WO2014185716A1 - Method for removing impurities by using calcium phosphate precipitation - Google Patents

Abstract

The present invention relates to a method for removing impurities in a sample containing proteins, comprising a step of causing calcium phosphate to precipitate by adding a calcium ion solution and a phosphate ion solution to the sample containing proteins. It is possible to effectively remove impurities such as protein aggregates within a short time and by a simple method if purifying proteins by using the present invention instead of chromatography, which takes a long time and has a complex equipment and method in a purification process after protein A chromatography.

Description

Impurity Removal Method Using Calcium Phosphate Precipitation

The present invention relates to a method for removing impurities using calcium phosphate precipitation, and more particularly, to a method for removing impurities by adding a calcium ion solution and a phosphate ion solution to a protein-containing sample to cause a calcium phosphate precipitation reaction.

Resin using calcium phosphate precipitation (e.g., Ceramic Hydroxyapatite resin, Hydroxyapatite resin: manufactured by BioRad, USA) chromatography binds or drops proteins to the resin in a specific buffered environment. How to separate.

When using ceramic hydroxyapatite or hydroxyapatite chromatography, high molecular weight protein aggregates, DNA, viruses, host cell proteins, protein A (Protein) A) can be removed.

However, resin chromatography using calcium phosphate precipitation is costly and time consuming for the following reasons. In order to proceed with protein purification using columns in production, it is necessary to first purchase and qualify columns and chromatography equipment. In addition, in order to perform chromatography, many kinds and a large amount of buffers are required. In particular, it takes a long time to prepare and perform chromatography (column packing, cleaning, buffer preparation, documentation).

The present inventors added calcium phosphate precipitates (Calcium phosphate precipitate, Ca ₅ ) by adding calcium ion (Ca ²⁺ ) solution and phosphate ion (PO ₄ ^3- ) solution to protein-containing samples without using column chromatography using resin. (PO ₄ ) ₃ (OH), Hydroxyapatite) was induced to devise a method for removing impurities. When calcium phosphate precipitate is formed, the protein is adsorbed with the calcium phosphate precipitate to precipitate together. The type of protein that binds to calcium phosphate precipitates depends on the concentration of calcium phosphate when added. Thus, the desired protein can be separated according to the concentration of calcium and phosphate. The method does not require as much equipment, buffers and long time as the protein purification using the chromatography column mentioned above.

As a result, it was confirmed that the use of calcium phosphate precipitation in protein purification could replace the uncomfortable, time-consuming and costly column chromatography process, and thus completed the present invention.

Accordingly, an object of the present invention is to provide a method for removing impurities in a protein-containing sample, comprising the step of forming a calcium phosphate precipitation reaction by adding a calcium ion solution and a phosphate ion solution to a protein-containing sample.

Another problem to be solved by the present invention is to provide a method for producing a protein formulation comprising a purification step using the above method.

In order to solve the above problems, one embodiment of the present invention provides a method for removing impurities in a protein-containing sample comprising the following steps:

a) purifying the protein-containing sample by applying it to Protein A affinity chromatography;

b) adding a calcium ion solution and a phosphate ion solution to the purified protein to form a calcium phosphate precipitation reaction, and then removing the precipitate by centrifugation or filtration; And

c) ion exchange chromatography, hydrophobic interaction chromatography and mixed mode chromatography of the purified protein after step b) or between steps a) and b). Purifying by applying to any one of the chromatography selected from the group consisting of mixed mode chromatography.

In addition, another embodiment of the present invention provides a method for producing a protein formulation comprising a purification step using the above method.

Hereinafter, the present invention will be described in more detail.

In general, the purification process of the antibody is based on Protein A chromatography, which can obtain high purity and yield with only one step purification. (Fahrner RL, Knudsen HL, Basey CD, Galan W, Feuerhelm D, Vanderlaan M, et al. Industrial purification of pharmaceutical antibodies: development, operation and validation of chromatography processes. Biotechnol Genet Eng Rev 2001; 18: 301-27) . To obtain the desired purity and protein of interest, after the Protein A chromatography step, ion-exchange creams are used to remove high molecular weight protein multimers, DNA, viruses, host cell-derived proteins, leached Protein A and impurities. Additional purification steps such as ion-exchange chromatography, hydrophobic interaction chromatography, and mixed mode chromatography can be utilized (Liu HF, Ma J, Winter C, Bayer R. Recovery and purification process development for monoclonal antibody production.mAbs. 2010; 2: 480-499).

According to one embodiment of the invention, after the Protein A chromatography step, calcium phosphate precipitation may be used in place of equipment and time-consuming additional chromatography. In other words, by applying the calcium phosphate precipitation method instead of the above additional chromatography, large molecular weight protein multimers, DNA, viruses, host cell-derived protein, Protein A, etc. can be effectively removed in a short time with little labor.

One embodiment of the present invention, as shown in Figure 1, after the Protein A chromatography step in the general protein purification step, by applying a calcium phosphate precipitation method instead of an additional 1 ^st or 2 ^nd chromatography step, an additional chromatography step It is a way to get the purification effect.

When explaining the calcium phosphate precipitation method of the present invention as an example.

In order to produce antibodies using animal cells, and to purify the antibodies contained in the cell culture medium, the cells and cell residues are removed by centrifugation and filtration, and the filtered cell culture medium is removed. After loading and binding to the Protein A column in equilibrium, the solution is again washed with the equilibrium solution, and the antibody is eluted from the Protein A column with a low pH elution solution. The eluted antibody solution is titrated to pH 3.6 and stored at room temperature for 1 hour for low pH virus inactivation. Thereafter, the antibody solution is neutralized to pH 6.0. The neutralized antibody solution was further purified by anion exchange (Q sepharose Fast Flow; manufactured by GE Healthcare), and the intermediate antibody solution obtained from anion exchange chromatography was adjusted to pH 7.0, followed by 100 mM CaCl ₂ and 100 mM. Sodium phosphate pH 7.0 is titrated in the intermediate purified solution to a final concentration of 10 mM Ca ²⁺ and 10 mM phosphate to cause precipitation. After titration, it is maintained at room temperature for 3 hours to sufficiently combine calcium phosphate precipitate (Hydroxyapatite, Ca ₅ (PO ₄ ) ₃ (OH)) with antibody multimers and other impurities (DNA, host cell-derived protein, Protein A, etc.). Let's do it. Thereafter, centrifugation or filtration is performed to remove the precipitates from the antibody solution.

In addition, by neutralizing protein A chromatography and low pH virus inactivation in the same manner as above, prior to anion exchange chromatography, calcium phosphate precipitation may be applied to the intermediate purified antibody solution. The filtered antibody solution can then be subjected to anion exchange chromatography.

In other words, the impurity removal method using calcium phosphate precipitation according to an embodiment of the present invention is a step after the protein A chromatography, the sequence and chromatography techniques (eg, ion exchange chromatography, hydrophobic interaction chromatography, mixed mode chromatography) It is possible to apply instead of intermediate chromatography purification of proteins (including antibodies).

Therefore, in one embodiment according to the present invention, the present invention provides a method for removing impurities in a protein-containing sample, comprising adding a calcium ion solution and a phosphate ion solution to a protein-containing sample to form a calcium phosphate precipitation reaction. do. And, after the calcium phosphate precipitation reaction step may further comprise the step of removing the precipitate by centrifugation or filtration.

In addition, in one embodiment of the present invention, prior to the calcium phosphate precipitation reaction step may further comprise the step of applying a protein containing sample to protein A adsorption chromatography to purify.

Further, in one embodiment of the present invention, further purification of the protein-containing sample purified in the previous step after the calcium phosphate precipitation reaction step, or after the Protein A adsorption chromatography purification step and before the calcium phosphate precipitation reaction step It may include a step. In the further purification step, purification may be applied to any one of chromatography selected from the group consisting of ion exchange chromatography, hydrophobic interaction chromatography, and mixed mode chromatography, but is not limited thereto.

For example, the ion exchange chromatography may be cation ion exchange chromatography or anion ion exchange chromatography.

For example, the hydrophobic interaction chromatography may be Ether (eg, Toyopearl Ether-650M, Toyopearl Ether-650S, etc.), Butyl (eg, Butyl sepharose, Butyl S sepharose, Catpo Butyl, etc.), Hexyl (eg, Toyopearl Hexyl-650C). ), Octyl (e.g. Octyl sepharose, etc.), Pheyl (e.g. Toyopearl Phenyl-650M, Phenyl sepharose, Catpo Phenyl, Fractogel ^® EMD Phenyl (S), etc.), Propyl (e.g., Fractogel ^® EMD Propyl (S), etc.) Ligand, such as, but may be included, but is not limited thereto.

For example, the mixed mode chromatography may be a mixed mode chromatography (eg, Capto adhere, Capto MMC, MEP HyperCell, Eshmuno HCX, etc.), an ion exchange function, and a cation exchange function. Glyphography (eg, hydroxyapatite, Ceramic hydroxyapate, etc.) may be included, but is not limited thereto.

In addition, one embodiment of the present invention provides a method for removing impurities in a protein containing sample comprising the following steps:

a) purifying the protein-containing sample by applying it to Protein A affinity chromatography;

b) adding a calcium ion solution and a phosphate ion solution to the purified protein to form a calcium phosphate precipitation reaction, and then removing the precipitate by centrifugation or filtration; And

c) ion exchange chromatography, hydrophobic interaction chromatography and mixed mode chromatography of the purified protein after step b) or between steps a) and b). Purifying by applying to any one of the chromatography selected from the group consisting of mixed mode chromatography.

In one embodiment of the present invention, when the protein-containing sample is a sample containing phosphate ions can be added only calcium ion solution, on the contrary, in the case of a sample containing calcium ion it can be added only a phosphate solution.

In one embodiment of the present invention, the calcium ion solution may be CaCl ₂ , or CaCO ₃ , but is not limited thereto.

In one embodiment of the present invention, the phosphate ion solution may be Na ₃ PO ₄ , NaH ₂ PO ₄ , Na ₂ HPO ₄ or KH ₂ PO _4, K ₂ HPO _4, K ₃ PO ₄ , but is not limited thereto. It is not.

In one embodiment of the present invention, the concentration of calcium ions added to the calcium phosphate precipitation reaction is 1 to 100 mM, 1 to 90 mM, 1 to 80 mM, 1 to 70 mM, 1 to 60 mM, 1 to 50 mM , 1 to 40 mM, 1 to 30 mM, 1 to 20 mM, 1 to 10 mM, or 1 to 5 mM, but is not limited thereto. For example, the concentration of calcium ions added to the calcium phosphate precipitation reaction may be 1 to 20 mM.

In one embodiment of the present invention, the concentration of phosphate ions added to the calcium phosphate precipitation reaction is 1 to 100 mM, 1 to 90 mM, 1 to 80 mM, 1 to 70 mM, 1 to 60 mM, 1 to 50 mM , 1 to 40 mM, 1 to 30 mM, 1 to 20 mM, 1 to 10 mM, or 1 to 5 mM, but is not limited thereto. For example, the concentration of phosphate ions added to the calcium phosphate precipitation reaction may be 1 to 20 mM.

According to one embodiment of the present invention, the concentration of calcium ions added to the calcium phosphate precipitation reaction may be 1 to 100 mM, and the concentration of phosphate ions may be 1 to 100 mM.

According to another embodiment of the present invention, the concentration of calcium ions added to the calcium phosphate precipitation reaction may be 1 to 20 mM, and the concentration of phosphate ions may be 1 to 20 mM.

In one embodiment of the present invention, the calcium phosphate precipitation reaction is pH 5.0 to 10.0, pH 5.0 to 9.5, pH 5.0 to 9.0, pH 5.0 to 8.5, pH 5.0 to 8.0, pH 5.0 to 7.5, pH 5.0 to 7.0, pH It may be performed at 5.0 to 6.5, pH 5.0 to 6.0, or pH 5.0 to 5.5, but is not limited thereto.

According to one embodiment of the invention, the calcium phosphate precipitation reaction may be carried out at pH 5.5 to 7.5.

According to another embodiment of the present invention, the concentration of calcium ions added to the calcium phosphate precipitation reaction is 1 to 20 mM, the concentration of phosphate ions is 1 to 20 mM, and the calcium phosphate precipitation reaction is pH 5.5 to 7.5. This can be done at

In one embodiment of the present invention, the calcium phosphate precipitation reaction time is 0 to 10 hours, 0 to 9 hours, 0 to 8 hours, 0 to 7 hours, 0 to 6 hours, 0 to 5 hours, 0 to 4 hours, 0 to 3 hours, 0 to 2 hours, 0 to 1 hour, or 0 to 0.5 hours, but is not limited thereto.

According to one embodiment of the present invention, the calcium phosphate precipitation reaction time may be 3 to 5 hours.

According to another embodiment of the present invention, the concentration of calcium ions added to the calcium phosphate precipitation reaction is 1 to 20 mM, the concentration of phosphate ions is 1 to 20 mM, and the calcium phosphate precipitation reaction time is 3 to 5 It can be time.

In one embodiment of the present invention, the concentration of the protein-containing sample during the calcium phosphate precipitation reaction 1 to 10 times, 1 to 9 times, 1 to 8 times, 1 to 7 times, 1 to 6 times, 1 to 5 times, 1 to 4 times, 1 to 3 times, 1 to 2 times, or 1 to 1.5 times dilution, but is not limited thereto.

According to one embodiment of the invention, the calcium phosphate precipitation reaction may be diluted 1 to 3 times the concentration of the protein-containing sample.

According to another embodiment of the present invention, the concentration of calcium ions added to the calcium phosphate precipitation reaction is 1 to 20 mM, the concentration of phosphate ions is 1 to 20 mM, and the calcium phosphate precipitation reaction time is 3 to 5 In time, the calcium phosphate precipitation reaction may be diluted 1 to 3 times the concentration of the protein-containing sample.

According to another embodiment of the present invention, the concentration of calcium ions added to the calcium phosphate precipitation reaction is 1 to 20 mM, the concentration of phosphate ions is 1 to 20 mM, and the calcium phosphate precipitation reaction is pH 5.5 to 7.5. Carried out in, and the calcium phosphate precipitation reaction time may be 3 to 5 hours.

According to another embodiment of the present invention, the concentration of calcium ions added to the calcium phosphate precipitation reaction is 1 to 20 mM, the concentration of phosphate ions is 1 to 20 mM, and the calcium phosphate precipitation reaction is pH 5.5 to 7.5. The calcium phosphate precipitation reaction time is carried out at 3 to 5 hours, and the calcium phosphate precipitation reaction may be diluted 1 to 3 times the concentration of the protein-containing sample.

In one embodiment of the invention, the impurities include, but are not limited to, protein multimers, protein fragments, DNA contaminants, viruses, host cell derived proteins or protein A.

In one embodiment of the present invention, the protein contained in the protein-containing sample to remove impurities may be a physiologically active protein, for example, the protein may be an antibody or a fusion protein, but is not limited thereto. For example, the antibody may be a monoclonal antibody, for example, the antibody may be an anti-CD20 antibody or an anti-ErbB3 antibody. In addition, the fusion protein may be an Fc fusion protein. For example, the Fc fusion protein may be a fusion protein of a TNFα receptor fragment and an antibody Fc fragment.

More specifically, the protein contained in the protein-containing sample to be purified by the method according to one embodiment of the present invention, for example, monoclonal antibody, Fc fusion protein, granulocyte colony stimulating factor (G-CSF), granulocyte macrophage colony Hematopoietic factors such as stimulating factor (GM-CSF), erythropoietin (EPO), thrombopoietin, cytokines such as interferon, IL-1, IL-6, serum albumin, blood coagulant VIII factor, blood coagulant IX Factors include, but are not limited to, insulin, stem cell growth factor (SCF), and the like. In one embodiment of the invention carried out using Protein A adsorption chromatography, the protein to remove impurities may be a monoclonal antibody or an Fc fusion protein (eg, a TNFRc-Fc fusion protein). In one embodiment of the invention, the purifiable antibody comprises all IgG, IgA, IgE, IgD, or IgM.

As used herein, the term "physiologically active protein" refers to a substance having a biological activity substantially the same as that of a physiologically active protein of a mammal, for example, a human being, or one obtained by natural or genetic recombination. Bioactive proteins by genetic recombination include bacterial cells such as E. coli; Chinese hamster ovary (CHO) cells, BHK cells, COS cells, cells derived from animals, such as human-derived cells can be prepared, and can be isolated and purified by various methods before use. Proteins obtained by genetic recombination include those having the same amino acid sequence as the native protein and those that retain the biological activity as deletions, substitutions, and additions of one or more amino acids in the amino acid sequence. In addition, physiologically active proteins include those chemically modified by PEG and the like.

When the bioactive protein is a monoclonal antibody, the monoclonal antibody may be produced by any method. The monoclonal antibody basically immunizes the sensitizing antigen according to a conventional immunization method using a known technique, fuses the obtained immune cells with known parental cells by a conventional cell fusion method, and then performs a conventional screening method. It can produce by screening antibody producing cells which are monoclonal.

Alternatively, the antibody genes can be cloned from the hybridomas and inserted into a suitable vector and introduced into the host using gene recombination techniques, and the resulting recombinant antibodies can also be used in one embodiment of the present invention. Specifically, cDNA of the variable region (V region) of an antibody is synthesize | combined from the mRNA of a hybridoma using reverse transcriptase. When DNA encoding the V region of the antibody of interest is obtained, the DNA is linked to the DNA encoding the desired antibody constant region (C region) and inserted into the expression vector. Alternatively, the DNA encoding the V region of the antibody may be inserted into an expression vector containing the DNA of the antibody C region. The DNA construct is inserted into the expression vector so that it is expressed under the control of an expression control region, eg, an enhancer or a promoter. Next, the host cell can be transformed with this expression vector to express the antibody.

In one embodiment of the present invention, artificially modified genetic recombinant antibodies, such as chimeric antibodies, humanized antibodies, etc., may be used for the purpose of lowering heterologous antigenicity to humans. These modified antibodies can be produced using a known method.

A chimeric antibody is an antibody consisting of a variable region of a heavy chain and a light chain of a mammal other than human, for example, a mouse antibody, and a constant region of a heavy chain and a light chain of a human antibody, and the DNA encoding the variable region of a mouse antibody is invariant of a human antibody. It can be obtained by linking with a DNA encoding a region, inserting it into an expression vector, and introducing into a host to generate an antibody.

Humanized antibodies are obtained by implanting a complementarity determining region (CDR) of a mammal other than a human, eg, a mouse antibody, into the complementarity determining region of a human antibody. Its standard genetic recombination procedures are also known. Specifically, a DNA sequence designed to connect the CDRs of a mouse antibody and a framework region (FR) of a human antibody is synthesized from several oligonucleotides prepared to have a portion overlapping with a terminal portion by the PCR method. The DNA thus obtained is linked with the DNA encoding the human antibody constant region, and then inserted into an expression vector, which is introduced into a host to generate an antibody. The FRs of human antibodies linked via CDRs are chosen such that the complementarity determining regions form good antigen binding sites. If desired, amino acids of the framework regions of the variable regions of the antibody can be substituted such that the complementarity determining regions of the reconstituted humanized antibody form the appropriate antigen binding site.

Moreover, the acquisition method of a human antibody is also known. For example, human lymphocytes are sensitized in vitro with cells expressing a desired antigen or a desired antigen, and sensitized lymphocytes are fused with human myeloma cells, eg, U266, to obtain a desired human antibody having binding activity to the antigen. Can be. In addition, a desired human antibody can be obtained by immunizing a transgenic animal having all repertoires of the human antibody gene with an antigen. In addition, techniques for obtaining human antibodies by panning using human antibody libraries are also known. For example, a phage that expresses the variable region of a human antibody on the surface of the phage by the phage display method as a single-chain antibody (scFv) can be selected. Interpreting the gene of the selected phage can determine the DNA sequence encoding the variable region of the human antibody that binds the antigen. Once the DNA sequence of the scFv that binds to the antigen is known, a suitable expression vector can be prepared based on the sequence to obtain a human antibody. Such methods are already well known in the art.

In addition, antibodies that can be used in one embodiment of the present invention also include antibody fragments such as Fab, (Fab ') ₂ , Fc, Fc', Fd, and reconstituted antibodies such as monovalent or bivalent single-chain antibodies (scFv). .

As used herein, the term "physiologically active protein-containing sample" or "protein-containing sample" or "antibody-containing sample" means a mammalian cell such as a CHO cell containing a bioactive protein molecule or a protein or antibody molecule obtained by culturing. The culture medium or the thing which gave constant treatment, such as partial purification, to this.

As used herein, the term “impurity” includes any substance as long as it is a substance other than the protein of interest, and examples of impurities include protein aggregates, protein fragments, DNA contaminants, viruses, protein A (from columns). Eluted), host cell protein, endotoxin, Hy-Fish (FL) as a medium component, IGF, and the like, but are not limited thereto.

As used herein as an example of impurities, the term "protein aggregate" refers to the binding of two or more proteins of interest. Here, the binding is irreversible binding, the bound protein is not broken down into each. Protein multimers can produce neutralizing antibodies to the protein in the human body, thereby disappearing the efficacy for that protein. Therefore, it is an impurity to be removed in the purification process.

As used herein as an example of an impurity, the term “protein fragment” refers to a fragment protein of a protein that is poorly synthesized and degraded by proteolytic enzymes of that protein, incorrectly synthesized during the protein synthesis process. Means. They are considered impurities because they are ineffective.

As used herein as an example of impurities, the term "DNA contaminant" means DNA in a biologically active protein-containing sample, and includes DNA from a host and DNA from a contaminated virus.

The term "DNA contaminant" used herein as an example of impurities is not particularly limited and includes DNA viruses and RNA viruses. RNA viruses include retroviruses such as X-MuLV, Leo viruses such as Reo3, and parvoviruses such as MVM. Specific examples of viruses removed by the method of the present invention include, for example, X-MuLV, PRV, Reo3, MVM, VSV, herpes simplex, CHV, Sindbis, mumps, vaccinia, Measle, Rubella, influenza, herpes zoster , Cytomegalo, parainfluenza, EB, HIV, HA, HB, NANB, ATL, ECHO, and parvovirus.

As used herein as an example of an impurity, the term "protein A" is a protein A (leached protein A) leached from the Protein A chromatography resin, which is an impurity that can cause an exothermic reaction in the human body.

The present inventors used a statistical experiment (Design of Experiment, DOE) to identify the factors affecting the impurities removal effect according to the calcium phosphate precipitation method of the present invention. DoE means developing an experimental plan to obtain "maximum information" with "minimum experiment" through statistical methods. In general, the purpose of DoE is to organize factors that can be controlled to identify important factors and to find optimal conditions. The inventors planned a statistical experiment (DoE) using a statistical program (Design expert software. Ver. 7.1.1) and performed a statistical analysis (Analysis of variance, ANOVA, variance analysis). The analysis of variance (ANOVA) used in one embodiment of the present invention is used to compare two or more large groups in statistics between groups resulting from the variance in the group, the difference between the total mean and the mean of each group. The F distribution produced by comparing the variances, where the F distribution is the distribution ratio obtained by comparing the variances. This ratio is used to test whether there is a difference in the population variance of each group and whether the population mean is different. Hypothesis test using the Created by statistician and geneticist Ronald Fisher (R.A. Fisher) from the 1920s to the 1930s. The statistical program (Design expert software. Ver. 7.1.1) used for the statistical analysis is widely used for the development and production of pharmaceuticals in the pharmaceutical field.

In the case of purifying proteins using the present invention, in the purification process after Protein A Chromatography, instead of complicated chromatography, a long time and a simple method of effectively removing impurities such as protein multimers Can be removed with

1 illustrates a general antibody purification step. In the antibody purification step of FIG. 1, a calcium phosphate precipitation method according to an embodiment of the present invention may be applied instead of an additional 1 ^st or 2 ^nd chromatography step after the Protein A chromatography step.

Figure 2 shows a purification preparation step for proceeding purification through calcium phosphate precipitation according to an embodiment of the present invention.

Figure 3a and 3b is further purified through calcium phosphate precipitation reaction according to an embodiment of the present invention for the TNFα receptor-Fc fusion protein, the purified anti-ErbB3 antibody for analysis size exclusion chromatography (SEC-HPLC Results of analyzing antibody multimers

Figure 3a: before calcium phosphate precipitation reaction

3b: after calcium phosphate precipitation reaction

Figures 4a and 4b is further purified through the calcium phosphate precipitation reaction according to an embodiment of the present invention for the anti-ErbB3 antibody, the purified TNFα receptor-Fc fusion protein for analysis size exclusion chromatography (SEC-HPLC This is the result of analyzing the protein multimer through

Figure 4a: before calcium phosphate precipitation reaction

4b: after calcium phosphate precipitation reaction

Hereinafter, the present invention will be described in more detail with reference to Examples. These examples are only for illustrating the present invention, but the scope of the present invention is not limited to these examples.

Example

Example 1 Preparation of Intermediate Purified Material for Purification by Calcium Phosphate Precipitation

1-1. Intermediate Purification of Anti-CD20 Antibodies

The CHO (Chines Hamster Ovary Cell) cell culture containing an anti-CD20 antibody was purified by protein A chromatography (MabSelect SuRe Protein A chromatography, GE Healthcare) (see FIG. 2 (A)). The anti-CD20 antibody was eluted from the Protein A column with 50 mM acetate, pH 3.5 buffer, and the eluate was adjusted to pH 3.5 with 1M acetic acid and stored at room temperature for 1 hour. Thereafter, the pH 3.5 eluate was brought back to pH 8.0 with a trisbase, and further purified by anion chromatography (Q sepharose Fast Flow; manufactured by GE Healthcare). The pH 8.0 solution containing the antibody flows out without binding to the anion. That is, purification was performed by a flow through method. The pH 8.0 solution containing the antibody released without binding was adjusted to pH 7.0 using 1 M acetic acid, and the intermediate purification step was further purified as in Example 2 using the calcium phosphate precipitation method of the present invention. did.

1-2. Intermediate Purification of TNFα Receptor-Fc Fusion Proteins

Animal cell cultures containing TNFα receptors and Fc fusion proteins (hereinafter referred to as TNFRc-Fc fusion proteins) were purified by protein A chromatography (see FIG. 2 (B)). TNFRc-Fc fusion protein was eluted from the Protein A column using 50 mM acetate, pH 3.5 buffer, and this eluate was brought to pH 7.0 using trisbase. Further purification was carried out as in Example 2 using the precipitation method.

Example 2: Calcium Phosphate Precipitation Conditions Confirmation Test for Impurity Removal

After adjusting the intermediate purified solution obtained by purification as described in Example 1, each culture solution containing an anti-CD20 antibody or TNFRc-Fc fusion protein produced using animal cells, such as antibody or protein multimer In order to find the factors to remove impurities, the experiment was conducted by planning the experiment as shown in Table 1 using a statistical program (Design expert software ver. 7.1.1 of Stat-Ease, Inc.).

Table 1 Calcium Phosphate Precipitation Condition Confirmation Test for Impurity Removal Exam number Final Na ₃ PO ₄ (mM) Final CaCl ₂ (mM) Response time (hr) Antibody or Protein Concentration Dilution Multiples One 20 20 6 3 2 20 20 0 3 3 0 0 0 One 4 0 0 6 One 5 20 0 0 3 6 ¹⁾ 10 10 3 2 7 0 0 0 3 8 20 20 6 One 9 20 0 0 One 10 20 0 6 One 11 20 0 6 3 12 0 20 6 3 13 0 20 6 One 14 ¹⁾ 10 10 3 2 15 ¹⁾ 10 10 3 2 16 0 0 6 3 17 20 20 0 One 18 0 20 0 3 19 0 20 0 One

¹⁾ The median of all factors was performed three times for statistical analysis.

Example 3: Evaluation of Purification Effect According to Calcium Phosphate Precipitation Conditions

After the test as shown in Table 1, the antibody yield, antibody monomer, antibody multimer, and antibody fragment ratio were measured for the primary evaluation by the calcium phosphate precipitation method.

In the case of antibody yield, the antibody concentration was calculated by substituting the formula 1 below by measuring the absorbance at UV280.

Equation 1

Antibody monomer ratio was determined by analytical size exclusion chromatography (Tosoh's TSKgel G3000SWL Size Exclusion Chromatography Column). High Molecular Weight, Antibody Monomer and Low Molecular Weight (Low Molecular Weight).

Through the first evaluation of the purification method using calcium phosphate precipitation, the results as shown in Table 2 and Table 3 were obtained.

TABLE 2 Anti-CD20 Antibody Yield and Antibody Multimer / Monomer / Segment Ratio According to Calcium Phosphate Precipitation Conditions Exam number Final Na ₃ PO ₄ (mM) Final CaCl ₂ (mM) Reaction time (hr) Dilution factor ¹⁾ Yield after centrifugation (%) Yield after filtration (%) After centrifugation (1000 rpm) After 0.2 μm filtration Antibody multimer ²⁾ (%) Antibody monomer ³⁾ (%) Antibody fragment ⁴⁾ (%) Antibody Multimer (%) Antibody Monomer (%) Antibody Segment (%) One 20 20 6 3 75.12 75.00 0.14 99.86 0.00 0.06 99.94 0.00 2 20 20 0 3 77.40 80.00 0.4 99.60 0.00 0.20 99.80 0.00 3 0 0 0 One 102.56 100.83 1.18 98.82 0.00 1.17 98.83 0.00 4 0 0 6 One 96.41 97.56 1.17 98.83 0.00 1.19 98.81 0.00 5 20 0 0 3 102.27 97.99 1.19 98.81 0.00 1.18 98.82 0.00 6 ⁵⁾ 10 10 3 2 78.33 78.81 0.28 99.72 0.00 0.39 99.61 0.00 7 0 0 0 3 98.17 98.95 1.16 98.84 0.00 1.18 98.82 0.00 8 20 20 6 One 79.74 76.00 0.21 99.79 0.00 0.19 99.81 0.00 9 20 0 0 One 98.58 100.99 1.2 98.80 0.00 1.17 98.83 0.00 10 20 0 6 One 98.74 98.95 1.18 98.82 0.00 1.19 98.81 0.00 11 20 0 6 3 97.66 98.54 1.17 98.83 0.00 1.18 98.82 0.00 12 0 20 6 3 98.38 96.10 1.14 98.86 0.00 1.18 98.82 0.00 13 0 20 6 One 98.06 98.31 1.19 98.81 0.00 1.19 98.81 0.00 14 ⁵⁾ 10 10 3 2 77.36 79.40 0.42 99.58 0.00 0.36 99.64 0.00 15 ⁵⁾ 10 10 3 2 76.04 77.06 0.32 99.68 0.00 0.33 99.67 0.00 16 0 0 6 3 96.00 95.64 0.15 99.85 0.00 0.17 99.83 0.00 17 20 20 0 One 84.04 85.79 0.42 99.58 0.00 0.40 99.60 0.00 18 0 20 0 3 99.64 97.49 1.18 98.82 0.00 1.19 98.81 0.00 19 0 20 0 One 98.38 96.14 1.19 98.81 0.00 1.18 98.82 0.00

^{1) The} anti-CD20 antibody concentration before dilution was 4.8906 mg / mL.

²⁾ The multimer ratio to the initial anti-CD20 antibody was 1.19%.

³⁾ The monomer ratio to the initial anti-CD20 antibody was 98.81%.

⁴⁾ The segment ratio for the initial anti-CD20 antibody was 0.00%.

⁵⁾ The median of all factors was performed three times for statistical analysis.

TABLE 3 TNFRc-Fc Fusion Protein Yield and Antibody Multimer / Monomer / Segment Ratio According to Calcium Phosphate Precipitation Conditions Exam number Final Na ₃ PO ₄ (mM) Final CaCl ₂ (mM) Response time (hr) Dilution factor ¹⁾ Yield after centrifugation (%) Yield after filtration (%) After centrifugation (200 rpm) After 0.2 μm filtration Antibody multimer ²⁾ (%) Antibody monomer ³⁾ (%) Antibody fragment ⁴⁾ (%) Antibody Multimer (%) Antibody Monomer (%) Antibody Segment (%) One 20 20 6 3 68.64 69.99 0.08 99.32 0.60 0.03 99.27 0.70 2 20 20 0 3 80.61 80.13 0.24 99.16 0.60 0.09 99.19 0.72 3 0 0 0 One 99.56 101.84 1.64 97.61 0.75 1.59 97.71 0.70 4 0 0 6 One 101.30 102.21 1.67 97.65 0.68 1.60 97.69 0.71 5 20 0 0 3 99.12 100.05 1.46 97.93 0.61 1.59 97.65 0.76 6 ⁵⁾ 10 10 3 2 80.72 81.15 0.07 99.21 0.72 0.26 99.17 0.57 7 0 0 0 3 100.74 101.15 1.63 97.58 0.79 1.56 97.67 0.77 8 20 20 6 One 82.72 83.10 0.31 99.15 0.54 0.29 99.17 0.54 9 20 0 0 One 99.90 98.96 1.49 97.86 0.65 1.54 97.73 0.73 10 20 0 6 One 99.49 99.72 1.47 97.87 0.66 1.55 97.74 0.71 11 20 0 6 3 96.13 96.76 1.44 97.84 0.72 1.53 97.72 0.75 12 0 20 6 3 95.39 98.75 1.6 97.67 0.73 1.58 97.7 0.72 13 0 20 6 One 99.83 100.09 1.54 97.64 0.82 1.51 97.74 0.75 14 ⁵⁾ 10 10 3 2 82.37 82.28 0.28 99.20 0.52 0.26 99.16 0.58 15 ⁵⁾ 10 10 3 2 86.29 86.84 0.29 99.13 0.58 0.28 99.16 0.56 16 0 0 6 3 98.77 99.17 1.64 97.63 0.73 1.51 97.69 0.80 17 20 20 0 One 88.42 90.39 0.63 98.61 0.76 0.79 98.59 0.62 18 0 20 0 3 98.97 98.28 1.57 97.63 0.8 1.52 97.71 0.77 19 0 20 0 One 100.28 100.32 1.57 97.66 0.77 1.54 97.70 0.76

¹⁾ TNFRc-Fc fusion protein concentration before dilution was 4.70 mg / mL.

^{2) The} multimer ratio to the initial TNFRc-Fc fusion protein was 1.59%.

^{3) The} monomer ratio to the initial TNFRc-Fc fusion protein was 97.71%.

^{4) The} segment ratio for the initial TNFRc-Fc fusion protein was 0.70%.

⁵⁾ The median of all factors was performed three times for statistical analysis.

Example 4 Statistical Analysis of Calcium Phosphate Precipitation Conditions

4-1. Anti-CD20 antibody

Using the results of Table 2, statistical analysis (Analysis of variance, ANOVA, variance analysis) was performed using a statistical program (Design expert software. Ver. 7.1.1) to find factors for purification of anti-CD20 antibodies. . As shown in Table 2, there was no significant difference between the yield of the sample subjected to centrifugation and 0.2 μm filtration and the result of the analytical size exclusion chromatograph. Proceeded with.

If the result wherein -CD20 antibody yield as shown in Table 4, the addition of phosphate concentration (P <0.0051), CaCl ₂ concentration (P <0.0033), and the reaction time (P <0.0453) and then the addition of the phosphate concentration and CaCl ₂ There was a statistical significance of P <0.05 for the correlation of concentrations ( P <0.0042).

That is, in the case of anti-CD20 antibody yield, the phosphate concentration added, the added CaCl ₂ concentration, the reaction time and the correlation between the added phosphate ion concentration and calcium ion concentration are affected.

Table 4 ANOVA Analysis of Anti-CD20 Antibody Yields According to Calcium Phosphate Precipitation Conditions Response; Yield ANOVA for selected factorial model Analysis of variance table [Partial sum of squares-Type III] Source Sum of squares df Mean square F Value p-value Prob> F Model 1178.7 15.0 78.6 53.1 0.0186 A-Phosphate 287.0 1.0 287.0 193.8 0.0051 B-CaCl ₂ 447.5 1.0 447.5 302.2 0.0033 C-Reaction (hour) 30.5 1.0 30.5 20.6 0.0453 D-Protein (mg / mL) 13.8 1.0 13.8 9.3 0.0926 AB 349.1 1.0 349.1 235.7 0.0042 AC 6.9 1.0 6.9 4.6 0.1642 AD 1.9 1.0 1.9 1.3 0.3731 BC 2.2 1.0 2.2 1.5 0.3467 BD 0.0 1.0 0.0 0.0 0.9362 CD 0.9 1.0 0.9 0.6 0.5187 ABC 26.7 1.0 26.7 18.0 0.0513 ABD 2.5 1.0 2.5 1.7 0.3237 ACD 7.5 1.0 7.5 5.1 0.1528 BCD 0.1 1.0 0.1 0.1 0.8117 ABCD 2.0 1.0 2.0 1.4 0.3609 Curvature 566.1 1.0 566.1 382.2 0.0026 Pure Error 3.0 2.0 1.5 - - Cor total 1747.7 18.0 - - - Std. Dev. 1.2 - R-Squared 1.00 - Mean 91.0 - Adj R-Squared 1.00 - CV% 1.3 - Pred R-Squared N / A - PRESS N / A - Adeq Precision 22.6 -

And through analytical size exclusion chromatography (SEC) results for the anti-CD20 antibody, as shown in Table 5, added phosphate concentration (P <0.0017), added CaCl ₂ concentration (P <0.0024), reaction time (P <0.0082), anti-CD20 antibody concentration ( P <0.0140), correlation between added phosphate concentration and CaCl ₂ concentration ( P <0.0006), correlation between added phosphate concentration and reaction time ( P <0.0315), added Correlation between phosphate concentration and anti-CD20 antibody concentration ( P <0.0298), correlation of added CaCl ₂ concentration and reaction time ( P <0.0355), correlation between added CaCl ₂ concentration and anti-CD20 antibody concentration ( P <0.0298), the correlation between reaction time and anti-CD20 antibody concentration (P <0.0141), added phosphate concentration and added CaCl ₂ concentration, Correlation between reaction time (P <0.0077), added phosphate concentration and added CaCl ₂ concentration, Correlation with anti-CD20 antibody concentration (P <0.0079), added phosphate concentration and reaction time, correlation with anti-CD20 antibody concentration (P <0.0117), added CaCl ₂ concentration and reaction time, and Wherein the mutual relationship between the -CD20 antibody concentration (P <0.0117) and then added with the addition of a phosphate concentration of CaCl ₂ concentration, reaction time and, There was a statistical significance of P <0.05 for the correlation (P <0.0166) with the anti-CD20 antibody concentration.

In other words, factors affecting the removal of the anti-CD20 antibody multimers are the added calcium ion concentration, reaction time, the concentration of anti-CD20 antibody, the correlation between the added phosphate and calcium ion concentrations, and the added phosphate ion concentration. Relationship between the reaction time and the reaction time, the correlation between the added phosphate concentration and the anti-CD20 antibody concentration, the correlation between the added calcium ion concentration and the reaction time, the correlation between the added calcium ion concentration and the anti-CD20 antibody concentration, and the reaction Correlation between time and anti-CD20 antibody concentration, added phosphate and calcium ion concentrations, and Correlation between reaction time , added phosphate and calcium ion concentrations, Correlation with anti-CD20 antibody concentration, added phosphate concentration and reaction time, and correlation with anti-CD20 antibody concentration, added calcium ion concentration and reaction time, and And wherein the antibody concentration -CD20 interaction, and then added with the addition of a phosphate ion concentration, calcium ion concentration, reaction time, and Correlation with Anti-CD20 Antibody Concentrations Affects anti-CD20 antibody multimer removal.

Table 5 ANOVA Analysis of Anti-CD20 Antibody Ratio According to Calcium Phosphate Precipitation Conditions Response; SEC ANOVA for selected factorial model Analysis of variance table [Partial sum of squares-Type III] Source Sum of squares df Mean square F Value p-value Prob> F Model 3.3 15.0 0.2 248.0 0.0040 A-Phosphate 0.5 1.0 0.5 576.0 0.0017 B-CaCl ₂ 0.5 1.0 0.5 560.1 0.0018 C-Reaction (hour) 0.1 1.0 0.1 121.0 0.0082 D-Protein (mg / mL) 0.1 1.0 0.1 124.7 0.0079 AB 1.5 1.0 1.5 1667.4 0.0006 AC 0.0 1.0 0.0 30.2 0.0315 AD 0.0 1.0 0.0 32.1 0.0298 BC 0.0 1.0 0.0 26.7 0.0355 BD 0.0 1.0 0.0 32.1 0.0298 CD 0.1 1.0 0.1 69.4 0.0141 ABC 0.1 1.0 0.1 128.4 0.0077 ABD 0.1 1.0 0.1 124.7 0.0079 ACD 0.1 1.0 0.1 84.0 0.0117 BCD 0.1 1.0 0.1 84.0 0.0117 ABCD 0.1 1.0 0.1 58.8 0.0166 Curvature 0.7 1.0 0.7 748.1 0.0013 Pure Error 0.0 2.0 0.0 - - Cor total 4.0 18.0 - - - Std. Dev. 0.0 - R-Squared 1.00 - Mean 99.2 - Adj R-Squared 1.00 - CV% 0.0 - Pred R-Squared N / A - PRESS N / A - Adeq Precision 39.82 -

4-2. TNFRc-Fc fusion protein

Using the results of Table 3, statistical analysis (Analysis of variance, ANOVA, variance analysis) was performed using a statistical program (Design expert software. Ver. 7.1.1) to find the factors for the purification of TNFRc-Fc fusion protein. It was. As shown in Table 3, there was no significant difference between the yield of the sample subjected to centrifugation and 0.2 μm filtration and the results of the analytical size exclusion chromatograph. Proceeded with.

As a result, in the TNFRc-Fc fusion protein yield as shown in Table 6, the concentration of the added phosphate ( P <0.0206), the added CaCl ₂ concentration ( P <0.0226), and the added phosphate and added CaCl ₂ concentration There was a statistical significance of P <0.05 for the correlation with ( P <0.0327).

That is, the yield of TNFRc-Fc fusion protein is affected by the correlation between the added phosphate concentration, the added calcium ion concentration, and the added forest ion concentration and calcium ion concentration.

Table 6 ANOVA Analysis on the Yield of TNFRc-Fc Fusion Proteins According to Calcium Phosphate Precipitation Conditions Response; Yield ANOVA for selected factorial model Analysis of variance table [Partial sum of squares-Type III] Source Sum of squares df Mean square F Value p-value Prob> F Model 1309.0 15.0 87.3 9.6 0.0981 A-Phosphate 427.6 1.0 427.6 47.1 0.0206 B-CaCl ₂ 388.2 1.0 388.2 42.8 0.0226 C-Reaction (hour) 28.4 1.0 28.4 3.1 0.2187 D-Protein (mg / mL) 65.4 1.0 65.4 7.2 0.1152 AB 263.7 1.0 263.7 29.1 0.0327 AC 21.6 1.0 21.6 2.4 0.2628 AD 20.5 1.0 20.5 2.3 0.2713 BC 10.6 1.0 10.6 1.2 0.3920 BD 28.0 1.0 28.0 3.1 0.2213 CD 4.6 1.0 4.6 0.5 0.5515 ABC 17.5 1.0 17.5 1.9 0.2990 ABD 29.8 1.0 29.8 3.3 0.2115 ACD 1.7 1.0 1.7 0.2 0.7056 BCD 1.1 1.0 1.1 0.1 0.7580 ABCD 0.2 1.0 0.2 0.0 0.8921 Curvature 341.9 1.0 341.9 37.7 0.0255 Pure Error 18.1 2.0 9.1 - - Cor total 1669.1 18.0 - - - Std. Dev. 3.0 - R-Squared 0.99 - Mean 93.2 - Adj R-Squared 0.88 - CV% 3.2 - Pred R-Squared N / A - PRESS N / A - Adeq Precision 11.31 -

And, as shown in Table 7 through the analytical size exclusion chromatography (SEC) results for the TNFRc-Fc fusion protein, added phosphate concentration ( P <0.0001), CaCl ₂ concentration ( P <0.0001), reaction time, TNFRc-Fc fusion protein concentration ( P <0.0072), added phosphate concentration and CaCl ₂ concentration correlation ( P <0.0007), added phosphate concentration and reaction time correlation ( P <0.0001), added phosphate concentration and TNFRc Correlation between -Fc fusion protein concentration ( P <0.0011), Correlation between added CaCl ₂ concentration and reaction time ( P <0.0008), Correlation between added CaCl ₂ concentration and TNFRc-Fc fusion protein concentration ( P <0.0006), correlation between reaction time and TNFRc-Fc fusion protein concentration ( P <0.0013), correlation between added phosphate concentration and CaCl ₂ concentration and reaction time ( P <0.0099), reaction with added phosphate concentration mutual relationship between the time and the TNFRc-Fc fusion protein concentrations (P <0.0011), was added Phosphate concentration and CaCl ₂ concentrations, and TNFRc-Fc interaction with the fusion protein concentration (P <0.0019), the addition of CaCl ₂ concentration and the reaction time and the mutual relationship between the TNFRc-Fc fusion protein concentrations (P <0.0008), was added There was a statistically significant difference of P <0.05 for the correlation between phosphate concentration, CaCl ₂ concentration, reaction time and TNFRc-Fc fusion protein concentration ( P <0.0020).

In other words, the factors affecting the removal of TNFRc-Fc fusion protein multimers were the concentration of added phosphate concentration, added calcium ion concentration, reaction time, TNFRc-Fc fusion protein concentration, added phosphate ion concentration and calcium ion concentration. Correlation, correlation between added phosphate ion concentration and reaction time, correlation between added phosphate ion concentration and TNFRc-Fc fusion protein concentration, correlation between added calcium ion concentration and reaction time, added calcium ion concentration And the relationship between TNFRc-Fc fusion protein concentration, reaction time and TNFRc-Fc fusion protein concentration, added phosphate and calcium ion concentration and reaction time, added phosphate ion concentration and reaction time And the relationship between TNFRc-Fc fusion protein concentration, added phosphate ion concentration and reaction time, TNFRc-Fc fusion protein concentration, added calcium ion concentration and reaction time, and Correlation with TNFRc-Fc fusion protein concentration, added phosphate and calcium ion concentrations, reaction time and TNFRc-Fc fusion protein concentration.

TABLE 7 ANOVA Analysis of Monomer Ratio in TNFRc-Fc Fusion Proteins According to Calcium Phosphate Precipitation Conditions Response; SEC ANOVA for selected factorial model Analysis of variance table [Partial sum of squares-Type III] Source Sum of squares df Mean square F Value p-value Prob> F Model 5.8 15.0 0.4 11552.5 <0.0001 A-Phosphate 1.9 1.0 1.9 55692.2 <0.0001 B-CaCl ₂ 1.9 1.0 1.9 56101.7 <0.0001 C-Reaction (hour) 0.0 1.0 0.0 1111.7 0.0009 D-Protein (mg / mL) 0.0 1.0 0.0 526.7 0.0019 AB 1.7 1.0 1.7 52470.2 <0.0001 AC 0.0 1.0 0.0 945.2 0.0011 AD 0.0 1.0 0.0 841.7 0.0012 BC 0.0 1.0 0.0 697.7 0.0014 BD 0.0 1.0 0.0 1230.2 0.0008 CD 0.0 1.0 0.0 379.7 0.0026 ABC 0.0 1.0 0.0 567.2 0.0018 ABD 0.0 1.0 0.0 1170.2 0.0009 ACD 0.0 1.0 0.0 346.7 0.0029 BCD 0.0 1.0 0.0 792.2 0.0013 ABCD 0.0 1.0 0.0 414.2 0.0024 Curvature 3.2 1.0 3.2 95318.1 <0.0001 Pure Error 0.0 2.0 0.0 - - Cor total 9.0 18.0 - - - Std. Dev. 0.0 - R-Squared 1.00 - Mean 98.2 - Adj R-Squared 1.00 - CV% 0.0 - Pred R-Squared N / A - PRESS N / A - Adeq Precision 296.64 -

Example 5 pH Condition Confirmation Test of Calcium Phosphate Precipitation

Based on the results of the statistical analysis (ANOVA, variance analysis) of Tables 4 to 7, in order to enhance the protein yield and protein multimer elimination effect, further tests, such as Table 8 below, were conducted.

Table 8 PH Condition Confirmation Test of Calcium Phosphate Precipitation Exam number Final Na ₃ PO ₄ (mM) Final CaCl ₂ (mM) Response time (hr) Antibody / protein concentration dilution factor ¹⁾ pH One 5 10 3 3.0 5.50 2 5 10 3 1.0 5.50 3 5 10 5 3.0 5.50 4 5 10 3 3.0 7.50 5 ²⁾ 5 10 4 2.0 6.50 6 ²⁾ 5 10 4 2.0 6.50 7 5 10 5 1.0 7.50 8 5 10 3 1.0 7.50 9 5 10 5 1.0 5.50 10 5 10 5 3.0 7.50 11 ²⁾ 5 10 4 2.0 6.50

^{1) The} pre-dilution anti-CD20 antibody and TNFRc-Fc fusion protein concentrations were 4.8906 mg / mL and 4.7047 mg / mL, respectively.

²⁾ The median of all factors was performed three times for statistical analysis.

Example 6: Statistical Evaluation of pH Condition Identification Test for Calcium Phosphate Precipitation

6-1. Anti-CD20 antibody

Results of statistical analysis (ANOVA, variance analysis) of the test were obtained as shown in Table 9 below.

Table 9 Additional Calcium Phosphate Precipitation Test Results for Anti-CD20 Antibody Multimer Removal Exam number Final Na ₃ PO ₄ (mM) Final CaCl ₂ (mM) Reaction time (hr) Dilution drainage ¹⁾ pH After 0.2 μm filtration Yield (%) Antibody Multimer (%) Antibody monomer ³⁾ (%) Antibody Segment (%) One 5 10 3 3.0 5.50 100.29 0.59 99.41 0.00 2 5 10 3 1.0 5.50 95.44 0.40 99.60 0.00 3 5 10 5 3.0 5.50 94.86 0.64 99.36 0.00 4 5 10 3 3.0 7.50 75.05 0.32 99.68 0.00 5 ²⁾ 5 10 4 2.0 6.50 82.42 0.22 99.78 0.00 6 ²⁾ 5 10 4 2.0 6.50 85.29 0.16 99.84 0.00 7 5 10 5 1.0 7.50 44.82 0.00 100.00 0.00 8 5 10 3 1.0 7.50 41.64 0.00 100.00 0.00 9 5 10 5 1.0 5.50 99.20 0.53 99.47 0.00 10 5 10 5 3.0 7.50 76.69 0.35 99.65 0.00 11 ²⁾ 5 10 4 2.0 6.50 85.34 0.23 99.77 0.00

^{1) The} anti-CD20 antibody concentration before dilution was 4.8906 mg / mL, respectively.

²⁾ The median of all factors was performed three times for statistical analysis.

³⁾ The multimer (%) and monomer (%) of the untreated anti-CD20 antibody were 0.78% and 99.28%, respectively.

Table 9 with the result of the CaCl ₂ and insanyeomwa additional factors to find statistics program added for purification (. Design expert software. Ver 7.1.1 ) utilized in the statistical analysis (Analysis of variance, ANOVA, analysis of variance) a Proceeded.

As a result, as shown in Table 10, at 5 mM phosphate concentration and 10 mM CaCl ₂ concentration, the anti-CD20 antibody yield was determined by the reaction pH ( P <0.0009), the concentration of anti-CD20 antibody ( P <0.0058), There was statistical significance of P <0.05 for the correlation ( P <0.0047) of the anti-CD20 antibody.

In other words, the anti-CD20 antibody yield in the purification method of anti-CD20 antibody by calcium phosphate precipitation, in addition to the added calcium and phosphate ions, the concentration of the anti-CD20 antibody, reaction pH, and anti-CD20 antibody concentration and reaction pH Affected by the interrelationship of

Table 10 ANOVA Analysis on the Yield of Anti-CD20 Antibodies According to Calcium Phosphate Precipitation Conditions Response; Yield ANOVA for selected factorial model Analysis of variance table [Partial sum of squares-Type III] Source Sum of squares df Mean square F Value p-value Prob> F Model 4119.71 7 588.53 210.62 0.0047 A-Hold Time (hour) 0.11 One 0.11 0.04 0.8593 B-Protein (mg / mL) 475.71 One 475.71 170.24 0.0058 C-pH 3029.92 One 3029.92 1084.32 0.0009 AB 5.49 One 5.49 1.97 0.2959 AC 14.02 One 14.02 5.02 0.1544 BC 592.88 One 592.88 212.18 0.0047 ABC 1.58 One 1.58 0.56 0.5311 Curvature 62.19 One 62.19 22.25 0.0421 Pure Error 5.59 2 2.79 - - Cor total 4187.49 10 - - - Std. Dev. 1.67 - R-Squared 1.00 - Mean 80.47 - Adj R-Squared 0.99 - CV% 2.08 - Pred R-Squared N / A - PRESS N / A - Adeq Precision 38.79 -

In addition, through analytical size exclusion chromatography (SEC) results for anti-CD20 antibodies as shown in Table 11, the anti-CD20 antibody multimer removal ability was fixed at 5 mM phosphate concentration and 5 mM CaCl ₂ concentration. There was a statistical significance of P <0.05 for the concentration of ( P <0.0120) and the reaction pH ( P <0.0051).

That is, the degree of anti-CD20 antibody multimer removal in the purification method of the anti-CD20 antibody by calcium phosphate precipitation is influenced by the concentration reaction pH of the anti-CD20 antibody in addition to the added calcium and phosphate ions.

Table 11 ANOVA Analysis of Anti-CD20 Antibody Monomer According to Calcium Phosphate Precipitation Conditions Response; SEC ANOVA for selected factorial model Analysis of variance table [Partial sum of squares-Type III] Source Sum of squares df Mean square F Value p-value Prob> F Model 0.42 7 0.06 42.10 0.0234 A-Hold Time (Hour) 0.01 One 0.01 3.85 0.1889 B-Protein (mg / mL) 0.12 One 0.12 82.06 0.0120 C-pH 0.28 One 0.28 193.61 0.0051 AB 0.00 One 0.00 0.22 0.6865 AC 0.00 One 0.00 1.96 0.2963 BC 0.02 One 0.02 11.94 0.0745 ABC 0.00 One 0.00 1.06 0.4123 Curvature 0.05 One 0.05 34.44 0.0278 Pure Error 0.00 2 0.00 - - Cor total 0.47 10 - - - Std. Dev. 0.04 - R-Squared 0.99 - Mean 99.69 - Adj R-Squared 0.97 - CV% 0.04 - Pred R-Squared N / A - PRESS N / A - Adeq Precision 18.69 -

Based on the results of Table 10 and Table 11, eight purification conditions as shown in Table 12 through the statistical program (Design expert software. Ver. 7.1.1) to optimize the optimization conditions for the anti-CD20 antibody purification using calcium phosphate precipitation method Got. These eight conditions are limited to reaction time (3-5 hours), anti-CD20 antibody concentration (1X-3X dilution), pH 6.0-7.0, wherein the anti-CD20 antibody yield is at least 90% and anti- These are the conditions obtained by statistical calculation when the ratio of the CD20 antibody monomer ratio is limited to the condition that can obtain 99.5% or more.

Table 12 Optimization of Anti-CD20 Antibody Purification Using Calcium Phosphate Precipitation Constraints Name Goal Lower limit Upper limit Lower weight Upper weight Importance Hold Time (Hour) 3-5 3 5 One One 3 Protein (mg / mL) 1-3 One 3 One One 3 pH 6.50 6 7 One One 3 Yield maximize 90 100.29 3 One 3 SEC maximize 99.5 100 5 One 5 Solutions Number Hold time Protein pH Yield SEC Desirability One 3 2.37 6.02 90.76 99.55 0.00029 2 3 2.35 6.02 90.72 99.55 0.00029 3 3 2.4 6.03 90.83 99.55 0.00029 4 3 2.34 6.02 90.64 99.55 0.00028 5 3 2.43 6.04 90.71 99.55 0.00027 6 3 2.45 6.05 90.74 99.55 0.00026 7 3 2.47 6.06 90.66 99.55 0.00025 8 3 2.44 6.01 91.34 99.54 0.00021

6-2. TNFRc-Fc fusion protein

Table 8 Results of statistical analysis (ANOVA, variance analysis) The results as shown in Table 13 below were obtained.

Table 13 Additional Calcium Phosphate Precipitation Tests for TNFRc-Fc Fusion Protein Multimer Removal Exam number Final Na ₃ PO ₄ (mM) Final CaCl ₂ (mM) Reaction time (hr) Dilution drainage ¹⁾ pH After 0.2 μm filtration Yield (%) Antibody Multimer (%) Antibody Monomer (%) Antibody Segment (%) One 10 10 3 3.0 5.50 99.20 1.77 97.72 0.51 2 10 10 3 1.0 5.50 97.37 1.64 97.83 0.53 3 10 10 5 3.0 5.50 97.71 1.84 97.43 0.73 4 10 10 3 3.0 7.50 83.23 0.57 98.66 0.77 5 ²⁾ 10 10 4 2.0 6.50 80.21 0.29 98.98 0.73 6 ²⁾ 10 10 4 2.0 6.50 80.58 0.28 98.98 0.74 7 10 10 5 1.0 7.50 68.68 0.15 98.50 1.35 8 10 10 3 1.0 7.50 80.00 0.09 98.75 1.16 9 10 10 5 1.0 5.50 97.69 1.6 97.63 0.77 10 10 10 5 3.0 7.50 87.24 0.68 98.29 1.03 11 ²⁾ 10 10 4 2.0 6.50 81.87 0.30 98.96 0.74

¹⁾ TNFRc-Fc fusion protein concentration before dilution was 4.7047 mg / mL.

²⁾ The median of all factors was performed three times for statistical analysis.

³⁾ The multimer (%), monomer (%) and segment (%) of the untreated TNFRc-Fc fusion protein were 2.11%, 96.58% and 1.3%, respectively.

Table 13 with the results of the addition of CaCl ₂ to the purified and phosphate and an additional factor to find statistical program (Design expert software. Ver. 7.1.1 ) to statistical analysis using the (Analysis of variance, ANOVA, analysis of variance) Proceeded.

As a result, as shown in Table 14, at 10 mM phosphate concentration and 10 mM CaCl ₂ concentration, the yield of TNFRc-Fc fusion protein was calculated as the concentration of TNFRc-Fc fusion protein ( P <0.0107), reaction pH ( P <0.0011). , The relationship between the reaction time and the concentration of TNFRc-Fc fusion protein ( P <0.0317), the relationship between the TNFRc-Fc fusion protein concentration and the reaction pH ( P <0.0149), the reaction time and the concentration of TNFRc-Fc fusion protein, and , P <0.05 for the correlation of reaction pH ( P <0.0201).

In other words, the yield of TNFRc-Fc fusion protein in the purification method of TNFRc-Fc fusion protein by calcium phosphate precipitation, in addition to the calcium and phosphate ions added, the concentration, reaction pH, reaction time and TNFRc-Fc of TNFRc-Fc fusion protein The relationship between the fusion protein, the TNFRc-Fc fusion protein concentration and the reaction pH, and the reaction time, the concentration of the TNFRc-Fc fusion protein, and the reaction pH are affected.

Table 14 ANOVA Analysis on the Yield of TNFRc-Fc Fusion Proteins According to Calcium Phosphate Precipitation Conditions Response; Yield ANOVA for selected factorial model Analysis of variance table [Partial sum of squares-Type III] Source Sum of squares df Mean square F Value p-value Prob> F Model 855.7 7.0 122.2 161.0 0.0062 A-Hold Time (Hour) 9.0 1.0 9.0 11.8 0.0751 B-Protein (mg / mL) 69.9 1.0 69.9 92.0 0.0107 C-pH 662.8 1.0 662.8 872.8 0.0011 AB 22.8 1.0 22.8 30.1 0.0317 AC 4.7 1.0 4.7 6.2 0.1304 BC 49.7 1.0 49.7 65.4 0.0149 ABC 36.7 1.0 36.7 48.4 0.0201 Curvature 139.8 1.0 139.8 184.0 0.0054 Pure Error 1.5 2.0 0.8 - - Cor total 996.9 10.0 - - - Std. Dev. 0.90 - R-Squared 1.00 - Mean 86.70 - Adj R-Squared 0.99 - CV% 1.02 - Pred R-Squared N / A - PRESS N / A - Adeq Precision 38.72 -

In addition, as shown in Table 15, TNFRc-Fc fusion protein multimers when the phosphate concentration and the CaCl ₂ concentration were fixed at 10 mM through analytical size exclusion chromatography (SEC) results for the TNFRc-Fc fusion protein, respectively. Scavenging ability was determined by the concentration of TNFRc-Fc fusion protein ( P <0.0004), reaction pH ( P <0.0001), correlation between reaction time and TNFRc-Fc fusion protein concentration ( P <0.007), and correlation between reaction time and reaction pH. (P <0.0014), for TNFRc-Fc fusion protein interaction (P <0.0011) of concentration and reaction pH, and reaction time and TNFRc-Fc fusion protein concentrations, then the reaction pH correlation (0.001) P < There was a statistical significance of 0.05.

In other words, TNFRc-Fc fusion protein multimer removal in the purification method of TNFRc-Fc fusion protein by calcium phosphate precipitation method, in addition to the added calcium and phosphate ions, TNFRc-Fc fusion protein concentration, reaction pH, reaction time and The relationship between TNFRc-Fc fusion protein concentration, reaction time and reaction pH, TNFRc-Fc fusion protein concentration and half pH, and reaction time and TNFRc-Fc fusion protein concentration, and reaction pH get affected.

Table 15 ANOVA Analysis of TNFRc-Fc Fusion Protein Monomers with Calcium Phosphate Precipitation Conditions Response; SEC ANOVA for selected factorial model Analysis of variance table [Partial sum of squares-Type III] Source Sum of squares df Mean square F Value p-value Prob> F Model 3.49 7 0.50 3741 0.0003 A-Hold Time (Hour) 0.00 One 0.00 11 0.0780 B-Protein (mg / mL) 0.32 One 0.32 2430 0.0004 C-pH 2.63 One 2.63 19751 <0.0001 AB 0.18 One 0.18 1373 0.0007 AC 0.09 One 0.09 710 0.0014 BC 0.12 One 0.12 919 0.0011 ABC 0.13 One 0.13 995 0.0010 Curvature 1.22 One 1.22 9133 0.0001 Pure Error 0.00 2 0.00 - - Cor total 4.71 10 - - - Std. Dev. 0.01 - R-Squared 1.00 - Mean 98.43 - Adj R-Squared 1.00 - CV% 0.01 - Pred R-Squared N / A - PRESS N / A - Adeq Precision 198.19 -

Based on the results of Table 14 and Table 15, 10 purification conditions as shown in Table 16 through the statistical program (Design expert software. Ver. 7.1.1), the optimization conditions for the purification of TNFRc-Fc fusion protein using calcium phosphate precipitation method Got. These 10 conditions are limited to reaction time (3-5 hours), TNFRc-Fc fusion protein concentration (1X-3X dilution), pH 6.0-7.0, where TNFRc-Fc fusion protein yield is 80% or higher and TNFRc The conditions obtained by statistical calculation when the -Fc fusion protein monomer ratio was limited to a condition that can obtain more than 98.5%.

Table 16 Optimization of TNFRc-Fc Fusion Protein Purification Using Calcium Phosphate Precipitation Constraints Name Goal Lower limit Upper limit Lower weight Upper weight Importance Hold time maximize 3 5 One One 3 Protein maximize One 3 One One 3 pH 5.5-7.5 5.5 7.5 One One 3 Yield maximize 80 99.2 One One 3 SEC maximize 98.5 98.98 5 One 5 Solutions Number Hold time Protein pH Yield SEC Desirability 1.0 3.1 3.0 7.5 83.4 98.6 0.0 2.0 3.1 3.0 7.5 83.4 98.6 0.0 3.0 3.1 3.0 7.5 83.4 98.6 0.0 4.0 3.1 3.0 7.5 83.4 98.6 0.0 5.0 3.1 3.0 7.5 83.1 98.6 0.0 6.0 3.1 3.0 7.5 83.5 98.6 0.0 7.0 3.1 3.0 7.5 83.0 98.6 0.0 8.0 3.2 3.0 7.5 83.6 98.6 0.0 9.0 3.1 2.9 7.5 82.3 98.7 0.0 10.0 3.1 2.8 7.5 81.4 98.7 0.0

Example 7 TNFRc-Fc Fusion Protein Purification Effect Test by Calcium Phosphate Precipitation

The TNFα receptor-Fc (TNFRc-Fc) fusion protein tested in the above Example was subjected to calcium phosphate precipitation reaction under the conditions shown in Table 17, and the multimer weight of the TNFRc-Fc fusion protein before and after the reaction was compared.

As a result, as shown in Figure 3a, 3b, it was confirmed that the multimer of the TNFRc-Fc fusion protein was reduced by about 82% from 1.59% to 0.29%.

Table 17 Calcium Phosphate Precipitation Reaction Conditions for TNFRc-Fc Fusion Proteins Final Na ₃ PO ₄ (mM) Final CaCl ₂ (mM) Response time (hr) TNFRc-Fc Fusion Protein Concentration (mg / mL) 20 20 6 4.70 mg / mL

Example 8 Anti-ErbB3 Antibody Purification Effect Test by Calcium Phosphate Precipitation

In addition to the anti-CD20 antibody and the TNFRc-Fc fusion protein tested in the above example, it was confirmed with the anti-ErbB3 antibody whether the calcium phosphate precipitation method is effective. Animal cell cultures containing anti-ErbB3 antibodies were purified by Protein A chromatography. Anti-ErbB3 antibody was eluted from the Protein A column using 50 mM acetate and pH 3.5 buffer, and the eluate containing the anti-ErbB3 antibody was adjusted to pH 7.0 with trisbase, and calcium phosphate precipitated under the conditions shown in Table 18 below. The reaction was performed to compare the amount of anti-ErbB3 antibody multimers before and after the reaction.

As a result, as shown in Figures 4a, 4b, it was confirmed that the multimer of the anti-ErbB3 antibody was reduced by about 45% from 4.18% to 2.28%.

Table 18 Calcium Phosphate Precipitation Reaction Conditions for Anti-ErbB3 Antibodies Final Na ₃ PO ₄ (mM) Final CaCl ₂ (mM) Response time (hr) Anti-ErbB3 Antibody Concentration (mg / mL) 10 10 3 12.2280

Example 9: Test to confirm the effect of removing impurities other than the protein multimer by calcium phosphate precipitation

In order to confirm that the purification method using calcium phosphate precipitation is effective in removing impurities other than antibody or protein multimer removal, the corresponding antibodies obtained in Table 12 and Table 16 using anti-CD20 antibody and TNFRc-Fc fusion protein Or selected under conditions optimal for multimer removal of the protein.

As shown in Table 19 below, calcium phosphate precipitation was performed on the anti-CD20 antibody and TNFRc-Fc fusion protein and analyzed for impurities such as host cell derived protein, DNA, and protein A residue.

As a result, it was confirmed that the purification method using calcium phosphate precipitation, as shown in Table 20, is effective in removing host cell-derived protein, DNA, and protein A residues in addition to removing multimers of antibodies. However, the yield, multimer removal capacity, and impurities removal ability were different depending on the protein. In other words, the reaction time, the added calcium ion concentration, the added phosphate ion concentration means that the conditions may vary depending on the protein.

Table 19 Calcium Phosphate Precipitation Reaction Conditions for Anti-CD20 Antibodies and TNFRc-Fc Fusion Proteins protein Calcium phosphate precipitation conditions Final Na ₃ PO ₄ (mM) Final CaCl ₂ (mM) Protein concentration (mg / mL) Reaction pH Response time (hr) Anti-CD20 antibody 5 10 4.8906 6.03 2.4 TNFRc-Fc fusion protein 10 10 4.7047 7.5 3.1

Table 20 Additional Purification Effect by Calcium Phosphate Precipitation protein Anti-CD20 antibody TNFRc-Fc fusion protein Before the reaction After the reaction Before the reaction After the reaction Antibody Yield (%) - 98.8 - 79.18 Monomer (%) 98.91 99.28 96.48 98.61 Host Cell Derived Protein (ppm) 3 One 655 21 DNA (ppb) LOQ ⁽¹⁾ LOQ ⁽¹⁾ 15.8 <3.2 (below detection limit) Protein A (ppm) 1.01 LOQ ⁽²⁾ 92.71 2.46

LOQ; Limit of Quantification

⁽¹⁾ Since the amount of residual DNA in the sample before the anti-CD20 antibody reaction was below the detection limit, DNA analysis was not performed on the samples before and after the reaction.

^{(2) Since the} amount of residual protein A in the sample before the anti-CD20 reaction was too low at 1.01 ppm, the analysis was not performed on the sample after the reaction.

Claims (9)

A method for removing impurities in a protein containing sample comprising the following steps: a) purifying the protein-containing sample by applying it to Protein A affinity chromatography; b) adding a calcium ion solution and a phosphate ion solution to the purified protein to form a calcium phosphate precipitation reaction, and then removing the precipitate by centrifugation or filtration; And c) ion exchange chromatography, hydrophobic interaction chromatography and mixed mode chromatography of the purified protein after step b) or between steps a) and b). Purifying by applying to any one of the chromatography selected from the group consisting of mixed mode chromatography. The method of claim 1, The calcium ion solution is CaCl ₂ or CaCO ₃ , The phosphate ion solution is characterized in that Na ₃ PO ₄ , NaH ₂ PO ₄ , Na ₂ HPO _4, KH ₂ PO _4, K ₂ HPO _4, or K ₃ PO ₄ How to. The method of claim 2, The concentration of calcium ions added to the calcium phosphate precipitation reaction is 1 to 100 mM, the concentration of phosphate ions is 1 to 100 mM. The method of claim 3, The calcium phosphate precipitation reaction is characterized in that carried out at pH 5.0 to 10.0. The method of claim 4, wherein The calcium phosphate precipitation reaction time is characterized in that 0 to 10 hours. The method of claim 5, The calcium phosphate precipitation reaction method characterized in that the concentration of the protein containing sample diluted 1 to 10 times. The method according to any one of claims 1 to 6, Wherein said impurities are protein multimers, protein fragments, DNA contaminants, viruses, host cell derived proteins or protein A. The method according to any one of claims 1 to 6, The protein is an antibody or an Fc fusion protein. A method for producing a protein preparation comprising a purification step using the method of any one of claims 1 to 6.

Images