified as a drug both at home and abroad and has achieved ideal effects in clinical practice. Bayer AG in Germany has used the crude product of hK1 extracted from urine as an antihypertensive drug since 1930, and its trade name is Padutin. Guangzhou Tianpu Biopharmaceutical Co., Ltd. has used purified hK1 from human urine in clinical practice and was approved for the treatment of acute ischemic stroke (AIS) in 2005, and its trade name is Kailikang. In addition, KLK1 derived from porcine pancreas has been approved for the treatment of diabetic kidney disease (DKD) and hypertension in China, Japan, and South Korea.

[0005] The production method of urine extraction and purification has potential problems in terms of source, biosafety, and inter-batch stability. Therefore, as early as 1988, Amgen attempted to clone and express human KLK1 protein, whose sequence was derived from the natural KLK1 gene. The eukaryotic expression vector pDSHK1 containing the SV40 late promoter and the human KLK gene or its fragment was transfected into African green monkey kidney cells, COS-1, and CHO cells for expression (European Patent EP0297913B1). Limited by the technical elements such as the sequences, expression vectors, and cells used in this invention, the yield of KLK1 produced by recombinant expression was low, and the cost problem could not be solved. Shanghai Wanxing Biopharmaceutical Co., Ltd. declared the clinical research of recombinant human kallidinogenase-1 injection expressed by Pichia pastoris in 2009 (CXSL0700032), and then the company continued to cooperate with Shanghai Tengrui Pharmaceutical Co., Ltd. to develop this variety. Guangzhou Tianpu attempted to express the fusion protein of KLK1 and human antibody Fc in the form of a fusion protein, and the obtained expression level was low, only 11 mg / L, and the specific activity was estimated to be 7.8 pNA / mg. After adding two linker peptides between KLK1 and Fc, the reported enzyme activity increased to 9.2 pNA / mg, and the yield of stably transfected cells was 0.7 mg / L, and the specific enzyme activity relative to the natural protein was 13.3 pNA / mg, which was also of no practical value. DM199 developed by DiaMedica Therapeutics, Inc. in Canada is a recombinant human tissue kallidinogenase 1. DM199 is a recombinant human tissue kallidinogenase (rhKLK1) for acute ischemic stroke (AIS) and diabetic kidney disease (DKD). In 2018, this product was used for the treatment of acute ischemic stroke, and it was expected to be used for the treatment of diabetic kidney disease (DKD) and acute ischemic stroke (AIS) through mechanisms such as vasodilation, anti-inflammation, cell repair, and reduction of apoptosis. The above-mentioned Phase Ib bridging clinical study reported the pharmacokinetics and pharmacodynamics of DM199 produced by intravenous injection, that is, the PK / PD activity was equivalent to that of the reference drug with the trade name Kelikang produced by Guangdong Tianpu. This variety has completed a Phase II clinical trial in Australia. A large-scale Phase II / III clinical study to verify patients with small vessel stroke without thrombolysis and thrombectomy has been carried out in the United States. The FDA granted DM199 fast track designation for the treatment of AIS in September 2021. Changzhou Qianhong Pharmaceutical Co., Ltd. developed a PEGylated long-acting recombinant pancreatic kallidinogenase, numbered QHRD106. This drug completed a Phase I clinical trial in December 2023, mainly for the safety assessment and drug metabolism evaluation of healthy people, and currently, the recruitment for Phase II clinical trial is in progress.

[0006] Currently, all marketed KLK1 drugs are natural purified proteins, mainly derived from human urine and porcine pancreas. The sources of natural extracted proteins are unstable, the purification process is complex, the yield is low, and the drug price is expensive. In addition, host residues cannot be completely removed during the drug purification process, resulting in high immunogenicity to the human body during drug use. Preparing KLK1 by recombinant expression can partially solve the above problems. Guangdong Tianpu is also trying to express recombinant human kallikrein in CHO. Its patent CN101134953A discloses a high-molecular-weight KLK1 containing 3 glycosylation modification sites. The patent US20130323222A of DiaMedica discloses the expression of recombinant human kallikrein in CHO, and highly glycosylated KLK1 and low-glycosylated KLK1 are separately purified. The in vivo activities of the two are not very different, but the activity of the mixture of high- and low-glycosylated KLK1 in a 1:1 ratio is higher than that of pure high-glycosylated KLK1 or low-glycosylated KLK1. The patent CN116135973A of Jiangsu Zhonghong Bioengineering Innovation Research Institute Co., Ltd. discloses a single glycosylation site mutant KLK1 with N-terminal site-directed PEG modification. It is reported that the in vitro enzyme activity of this mutant protein is higher than that of the wild type, and the molecular weight distribution is more concentrated than that of the wild type sequence during SDS-PAGE electrophoresis detection. After PEG modification, it is expected to improve the stability and in vivo half-life of the protein.

[0007] The above studies all indicate that the recombinant production of KLK1 has high economic value and is feasible. However, for decades, the production of recombinant KLK1 has not been applied. From the research process and results, the technical reasons can be roughly attributed to three reasons. One is the yield limitation. The yield of recombinant expression is low and the cost is high, which is not enough to replace biological extraction. The second is the diversity of post-translational modifications of the expression products, making it difficult to form a uniform product, which affects purification and application. Especially glycosylation modification causes the dispersion of molecular weight and isoelectric point, and different modification levels also affect the enzyme activity stability and specific enzyme activity level. The third is that the recombinant form of KLK1 has an important impact on the properties of the product. Whether it is the zymogen with a leader peptide or the mature zymogen sequence, the enzyme activity will be affected by the activation efficiency and sequence integrity of the analysis. To solve the recombinant expression and production application of KLK1, breakthroughs must be made in the above three aspects. Summary of the Invention

[0008] The first aspect of the present invention is to solve the problem of sequence modification of highly active kallikrein.

[0009] Kallikrein is a class of serine proteases that are highly conserved in evolution. Human tissue kallikrein is synthesized in the form of zymogen (prokallikrein), and the enzyme must be activated by hydrolyzing the seven amino acid propeptides after the signal peptide. This enzyme can be activated by thermolysin, trypsin, and human plasma kallikrein in vitro, but the in vivo activation mechanism is unclear. The human KLK gene family, to which this enzyme belongs, has a total of 15 members. The residues of the enzyme active center (His, Asp, Ser) and the surrounding amino acids within the family are highly conserved. Each family member contains at least one glycosylation motif Asn-Xaa-Ser / Thr, where Xaa ≠ Pro. This means that the first amino acid of the three-amino-acid sequence is Asn, the second amino acid can be any of the 19 amino acids other than Pro, and the third can be either Ser or Thr. The sequences of tissue kallikrein 1 (Kallikrein 1, KLK1) from different mammalian sources are highly similar, highly consistent in disulfide bonds, glycosylation sites, and enzyme active sites, and also highly consistent in enzymatic properties. Taking the human KLK1 protein with the accession number P06870 in the Uniprot protein database as the reference sequence, in the description and examples of the present invention, the amino acid positions of KLK1 are based on the amino acid sequence numbers of the mature kallikrein protein numbered P06870 in the Uniprot database. Both the 102nd Asn and 108th Asn of KLK1 conform to the glycosylation motif Asn-Xaa-Ser / Thr rule, while the 165th Asn glycosylation site is only conservatively present in primates. In addition to the above N-glycosylation modification, human KLK1 also has 3 potential O-glycosylation sites, S93, S104, and S167. Additionally, disulfide bonds also have an important impact on the high-level structural stability and enzyme activity of KLK1. Genetic variation of KLK1 is the cause of reduced urokinase activity. In addition to the three natural variants AAU12569, AAP35917, and AAH05313, the single amino acid R77H mutant caused by single nucleotide polymorphism (SNP) also causes a 50% to 60% reduction in kallikrein activity. Glycosylation of proteins is crucial for enzyme activity and is also important for maintaining the activity stability of enzymes. Due to the complexity of the enzymes and processing processes involved in post-translational modifications such as glycosylation in mammalian cells, the form of glycosylation, the molecular composition and ratio of sugar chains cannot currently be determined by amino acid sequences. For this reason, both naturally extracted and recombinantly expressed glycoproteins have differences in molecular weight and isoelectric point, manifested as non-uniform molecular weight and varying degrees of molecular weight and charge heterogeneity phenomena.In the present invention, single-point mutations and combined mutations were carried out on all important modification sites that may affect the function of KLK1. The results showed that the glycosylation modification at position N102 is crucial for the stability, correct folding and secretion of the protein. After removing the glycosylation modification by mutating N102, the expression level of the single-point or combined mutant protein was extremely low or there was no expression. However, the deglycosylation mutation at position N165 had no significant effect on the activity and expression level of the protein, which is consistent with the highly conserved nature of position N102 in evolution while position N165 is only conserved in primate KLK1. The glycosylation site at position N108 is also a conserved site in different species. Theoretically, it is speculated that the glycosylation modification at this site also plays an important role in enzyme activity and stability. Surprisingly, in the activity experiment of the mutants designed at this site, it was found that the enzyme activity of the single-point mutant at position N108 was the same as that of the wild type, and it had similar activity to the deglycosylation mutant at position N165 which is not conserved except in primates. On this basis, different mutant amino acid selections and combinations at the three sites were screened and their enzyme activities were measured. Unexpectedly, it was found that the in vitro enzyme activity of the double-point mutant at positions N108 and N165 was the same as that of the wild type, and the protein expression level was higher than that of the wild type. This may be due to the fact that the protein folding and structure are more stable after introducing the mutations, and different amino acid combinations provide more structural combinations.

[0010] The present invention provides a novel recombinant human kallikrein (KLK1) mutant protein. This mutant has a single structure, good stability, and has an activity similar to that of wild-type kallikrein. In a specific combination, it has an enzyme activity 120% of that of the wild-type enzyme. Specifically, the present invention includes the following aspects:

[0011] The first aspect of the present invention relates to a KLK1 mutant, and the mutation positions involve C31, C66, R77, S93, N102, M103, S104, N108, H109, T110, D114, V139, E145, N165, F166, S167, K186, A188, V193. Preferably, the mutant amino acid positions are S93, N102, S104, N108, N165 and S167. More preferably, the mutant amino acid positions are N102, N108 and N165.

[0012] In a preferred example, the amino acids at positions N102, N108 and N165 can be replaced by amino acids selected from glutamine, glutamate, lysine, aspartic acid, and tyrosine.

[0013] In another preferred example, the amino acids for replacement are selected from: aspartic acid, glutamine, lysine, glycine, or a combination thereof.

[0014] In another preferred example, the amino acid replacement combinations of the mutant are selected from:

[0015] (1) The asparagine at position 102 is replaced by glutamine, the asparagine at position 108 is replaced by glutamine, and the asparagine at position

[0016] 165 is replaced by glutamine;

[0017] (2) The original asparagine is retained at position 102, the asparagine at position 108 is replaced by glutamine, and the asparagine at position 165 is replaced by aspartic acid;

[0018] (3) The original asparagine is retained at position 102, the asparagine at position 108 is replaced by aspartic acid, and the asparagine at position 165 is replaced by glutamine;

[0019] (4) The original asparagine is retained at position 102, the asparagine at position 108 is replaced by aspartic acid, and the asparagine at position 165 is replaced by aspartic acid.

[0020] In another preferred embodiment, the KLK1 mutant of the present invention retains at least 80% or more (≥80%) of the enzyme activity of the wild type, preferably 90%, more preferably 100% - 120%.

[0021] In another preferred embodiment, the specific activity of the recombinant mutant of the KLK1 mutant of the present invention is 1000 U / mg, preferably 1100 U / mg, more preferably 1200 U / mg.

[0022] In the second aspect of the present invention, an isolated polynucleotide is provided, and the sequence encodes the KLK1 mutant described in the first aspect of the present invention.

[0023] In another preferred embodiment, the amino acid sequence of the mutant KLK1 encoded by the polynucleotide is as shown in SEQ ID NO.: 3 - 6, and the sequence is introduced with a recombinant mutation selected from the group consisting of: N102Q, N108Q, N165Q, N102D, N108D, N165D, or a combination of the above mutations.

[0024] In another preferred embodiment, the polynucleotide additionally adds an expression and purification element selected from the group consisting of:

[0025] A signal peptide for secretion, such as the native signal peptide sequence MWFLVLCLALSLGGTGA of KLK1, or a signal peptide from murine IgG, a human insulin signal peptide.

[0026] Propeptides, such as the native Propeptide sequence APPIQSR of KLK1, or propeptide sequences of KLK1 proteins from other species, such as APPIRSR in Pongo abelii, VFPIQSR in Bos taurus, and APPVQSR in rats.

[0027] Purification tag sequences, such as polyhistidine tags formed by six, or seven, or eight, or nine, or ten histidines in tandem. Other purification tags can also use polypeptide tags, such as a FLAG tag (DYKDDDDK) or tags formed by several FLAG tags in tandem, a StrepTag (WSHPQFEK), and protein tags. Other available tags can also be GST, SUMO, or Trx.

[0028] The third aspect of the present invention provides a vector that contains the polynucleotide encoding the KLK1 mutant described in the second aspect of the present invention.

[0029] The fourth aspect of the present invention provides a host cell that contains the vector described in the third aspect of the present invention or integrates the polynucleotide described in the second aspect of the present invention into the genome.

[0030] The fifth aspect of the present invention provides a method for producing the KLK1 mutant described in the first aspect of the present invention, including the following steps:

[0031] Under suitable expression conditions, culture the host cell described in the fourth aspect of the present invention, thereby expressing the mutant KLK1 described in the first aspect of the present invention. The host cells include, but are not limited to, Escherichia coli, yeast, and mammalian cells.

[0032] Within the scope of the present invention, the above-mentioned technical features of the present invention and the technical features specifically described below (such as in the examples) can be combined with each other to form new or preferred technical solutions.

[0033] Advantages of the Invention

[0034] The present invention obtains a novel recombinant human kallikrein mutant. Compared with the wild-type KLK1 enzyme, this mutant has fewer glycosylation sites in structure, a more uniform molecular weight in basic property detection, better stability, and higher biological activity than the wild type. As a biotherapeutic drug, due to changes in its sequence design and glycosylation modification method, it has a higher yield than the wild type. The expression level in stably transfected CHO-S cells reaches 1 g / L, significantly higher than the yield of the wild type at 0.5 g / L, and is more suitable for drug production. Brief Description of the Drawings

[0035] Figure 1 Alignment of amino acid sequences of KLK1 proteins from different species. The black boxes indicate the positions of glycosylation motifs where N102, N108, and N165 are located.

[0036] Figure 2 . Simulation results of the wild-type KLK1 structure, indicating the positions of potential mutation sites

[0037] Figure 3. Expression and purification results of different single-site mutants of KLK1. The left side shows the expression results in Expi-293 cells, and the right side shows the expression results in Expi-CHO cells. The mutation site situations are marked above the lanes.

[0038] Figure 4. Expression and purification results of multi-site combinatorial mutants of KLK1. The left side shows the electrophoresis results of the purified products of wild-type and mutant KLK1 expressed in Expi-293, and the right side shows the electrophoresis results of the purified products of wild-type and mutant KLK1 expressed in Expi-CHO cells. Among them, MW is the protein molecular weight marker, WT is wild-type KLK1, M1 is the N108D single-site mutant, M2 is the N165D single-site mutant, M3 is the N108D and N165D double-site mutant, M4 is the N108D and N165Q double-site mutant, M5 is the N108Q and N165D double-site mutant. In both cells, the molecular weight of the wild-type is significantly higher than that of the single-site and double-site mutants due to the highest degree of glycosylation. The molecular weight of the 108-site deglycosylation mutant decreases, but two bands still appear, indicating diverse glycosylation modifications at the remaining 165-site glycosylation site. The 165-site single-site deglycosylation mutant shows one band, but the molecular weight is similar to that of the wild-type, indicating that the sugar chain at the 108-site occupies a certain molecular weight. The molecular weights of the three double-site deglycosylation mutants M3, M4, and M5 are similar and are all smaller than those of the wild-type and single-site mutants, indicating that their structures are more consistent.

[0039] Figure 4. Protein gel images of KLK1 mutants after activation by Thermolysin ( Figure 4a ) and Trypsin ( Figure 4b ). Figure 4a The left side shows the electrophoresis results of the purified products of wild-type and mutant KLK1 expressed in Expi-293, and the right side shows the electrophoresis results of the purified products of wild-type and mutant KLK1 expressed in Expi-CHO cells. Among them, MW is the protein molecular weight marker, WT is wild-type KLK1, M1 is the N108D single-site mutant, M2 is the N165D single-site mutant, M3 is the N108D and N165D double-site mutant, M4 is the N108D and N165Q double-site mutant, M5 is the N108Q and N165D double-site mutant. Figure 4bElectrophoresis results of the purified product of the KLK1 mutant expressed in CHO-S cells, where MW is the protein molecular weight marker, 1 is the unactivated zymogen, and 2 is the kallikrein after activation with trypsin.

[0040] Figure 5 Detection results of N-terminal site-directed PEG-conjugated kallikrein protein KLK1 by SDS-PAGE. MW is the protein molecular weight marker, 1 is the unconjugated kallikrein protein, and 2, 3, and 4 are the PEG-KLK1 proteins after conjugation for 1, 3, and 5 hours, respectively.

[0041] Figure 6 TTC staining results of rat brain tissue cell sections

[0042] Figure 7 Statistical analysis results of the cerebral infarction area in the rat ischemia-reperfusion experiment

[0043] Example 1: Determination of mutation sites

[0044] The following factors were considered when screening for mutation sites and mutation methods: (1) whether the amino acid residue is within the motif of post-translational modification, (2) whether the amino acid residue is far from the active center, substrate, or cofactor binding site, (3) whether the mutation disrupts existing helices, hydrogen bonds, or salt bridges, (4) whether the mutation can introduce new forms of interaction, (5) protein secondary structure, hydrophilicity / hydrophobicity, charge, immunogenicity, etc.

[0045] 1.1. Determination of potential mutation sites

[0046] (1) By aligning the KLK1 sequences of human-related mammals, especially primates, back-mutation of residues with low homology can be considered. The sequence alignment results are shown in Figure 1 as follows.

[0047] (2) Use bioinformatics software such as Discovery Studio, AMBER, and Rosetta for alanine scanning to identify hot-spot residues.

[0048] 1.2. Prediction of the function of mutation points. Search for information on mutation sites through the BioMute database and predict whether each site to be mutated is a public nuisance site through PolyPhen.

[0049] 1.3. Conservation calculation To further analyze the functions of the two mutation sites in the transcription factor, we need to analyze the evolutionary conservation of these two sites because evolutionarily conserved sites are of significant importance to the function and structure of proteins.

[0050] 1.4. Modification site analysis From the perspective of mutation types, multiple sites are involved in serine mutations, and serine is usually the phosphorylation modification site of proteins. The phosphorylated sequences were predicted using two software, NetPhos2.0 and GPS3.0. The results showed that the fragment where one of the mutation sites is located is likely to be a phosphorylation hot spot region of the protein.

[0051] 1.5. Protein structure simulation To further verify this hypothesis, we performed protein structure simulation analysis. We downloaded the PDB file and used PYMOL and Chimera to analyze whether the mutation results have an important impact on the surrounding structure.

[0052] Through a large number of bioinformatics analyses and spatial modeling, 19 mutation sites were initially selected, including 3 N-glycosylation sites, 3 O-glycosylation sites, 4 sites related to glycosylation conserved motifs, 2 cysteines related to disulfide bond formation, as well as other phosphorylation sites and polar amino acids. The specific positions are as Figure 2 shown. Table 1 shows the different mutation positions and substituted amino acids.

[0053] Table 1 KLK1 Mutation Positions and Substituted Amino Acids

[0054]

[0055] Example 2. Obtaining of KLK1 Wild Gene and Construction of Expression Vector

[0056] The gene was synthesized to encode the full-length sequence of human kallikreinogenase NM_002257.3. An EcoR I restriction site and a tandem repeat purification tag of 6 histidines were introduced upstream by PCR, and an EcoRV restriction site was introduced downstream. The primer sequences are as follows:

[0057] F1: F 5'-ACGAATTCGCACCATCATCATCATCATGCGCCCCCGATTCAG-3'

[0058] R1: F 5'-GAGATATCATTATTAGGAGTTCTCCGC-3'

[0059] After the PCR amplified fragment was recovered using an agarose gel electrophoresis recovery kit (Tiangen, DP209-03), it was cloned into the pCDNA3.4 vector (purchased from Thermo Fisher) that had also been double-digested with EcoR I and EcoRV to construct an expression vector.

[0060] Example 3. Construction of KLK1 Mutant Expression Vector

[0061] The KLK1 gene containing mutation sites described in Example 1 of full-length gene synthesis, wherein the single-site mutations N108Q and N165Q are amplified by Overlap PCR. Using the wild-type KLK1 gene as a template, three gene sequences are amplified and spliced into a full-length gene sequence by the overlap PCR method. The primer sequences for amplification are as follows:

[0062] N108Q-F: 5'-TGGAGCAGCACACCCG-3'

[0063] N108Q-R: 5'-GGTGTGCTGCTCCAGG-3'

[0064] N165Q-F: 5'-GAACCAGAGCAGTTCTCATTT-3'

[0065] N165Q-R: 5'-TGAGAACTGCTCTGGTTCG-3'

[0066] F1: 5'-ACGAATTCGCACCATCATCATCATCATGCGCCCCCGATTCAG-3'

[0067] R1: 5'-GAGATATCATTATTAGGAGTTCTCCGC-3'

[0068] After double digestion of the gene synthesis plasmid and the PCR amplified fragment, they were subcloned into the pCDNA3.4 vector that was also digested with EcoR I and EcoRV to construct a mutant expression vector.

[0069] Example 4. Expression and purification of KLK1 protein in mammalian cells

[0070] According to the manufacturer's instructions, Gibco transfection reagent was used to transiently express the KLK1 wild-type and mutant expression vectors constructed in Examples 2 and 3 in Expi293F cells and Expi-CHO-S cells. Briefly, Expi293F cells were transfected with the obtained vector using the ExpiFectamine293 Transfection Kit (Gibco), with a total transfection DNA of 25 μg / 25 mL, and Expi-CHO-S cells were transfected with the ExpiFectamine CHO Transfection Kit, with a total transfection DNA of 12.5 μg / 25 mL. At 37 °C, 5% CO 2After culturing in an incubator at 125 RPM for 4 - 5 days, the cell culture supernatant was collected, centrifuged at 3500 rpm for 5 minutes, and filtered through a 0.22 μm filter membrane to remove cell debris. The cell culture supernatant was incubated with nickel - ion affinity chromatography (GE) packing material overnight at 4°C. After eluting the miscellaneous proteins with 15 mM and 30 mM imidazole, it was eluted with 10 mM Tris - HCl buffer containing 500 mM NaCl, 300 mM imidazole, and pH 8.0. After ultrafiltration and concentration, the protein was stored in PBS buffer (pH 7.0), and its concentration was measured by NanoDrop. The results of partially purified protein are as shown in Figure 3a / 3b. Surprisingly, it was found that the amino acids at positions N102, C31, and C66 play important roles in the structural stability, in vitro translation and folding, and secretory expression of KLK1 protein. In particular, the mutation at the N102 site significantly reduced the protein expression level, down to 10% or even lower of the wild - type expression level. Therefore, the original amino acid sequences of these three sites were retained in subsequent studies.

[0071] Example 5. Activation of pro - KLK1

[0072] The 16 samples of KLK1 and its mutant proteins purified in Example 4 were replaced into 10 mM Tris - HCl buffer containing 150 mM NaCl and pH 8.0 through an ultrafiltration tube with a molecular weight cut - off of 10 kDa. The protein was placed at 37°C and incubated with autoactivation or thermolysin protease

[0073] (Sigma, T7902 - 100MG)) or trypsin to remove the activation peptide in prekallikrein, and the active kallikrein was released. The activated KLK1 was subjected to nickel - ion affinity chromatography again to remove free pro - peptides and unactivated residual pro - enzymes. The electrophoresis results of kallikrein after activation with thermolysin and trypsin are as shown in Figure 4a / 4b.

[0074] Example 6. Enzyme activity assay

[0075] The 16 samples of kallikrein activated in Example 5 and the standard product (Yikai TMPancreatic kininogenase for injection (Changzhou Qianhong) was diluted to 10 μg / mL and 10 U / mL with 10 mM Tris-HCl, 150 mM NaCl, pH 8.0 respectively. 25 mg of substrate S2266 (D-Val-Leu-Arg-pNA) was dissolved with 10 ml of Tris-HCl, 150 mM NaCl, pH 8.0. 50 μL of the test sample and standard product preheated at 37°C ± 0.5°C were taken respectively and added into a 96-well microplate, then 50 μL of the substrate solution was added respectively. The absorbance A0 at 0 minute and the absorbance A at 10 minutes were read rapidly at a wavelength of 405 nm. The standard product solution and the test sample solution were determined in parallel for 3 times, and the average values of the ΔA values (ΔA = A - A0) of the standard product solution and the test sample solution were obtained respectively and substituted into the enzyme activity formula for calculation. The results of enzyme activity determination are shown in Table 2. Among them, the enzyme activities of R77H and H109K were significantly lower than that of the wild type, so no further analysis was performed on the mutations at this site.

[0076]

[0077] In the above formula, P is the enzyme amount of kallikrein per 1 mg of the test sample, unit; △At is the average value of ΔA of the test sample solution; △A S is the average value of ΔA of the standard product solution; P S is the unit number of the standard product solution; n is the dilution factor, w is the mass of the test sample, unit is mg.

[0078] Table 2 Enzyme activities and expression levels of single-site mutations of KLK1

[0079]

[0080] Example 7. Characterization analysis of KLK1 protein

[0081] To verify the effects of each mutation site on protein stability and integrity, LC / MS method was used to perform integrity, disulfide bond and glycosylation characterization analysis on the 14 kallikrein enzymes obtained in Example 6 (U3000 HPLC chromatographic system and -QExactive mass spectrometer, Thermo). Briefly, for integrity analysis, each protein sample was subjected to reduction alkylation using the FASP digestion method. After removing the sugar chains with PNGaseF (NEB, p0704s), the samples were digested with Trypsin (Sigma, T0303-10G) and Chymotrypsin (Pierce, 90056) for 16 hours respectively. The peptide segments were dried by vacuum, re-dissolved with 0.1% FA and then subjected to mass spectrometry analysis. The disulfide bond analysis samples were divided into two parts. One part was subjected to reduction alkylation, and after removing the sugar chains, it was digested with Chymotrypsin. The other part was directly digested without reduction alkylation, and then detected by the instrument. By comprehensively analyzing the results, the presence of disulfide bonds was determined. For the glycopeptide samples, after each protein sample was subjected to reduction alkylation using the FASP digestion method, the sugar chains were not removed, and it was directly digested with Glu-C (Sigma, 2922-100UN). After the samples were enriched by SAX, they were detected by the instrument. The identification results of the integrity analysis part are shown in Tables 3 - 6. From the results, it can be seen that the activation of thermocin and trypsin has no significant effect on the N and C integrity of the single-site mutant proteins at 14 sites, while self-activation has a significant effect on the N-terminal integrity of the protein. As shown in Table 7, the proportion of the N-terminal incomplete sequences reached 34%. The N-terminal sequence integrity is very important for the activity of this enzyme, and partial deletion may cause a decrease in enzyme activity and protein instability. In addition to the N-terminal incompleteness, another potential problem of self-activation lies in incomplete activation. It was detected that the total proportion of the residual amino acids R, SR or QSR at the end of the propeptide at the N-terminal after self-activation reached more than 50%. The incomplete excision of the propeptide also has an obvious impact on the formation of mature peptide release enzyme activity. Therefore, self-activation may be an activation method of incomplete or inaccurate cleavage, which is affected by multiple factors such as digestion conditions and reaction components. The formation of amino acids and disulfide bonds plays an important role in protein conformation and tertiary structure stability. After the amino acid mutations at 14 sites, compared with the wild-type protein, there is no difference in the position of disulfide bonds, and the results are not shown. Glycosylation has important effects on the properties of proteins such as stability, immunogenicity, molecular weight homogeneity and isoelectric point. The results of chromatographic and mass spectrometric identification of the glycosylation forms and ratios of different mutant proteins are shown in Table 8. It is worth noting that since the N102 site is a key amino acid, mutation of it can affect protein folding and secretion. The reported sugar types at this site are the most abundant, including more than 10 kinds of glycosylation modifications. The glycosylation site at the N108 site is close to the N102 site, and there is no suitable protease cleavage site for independent analysis.As can be seen from the glycosylation modification identification results in Table 8, the mutations at the N108 and N165 glycosylation sites do not affect the protein production and basic properties. The expression of mutants in 293 cells and CHO-S cells is different. A2G0FB, A2G1FB, A2G0B, A2G2FB, and A2G2B are the main forms of sugar modification. The main glycoforms of the N108 / N165 double-site mutant in the two types of cells are not exactly the same, but the proportions of the top three glycoforms are 72.5% (293 cells) and 74.2% (CHO cells) respectively. Especially in CHO-S cells, the proportion of the main glycoform is significantly higher than that of the wild type (57.8%), which is consistent with the more concentrated molecular weights during electrophoresis.

[0082] Table 3 List of N-terminal peptide segments of KLK1-N165Q protein after activation by Thermolysin

[0083]

[0084] Table 4 List of C-terminal peptide segments of KLK1-N165Q protein after activation by Thermolysin

[0085]

[0086]

[0087] Table 5 List of N-terminal peptide segments of KLK1-N108DN165Q protein after activation by Trypsin

[0088]

[0089] Table 6 List of C-terminal peptide segments of KLK1-N108DN165Q protein after activation by Trypsin

[0090]

[0091] Table 7 Incomplete form peptide segments at the cleavage end of KLK1-N165D after self-activation by Chymotrypsin

[0092]

[0093] Table 8. Effects of mutations at different glycosylation sites on sugar chain composition

[0094]

[0095]

[0096] Example 8. Double-site and triple-site mutations

[0097] 19 amino acid mutation sites were designed in Example 1. Among them, N102, N108, and N165 were proven to play key roles in the stability and activity of the protein. The mutations of these three sites were studied in more depth, and double-site and triple-site mutations were designed. The methods described in Examples 2-6 were used for mutant protein expression and enzyme activity determination. The specific mutation positions and substituted amino acids are shown in Table 9. The results showed that the protein expression level or enzyme activity decreased significantly after single-site mutations of N102Q / N102D and N108D, while the single-site mutation of N165 had no significant effect on the protein expression level and enzyme activity. It was speculated that this result was related to the glycosylation level of the site. Further analysis of double-site and triple-site mutations surprisingly found that the expression level and enzyme activity of double mutations and triple mutations containing the N102 site decreased, which was consistent with the theoretical speculation, while the expression level and enzyme activity of the N108 / N165 double-site mutation did not decrease. Therefore, further research and analysis were carried out on the N108 / N165 double mutation and the N165 single mutation. The mutant protein products of N165D, N108D&N165D, N108D&N165Q, and N108Q&N165D were named RD200, RD201, RD202, and RD203 respectively, and the wild-type protein was named WT.

[0098] Table 9 Enzyme activity and expression level of KLK1 mutants

[0099]

[0100] Example 9. Hydrolysis of natural substrate by KLK1 mutants

[0101] The KLK1 protein is a tissue kallikrein, and its natural substrate in the human body is low molecular weight kininogen. At Met 379 -Lys 380 and Arg 389 -Ser 390Site-specific cleavage generates the bioactive polypeptide kallidin (Lys-BK: KRPPGFSPFR). To evaluate the in vitro biological activity of the mutant proteins, Expi-293F was selected as the host cell to recombinantly express human low molecular weight kininogen protein LMW (Uniprot Entry: P01042), the sequence of which is shown in SEQ NO ID2. The mutant proteins obtained in Example 8 were used to hydrolyze LMW in vitro. Briefly, LMWK was diluted to 1 mg / mL with 10 mM Tris-HCl, 0.15 M NaCl, pH 8.0. The final concentration of mature KLK1 and its mutant proteins was 0.1 μg / mL, and the hydrolysis experiment was carried out at 37°C. Samples were taken at 10 min, 20 min, and 40 min of incubation, and the reaction was terminated by adding trichloroacetic acid at a final concentration of 50 mmol / L. The enzyme activities of different mutants were accurately determined by synthesizing a stable isotope-labeled standard KRPPGFSPFR peptide and using multiple reaction monitoring (MRM) technology to measure the concentration of the generated polypeptide. This technology is essentially a scanning mode of mass spectrometry. Based on the specific parent ion and daughter ion pairs of the target protein, signals that conform to the target ion rules are selected for acquisition, and signals that do not conform to the rules are removed to interfere with the highly sensitive, accurate, and specific quantitative analysis of proteins or polypeptides. The concentration of the decapeptide Lys-BK generated by the hydrolysis of different enzymes on the substrate was measured on a 6500 mass spectrometer from SCIX, and then the activities of each mutant were calculated and compared. The specific steps of the MRM detection method are as follows:

[0102] Standard curve and quality control sample processing procedures

[0103] 1) Take 25 μL of the standard, 25 μL of the internal standard, and 1 μL of BSA (50 μg / μL, 150 mM NaCl) and add them to a rinsed 10k ultrafiltration tube. Centrifuge at 14000 g at 4°C for 20 min.

[0104] 2) Wash three times with 100 μL, 50 μL, and 50 μL of pure water respectively. Collect the permeate and washing solution, dry by suction and redissolve in 20 μL of 0.1% formic acid aqueous solution, and the sample loading volume is 8 μL.

[0105] Test sample processing procedures

[0106] 1) Take 20 μL of the sample, 20 μL of the internal standard, and 11 μL of Buffer (10 mM Tris-HCl, 150 mM NaCl, pH 8.0) and add them to a rinsed 10k ultrafiltration tube. Centrifuge at 14000 g at 4°C for 20 min.

[0107] 2) Wash three times with 100 μL, 50 μL, and 50 μL of pure water, collect the penetration and washing solutions, dry by suction and redissolve in 16 μL of 0.1% formic acid aqueous solution, with a sample loading volume of 8 μL. Sample analysis was performed on the SCIEX triple quadrupole mass spectrometry system Triple Quad TM 6500+. The results are shown in Table 10. In the table, taking the concentration released by each mutant at 5 minutes as the reference point, the concentration of Lys-BK released at different times was compared. The concentration at different time points minus the concentration at 5 minutes of reaction was expressed as Cx - C5 to calculate the polypeptide release rate of each sample to evaluate the enzyme activity. For the four mutants, RD200, RD201, RD202, and RD203, whether it is a single-site mutation or a double-site mutation, the Lys-BK polypeptide release rate is higher than that of the wild-type protein. The polypeptide release rate of the wild-type enzyme is 54.5 pmol / ml·min. The hydrolysis peptide release rate of the single-site mutant RD200 with N165D at 20 minutes is 83.3 pmol / ml·min. The hydrolysis peptide release rate of the double-site mutant RD203 with N108Q and N165D is 94 pmol / ml·min and reaches a plateau at about 20 min. Neither the double-site mutant RD202 with N108D and N165Q nor the single-site mutant RD200 reaches a plateau at 20 minutes. Continuing to compare their polypeptide release rates when reaching a plateau at 40 minutes, the rate of RD202 is 47.3 pmol / ml·min, and the rate of RD200 is 41.0 pmol / ml·min. However, the double-site mutant RD201 with N108D and N165D has a lower polypeptide hydrolysis and release rate than the single-site mutant at both 20 minutes and 40 minutes. Although the three double-site combination mutants all conform to the usual amino acid substitution principle, their enzyme activity performances are different, indicating that the change in enzyme activity not only depends on the glycosylation conservative site, but also the introduction of different amino acids has an important impact on enzyme activity.

[0108] Table 10 Comparison of the activities of different KLK1 enzymes in hydrolyzing the substrate to release Lys-BK peptide

[0109]

[0110] Example 10. PEG conjugation of KLK1 mutants

[0111] The half-life of KLK1 protein in human serum is approximately 3 - 4 hours. To extend the half-life of the target protein in plasma and improve the in vivo drug release efficiency, site-directed PEGylation modification of the N-terminus of KLK1 protein was performed. Briefly, KLK1 was diluted to a final concentration of 5 mg / ml with PBS pH5.5 solution, and mixed with M-PEG-ALD-20KD (Beijing KeyGen, A3001) at 6 times the molar mass of KLK1 and sodium cyanoborohydride (Sigma) at 100 times the molar mass. The mixture was inverted and suspended at room temperature for reaction. As Figure 5 shown, with the extension of the reaction time, the PEG coupling efficiency increased significantly. After 1 hour of coupling, the coupling efficiency was approximately 30%, and after 5 hours, the coupling efficiency was higher than 60%. The molecular weight of the protein increased from 33 kDa to 54 kDa. The increase in molecular weight greatly reduced its glomerular filtration during systemic drug administration, thus reducing its excretion through urine. In addition, the stability of the PEG-modified drug was improved in the systemic circulation, and the retention time was extended, which was beneficial to improving the distribution of the drug in vivo, especially facilitating the accumulation of macromolecular drugs in tumors and inflammatory sites with enhanced retention effects, thereby prolonging the in vivo treatment time of the drug.

[0112] Example 11. Screening of stable expression cell lines of KLK1 mutant proteins

[0113] The CHO-S system from Gibco was selected as the host cell for the development of stable cell lines. The mutant obtained by screening in Example 9 was used to construct an expression vector with pCHO1.0 as the backbone. After electroporation, two rounds of antibiotic pressure screening were carried out, followed by monoclonal plating screening and two rounds of limited dilution. Monoclonal cells that could be stably passaged were screened. Recombinant expression identification obtained unexpected results. The expression level of the stable cell line of the double-site mutant RD202 at positions N108 and N165 was 1 g / L, higher than the yield of the wild type at 0.5 g / L, and was more suitable for large-scale production of drugs.

[0114] Example 12. MCAO animal model test of KLK1 mutant proteins

[0115] A rat model of transient middle cerebral artery occlusion (tMCAO) was used to evaluate the in vivo efficacy of KLK1 protein. The experiment was designed as a short-term study lasting for 3 days, with an adaptation period of 3 - 7 days before the experiment. Healthy Sprague-Dawley (SD) rats, male, weighing 200 - 250 g, were selected for the experiment. The rats were divided into a sham operation group, a model group, a positive group, and dosing groups (RD200, RD201, RD202, RD203), with 10 rats in each group. The model establishment was carried out in two batches, with 5 rats in each batch, and the time interval between the two batches was one day. The first dose was administered 1 h after successful ischemia modeling, and then once a day on the 2nd and 3rd days after modeling. The body weight was measured every day during the experiment, and the neurological function and behavior were scored every day. At the end point, the sera of 5 rats were collected and the concentration of VEGF was determined by ELISA using a kit. Brain slices of 5 rats were stained with TTC, 5 rats' brain slices were stained with HE, 5 rats' brain slices were stained with IF TUNEL, and 5 rats' brain slices were stained with Nissl. Finally, the efficacy results of the test substances were judged by comprehensively analyzing the data of each index.

[0116] Transcranial Doppler blood flow monitoring was performed before the operation and 2 h after the operation, and the Longa score was given to the modeled rats 2 h after the operation. Through blood flow monitoring and Longa score, the rats were selected and grouped. When the cerebral blood flow in the ischemic area of the rat brain decreased by 70% (rCBF≥70%) and the Longa score was 2 - 3 points, it was determined that the modeling was successful and the rats were included in the group. After the operation, the body weights of the rats in the modeling group were all reduced compared with those in the sham operation group (Sham), with an average reduction of 5% - 10%. The specific scoring results and body weight changes are shown in Table 11.

[0117] Table 11 Longa score and grouping of rats

[0118]

[0119] At the end point of the experiment, the rats were euthanized by inhaling carbon dioxide, and the whole brain was taken and coronally sectioned (2 mm / brain slice) in a brain trough. After staining with 1% TTC solution in the dark at 37°C for 10 - 15 minutes, obvious red and white differences appeared in the infarcted and non-infarcted areas of the brain slices. The infarct area percentage was quantitatively analyzed using Image J software for the digitized scanned images. Dead rats were not stained with TTC. Infarct area percentage (%) = [contralateral hemisphere area - (ipsilateral hemisphere area - infarct area)] / contralateral hemisphere area × 100%. The results of the infarct area percentage in each group showed that the infarct area percentage in the model group was significantly higher than that in the sham operation group, with a significant difference (p < 0.001). The positive control group Huk and the dosing group RD202 were significantly lower than the model group, with statistical significance (p < 0.05). The specific results are shown in Figure 6 and Figure 7 .

[0120] At the end of the experiment, the sera of rats in each group were collected, and the expression level of VEGF protein was detected by ELISA. The results showed that compared with the model group, the expression levels of VEGF protein in the positive control group and the drug administration group were both increased, about 2-fold. There was no significant difference among different groups. The specific results are shown in Table 12.

[0121] Table 12 VEGF content in rat sera

[0122]

Claims

1. A low-glycosylated kininogenase or a derivative thereof, wherein the kininogenase is a primate kininogenase, and contains three N-glycosylation modification sites at the natural kininogenase NMS, NHT and NFS sequences, wherein NHT and NFS are not glycosylated or only slightly glycosylated, and the slightly glycosylated modification refers to a glycosylation modification ratio of the site ≤10%, ≤9%, ≤8%, ≤7%, ≤6%, ≤5%, ≤4%, ≤3%, ≤2%, ≤1%, ≤0.5% or ≤0.1%.

2. A recombinant kininogenase mutant or a derivative thereof, wherein the kininogenase is a primate kininogenase, and the recombinant kininogenase mutant or a derivative thereof has and has only one N-glycosylation modification site.

3. The recombinant kininogenase mutant or its derivative according to claim 2 retains the N-glycosylation modification site at the natural kininogenase amino acid sequence NMS, and does not contain the N-glycosylation modification site at the natural kininogenase amino acid sequence NHT.

4. The recombinant kininogenase mutant or its derivative according to claim 2 retains the N-glycosylation modification site at the natural kininogenase amino acid sequence NMS, and does not contain the N-glycosylation modification site at the natural kininogenase amino acid sequence NFS.

5. The recombinant kininogenase mutant or its derivative according to claim 2, wherein the asparagine at position NHT or position 108 of the natural human kininogenase amino acid sequence is mutated to any other amino acid. 6 . The recombinant kininogenase mutant or its derivative according to claim 2 , wherein the histidine at position NHT or position 109 of the natural human kininogenase amino acid sequence is mutated to proline.

7. The recombinant kininogenase mutant or its derivative according to claim 2, wherein the threonine at position NHT or position 110 of the natural human kininogenase amino acid sequence is mutated to any amino acid other than threonine.

8. The recombinant kininogenase mutant or its derivative according to claim 2, wherein the asparagine at position NFS or position 165 of the amino acid sequence of natural human kininogenase is mutated to any other amino acid.

9. The recombinant kininogenase mutant or its derivative according to claim 2, wherein the phenylalanine at position NFS or position 166 of the natural human kininogenase amino acid sequence is mutated to proline.

10. The recombinant kininogenase mutant or its derivative according to claim 2, wherein the serine at position NFS or position 167 of the natural human kininogenase amino acid sequence is mutated to any other non-serine amino acid.

11. The recombinant kininogenase mutant or derivative thereof according to any one of claims 3 to 10, wherein one, two or three amino acids are mutated among asparagine, histidine and threonine at the amino acid sequence NHT of the natural human kininogenase.

12. The recombinant kininogenase mutant or derivative thereof according to any one of claims 3 to 11, wherein one, two or three amino acids are mutated in asparagine, phenylalanine or serine at the amino acid sequence NFS of the natural human kininogenase.

13. The recombinant kininogenase mutant or its derivative according to claim 5, wherein the mutant is a mutant in which the amino acid at NHT or the 108th amino acid in the amino acid sequence of natural human kininogenase is mutated to a neutral polar amino acid, an acidic amino acid, a basic amino acid, or an aliphatic amino acid.

14. The recombinant kininogenase mutant or its derivative according to claim 13, wherein the mutant is a mutant in which the amino acid NHT or the 108th amino acid in the amino acid sequence of natural human kininogenase is mutated to glutamine (Gln) or aspartic acid (Asp).

15. The recombinant kininogenase mutant or its derivative according to claim 8, wherein the mutant is a mutant in which the amino acid NFS or the 165th amino acid in the amino acid sequence of natural human kininogenase is mutated to a neutral polar amino acid, an acidic amino acid, a basic amino acid, or an aliphatic amino acid.

16. The recombinant kininogenase mutant or its derivative according to claim 15, wherein the mutant is a mutant in which the amino acid NFS or the 165th amino acid in the amino acid sequence of natural human kininogenase is mutated to glutamine (Gln) or aspartic acid (Asp).

17. The recombinant kininogenase mutant or its derivative according to claim 14, wherein the amino acid sequence of the mutant is shown as SEQ ID No: 3, SEQ ID No: 4, SEQ ID No: 5 or SEQ ID No:

6.

18. The recombinant kininogenase mutant or its derivative according to claim 16, wherein the amino acid sequence of the mutant is shown as SEQ ID No: 3, SEQ ID No: 4, SEQ ID No: 5 or SEQ ID No:

6.

19. The recombinant kininogenase mutant or derivative thereof according to any one of claims 2 to 16, wherein the kininogenase is human kininogenase, the amino acid sequence of natural human kininogenase is such as GenBank accession number AAA36136.1 or NP_002248.1, and the accession numbers of naturally occurring variants are AAU12569, AAP35917 and AAH05313, respectively.

20. A composition comprising the recombinant kininogenase mutant or a derivative thereof according to any one of claims 2 to 16.

21. The recombinant kininogenase mutant or its derivative according to claims 2 to 16, characterized in that The specific enzyme activity of the activated zymogen is higher than that of the wild-type enzyme.

22. The recombinant kininogenase mutant or its derivative according to claims 2 to 16, characterized in that The main product is a single molecular weight protein.

23. The recombinant kininogenase mutant or its derivative according to claims 2 to 16 can be coupled with polyethylene glycol to form a polyethylene glycol. Alcohol-modified mutants.

24. The polyethylene glycol-modified mutation of claim 23, characterized in that Modification was achieved by N-terminal site-directed conjugation.

25. A recombinant expression vector that can be used to recombinantly express the sequence of claims 17-18.

26. The expression vector of claim 25 can be selected from pCDNA3.4, pCHO1.0, pEE6.4, and pEE12.

4.

27. An engineered cell comprising a gene encoding the protein mentioned in claims 17-18.

28. The engineered cells according to claim 27 are selected from Chinese hamster ovary cells CHO-S, CHO-K1, and mouse myeloma cells SP2 / 0, NS0.

29. A method for activating recombinant human kininogenase zymogen.

30. The activation method according to claim 29 is to cleave the leader peptide using trypsin, thermolysin, etc., or to activate the protein itself under specific conditions.

31. Use of the kininogenase according to claims 3 to 10 in the preparation of drugs for treating diseases such as stroke, hypertension, and diabetic nephropathy.

Images