Evaluation methods for protein-containing solutions

Abstract

A method for evaluating the clogging of a filter membrane containing a protein solution includes the following steps: step a), passing the protein-containing solution through the filter membrane; step b), obtaining a cross-section of the filter membrane after step a); step c), treating the protein-containing solution before step a), or treating the filter membrane before step b), or treating the cross-section of the filter membrane after step b), using at least one staining agent specific to protein aggregates; and step d), confirming the presence of protein aggregates in the cross-section of the filter membrane.

Description

Technical Field

[0001] This invention relates to a method for evaluating protein-containing solutions. Background Technology

[0002] In recent years, in addition to plasma fractionation preparations derived from human blood, the development of protein-based formulations, such as biopharmaceuticals utilizing gene recombination technology, has also flourished. Because these protein preparations are derived from biological sources, they contain many impurities, and their manufacturing process includes multiple purification stages. Purification utilizes separation techniques such as centrifugation, membrane separation, and column chromatography. Among these, membrane separation is a useful method for separating impurities at various sizes.

[0003] One of the problems in membrane separation is the reduced filtration rate due to membrane clogging. There are generally three mechanisms of membrane clogging. The first is complete blockage, where large particles clog the membrane pores. The second is standard blockage, where particles adsorb onto the pore walls, narrowing the pores. The third is cake filtration, where small particles concentrate and accumulate on the membrane surface, creating resistance (see, for example, Non-Patent Literature 1). To avoid these problems, those working in purification processes need to select appropriate membranes and study filtration conditions to optimize the process efficiency.

[0004] Existing technical documents

[0005] Patent documents

[0006] Patent Document 1: Japanese Patent Publication No. 2014-517047

[0007] Patent Document 2: International Publication No. 2015 / 156403

[0008] Non-patent literature

[0009] Non-patent literature 1: Membrane, 36(5), 211-216 (2011)

[0010] Non-patent document 2: Guidance for Industry Immunogenicity Assessment for Therapeutic Protein Products (2014, USDepartment of Health and Human ServicesFood and Drug Administration) Summary of the Invention

[0011] The problem the invention aims to solve

[0012] Proteins can aggregate due to various stress responses and stimuli. If protein aggregates (hereinafter sometimes simply referred to as "aggregates") are present, large aggregates can cause complete membrane clogging during membrane separation. Virus removal processes, often used in purification procedures, involve separating the target protein from viruses slightly larger than it using a membrane. Therefore, if aggregates of the target protein occur, and these aggregates exceed the size of the pores in the virus removal membrane, they may cause clogging of the membrane.

[0013] Furthermore, aggregates are often accompanied by denaturation. Hydrophobic amino acid residues that are folded inside the protein molecule in their native state may be exposed on the surface of the denatured protein. Therefore, it is known that aggregates readily adsorb onto the surface of hydrophobic membrane materials, potentially becoming a standard cause of membrane occlusion.

[0014] To suppress membrane occlusion caused by aggregates, a strategy of treating solutions containing aggregates with an aggregate-removing membrane, for example, in the pre-treatment stage of virus removal membranes, may be effective.

[0015] It is known that if proteins in their native, undenatured, and unaggregated state come into contact with hydrophobic surfaces, they will adsorb onto the membrane surface due to hydrophobic and electrostatic interactions. When this phenomenon is significant, it leads to a decrease in filtration rate due to standard membrane occlusion. It is believed that protein adsorption on the membrane surface is affected by the pH and salt content of the solution. For example, when the protein and membrane surface have opposite charges, adsorption occurs due to electrostatic interactions. However, if an appropriate amount of salt is present in the solution, the surface charges are masked, and adsorption is mitigated. Thus, even in cases of standard occlusion, adjusting the pH and salt concentration in the filtration process using virus removal membranes may be effective.

[0016] Patent Document 1 discloses a method for preventing clogging of a virus removal membrane by pre-treating a solution filtered through a virus removal membrane using a porous polyamide-containing molded body to remove biopolymer aggregates such as protein polymers from the solution. However, Patent Document 1 does not disclose or imply that clogging can occur through complete blockage, standard blockage, or cake filtration. Furthermore, Patent Document 1 does not disclose or imply a method for pre-confirming the presence or absence of biopolymer aggregates in the solution. If biopolymer aggregates are not present in the solution, and the cause of virus removal membrane clogging is not due to biopolymer aggregates, pre-treatment of the solution with a porous polyamide-containing molded body may not resolve the clogging issue.

[0017] Patent Document 2 discloses a method for capturing the positions of particles of various sizes within a virus removal membrane using colloidal gold of different sizes. However, the method in Patent Document 2 only determines the pore size of a specific virus removal membrane, and identifies the captured colloidal gold within the membrane after filtering colloidal gold of known sizes. Therefore, Patent Document 2 does not disclose or imply a method for determining the cause of blockage when a protein-containing liquid with unknown aggregates (such as the presence, amount, size, and shape of aggregates) is filtered through a virus removal membrane and the membrane becomes clogged.

[0018] Furthermore, if aggregates remain in the final formulation, there is a risk of side effects on patients. Therefore, regulatory authorities in various countries have published guidelines for their analysis. As illustrated in Non-Patent Literature 2, size exclusion chromatography is considered the primary method for aggregate analysis in the pharmaceutical industry, but its quantitation limit is around several μg / mL. However, even small amounts of aggregates can clog virus removal membranes. Analytical ultracentrifugation is a recently used technique, and its principle for protein detection is based on UV absorbance, similar to size exclusion chromatography. Therefore, the quantitation limit of analytical ultracentrifugation is considered to be similar to that of size exclusion chromatography. Thus, in the manufacturing process of biopharmaceuticals, it is not easy to confirm the presence of protein aggregates in protein-containing solutions, especially in the stage intended for virus removal.

[0019] Therefore, determining the cause of clogging in conventional filter membranes has been difficult. Thus, one object of the present invention is to provide a method for determining whether the cause of clogging in a filter membrane during the filtration of a protein-containing solution is aggregates.

[0020] Solution for solving the problem

[0021] The inventors believe that knowing whether the cause of membrane blockage is aggregates provides useful information for improving the process, and have conducted research on this. Specifically, the inventors have found that if the cause of blockage is aggregates, the problem can be solved, for example, by removing the aggregates beforehand. Furthermore, the inventors have found that if the cause of blockage is not aggregates, the problem can be solved by other means. If the cause of membrane blockage is unclear, various blockage countermeasures need to be implemented, and their respective inhibitory effects need to be verified. However, if the cause of membrane blockage is clear, the necessary countermeasures and their effects can be predicted.

[0022] According to the present invention, a method for evaluating the clogging of a filter membrane for a protein-containing solution is provided, comprising the following steps:

[0023] Step a) involves passing the protein-containing solution through a filter membrane;

[0024] Step b), after step a), obtain the cross-section of the filter membrane from the filter membrane;

[0025] Step c), using at least one staining agent specific to protein aggregates, to treat the protein-containing solution before step a), or to treat the filter membrane before step b), or to treat the filter membrane cross-section after step b); and

[0026] Step d) confirms the presence of protein aggregates in the cross-section of the filter membrane.

[0027] According to the present invention, a method for evaluating the clogging of a filter membrane for a protein-containing solution is provided, comprising the following steps:

[0028] Step a) involves passing the protein-containing solution through a filter membrane;

[0029] Step a-1): In step a), confirm the blockage of the filter membrane;

[0030] Step b), after step a), obtain the cross-section of the filter membrane from the filter membrane;

[0031] Step c), using at least one staining agent specific to protein aggregates, to treat the protein-containing solution before step a), or to treat the filter membrane before step b), or to treat the filter membrane cross-section after step b); and

[0032] Step d) confirms the presence of aggregates of the aforementioned proteins in the cross-section of the filter membrane.

[0033] According to the present invention, a method for evaluating the clogging of a filter membrane for a protein-containing solution is provided, comprising the following steps:

[0034] Step a) involves passing the protein-containing solution through a filter membrane;

[0035] Step a-2), measure the filtration rate in step a);

[0036] In step b'), when the filtration speed is reduced, the cross-section of the filter membrane is obtained from the filter membrane.

[0037] Step c') involves adding at least one staining agent specific for protein aggregates to the filter membrane; and

[0038] Step d) confirms the presence of protein aggregates in the cross-section of the filter membrane.

[0039] According to a method of the present invention, a method for confirming the presence of protein aggregates in a protein-containing solution is provided, comprising the following steps:

[0040] Step a) involves passing the protein-containing solution through a filter membrane;

[0041] Step b), after step a), obtain the cross-section of the filter membrane from the filter membrane;

[0042] Step c), using at least one staining agent specific to protein aggregates, to treat the protein-containing solution before step a), or to treat the filter membrane before step b), or to treat the filter membrane cross-section after step b); and

[0043] Step d) confirms the presence of protein aggregates in the cross-section of the filter membrane.

[0044] In the above method, a step of washing the filter membrane may be included after step a).

[0045] In the above method, a step of washing the filter membrane may be included after step c).

[0046] In the above method, a step of washing the filter membrane may be included after step c').

[0047] The above method may also include a step of adding a staining agent to the filter membrane to stain both non-aggregated and aggregated proteins.

[0048] The above method may also include a step of fixing the cross-section of the filter membrane.

[0049] In the above method, the cross-section of the filter membrane can be the cross-section of a slice cut from the filter membrane.

[0050] In the above methods, the staining agent that is specific to aggregates may contain a substance that specifically binds to the aggregation portion of protein aggregates and emits fluorescence.

[0051] In the above methods, the staining agent that is specific to aggregates can be at least one selected from Nile Red, Thiosulfate T, PROTEOSTAT and their derivatives.

[0052] In the above method, the staining agent that is specific to the aggregate can be a staining agent containing a binding substance that specifically binds to the aggregate portion of the protein aggregate, and the binding substance is labeled with a fluorescent or luminescent substance.

[0053] In the above method, the binding substance that specifically binds to the aggregate portion of the protein aggregate can be at least one selected from antibodies, antibody fragments, and peptides.

[0054] In the above methods, the staining agent that is specific to aggregates may contain:

[0055] (A) A primary binding substance that specifically binds to the aggregation sites of protein aggregates; and

[0056] (B) A secondary binding substance that specifically binds to a primary binding substance and contains a detectable substance.

[0057] In the above method, the primary binding substance can be at least one selected from antibodies, antibody fragments, and peptides.

[0058] In the above method, the secondary binding substance can be an antibody labeled with a fluorescent substance.

[0059] In the above method, the secondary binding substance is an enzyme-modified antibody. After step c), the method may further include adding a substrate that becomes a fluorescent or luminescent substance when decomposed by the enzyme to the filter membrane and detecting fluorescence or luminescence.

[0060] In the above method, the secondary binding substance is an enzyme-modified antibody. After step c'), the method may further include adding the substrate, which becomes a fluorescent or luminescent substance when decomposed by the enzyme, to the filter membrane to detect fluorescence or luminescence.

[0061] In the above method, the enzyme can be a peroxidase or an alkaline phosphatase.

[0062] In the above method, the filter membrane can be a virus removal membrane.

[0063] In the above method, process c) can be performed before process b).

[0064] In the above method, process c' can be performed before process b').

[0065] In the above method, after step c), a step of removing unreacted staining agent that is specific to the aggregate may also be included.

[0066] In the above method, after step c'), a step of removing unreacted staining agent that is specific to the aggregate may also be included.

[0067] In the above methods, unreacted staining agents specific to aggregates can be removed by dialysis, ultrafiltration, or gel filtration.

[0068] In the above method, in step a-1), the clogging of the filter membrane in step a) can be confirmed by measuring the filtration rate.

[0069] In the above method, in step a-1), the blockage of the filter membrane in step a) can be confirmed by measuring the pressure.

[0070] The effects of the invention

[0071] According to the present invention, a method is provided to determine whether the cause of clogging of the filter membrane during the filtration of a protein-containing solution is aggregates. Attached Figure Description

[0072] Figure 1 A graph representing the results obtained from measuring the dynamic light scattering of a solution of artificially adjusted protein aggregates.

[0073] Figure 2 A graph representing the size exclusion chromatography results of a solution of artificially adjusted protein aggregates.

[0074] Figure 3 This graph illustrates the relationship between the volume of solution processed and the filtration rate when filtering solutions containing and without protein aggregates using a virus removal filter.

[0075] Figure 4 Fluorescence microscopy images showing unused virus removal filters not stained with a staining agent specific to aggregates.

[0076] Figure 5 This is a fluorescence microscopy image showing a virus removal filter that has been filtered through a solution of globulins that do not contain aggregates, and is not stained with a staining agent that is specific to aggregates.

[0077] Figure 6 A fluorescence microscopy image showing a virus removal filter stained with a staining agent specific to aggregates after filtering a solution containing aggregates.

[0078] Figure 7 for Figure 4 The image shown is a linear profile of fluorescence intensity along the film thickness in a fluorescence microscope observation image. The horizontal axis represents the relative position with the inner surface set to 0 and the outer surface set to 1, and the vertical axis represents the brightness at that relative position.

[0079] Figure 8 for Figure 5 The image shown is a linear profile of fluorescence intensity along the film thickness in a fluorescence microscope observation image. The horizontal axis represents the relative position with the inner surface set to 0 and the outer surface set to 1, and the vertical axis represents the brightness at that relative position.

[0080] Figure 9 for Figure 6 The image shown is a linear profile of fluorescence intensity along the film thickness in a fluorescence microscope observation image. The horizontal axis represents the relative position with the inner surface set to 0 and the outer surface set to 1, and the vertical axis represents the brightness at that relative position.

[0081] Figure 10The image shows a fluorescence microscope image of a virus removal filter stained with a staining agent specific to the aggregates, as described in Example 1-1.

[0082] Figure 11 Fluorescence microscopy images of virus removal filters stained with a staining agent specific to the aggregates, as shown in Examples 1-2.

[0083] Figure 12 for Figure 10 The image shown is a linear profile of fluorescence intensity along the film thickness in a fluorescence microscope observation image. The horizontal axis represents the distance from the inner surface, and the vertical axis represents the brightness.

[0084] Figure 13 for Figure 11 The image shown is a linear profile of fluorescence intensity along the film thickness in a fluorescence microscope observation image. The horizontal axis represents the distance from the inner surface, and the vertical axis represents the brightness.

[0085] Figure 14 The image shows a fluorescence microscope image of a virus removal filter stained with a staining agent specific to the aggregates, as described in Example 2.

[0086] Figure 15 for Figure 14 The image shown is a linear profile of fluorescence intensity along the film thickness in a fluorescence microscope observation image. The horizontal axis represents the distance from the inner surface, and the vertical axis represents the brightness.

[0087] Figure 16 Fluorescence microscopy images of the virus removal filter stained with fluorescently labeled antiglobulin antibody, as shown in Comparative Example 1-1.

[0088] Figure 17 Fluorescence microscopy images of the virus removal filters stained with fluorescently labeled antiglobulin antibodies, used in Comparative Examples 1-2.

[0089] Figure 18 for Figure 16 The image shown is a linear profile of fluorescence intensity along the film thickness in a fluorescence microscope observation image. The horizontal axis represents the distance from the inner surface, and the vertical axis represents the brightness.

[0090] Figure 19 for Figure 17 The image shown is a linear profile of fluorescence intensity along the film thickness in a fluorescence microscope observation image. The horizontal axis represents the distance from the inner surface, and the vertical axis represents the brightness.

[0091] Figure 20 Fluorescence microscopy images of the virus removal filter stained with fluorescently labeled antiglobulin antibody, as shown in Comparative Example 1-1.

[0092] Figure 21 for Figure 20The image shown is a linear profile of fluorescence intensity along the film thickness in a fluorescence microscope observation image. The horizontal axis represents the distance from the inner surface, and the vertical axis represents the brightness. Detailed Implementation

[0093] The embodiments of the present invention will be described below. In the following description of the drawings, the same or similar parts are indicated by the same or similar reference numerals. The drawings are schematic diagrams and do not accurately show specific dimensions, etc. Therefore, specific dimensions, etc., should be determined by referring to the following description. Of course, the drawings also include parts with different dimensional relationships and ratios.

[0094] The method for evaluating membrane clogging caused by protein-containing solutions in this embodiment includes the following steps:

[0095] Step a) involves passing the protein-containing solution through a filter membrane;

[0096] Step b), after step a), obtain the cross-section of the filter membrane from the filter membrane;

[0097] Step c), treating the protein-containing solution prior to step a), using at least one staining agent specific to protein aggregates, or treating the filter membrane prior to step b), and treating the filter membrane cross-section after step b); and

[0098] Step d) confirms the presence of aggregates of the aforementioned proteins in the cross-section of the filter membrane.

[0099] The order of procedures listed here is not particularly limited unless otherwise specified, but it is preferable to perform them in the order described herein.

[0100] The method for addressing the specific cause of membrane clogging due to a protein-containing solution in this embodiment includes the following steps:

[0101] Step a) involves passing the protein-containing solution through a filter membrane;

[0102] Step a-2), measure the filtration rate in step a);

[0103] In step b'), when the filtration speed is reduced, the cross-section of the filter membrane is obtained from the filter membrane.

[0104] Step c') involves adding at least one staining agent specific for protein aggregates to the filter membrane; and

[0105] Step d) confirms the presence of protein aggregates in the cross-section of the filter membrane.

[0106] The order of the procedures listed here is not particularly limited, but it is preferable to perform them in the order described herein.

[0107] In this embodiment, the protein is not particularly limited, and examples include albumin, globulin, or fibrinogen, with antibody proteins being more preferred. As an example of a physiologically active substance, an antibody protein, as generally defined in biochemistry, is a glycoprotein molecule (also called gamma globulin or immunoglobulin) produced by B lymphocytes, the infection defense mechanism of vertebrates. For example, the purified antibody protein in this embodiment, used as a human drug, has a structure substantially identical to that of antibody proteins in the body of a human being to whom the drug is administered.

[0108] Antibody proteins can be human antibodies or antibodies derived from mammals other than humans, such as cattle and mice. Alternatively, antibody proteins can be chimeric antibodies with human IgG or humanized antibody proteins. Chimeric antibody proteins with human IgG refer to antibody proteins whose variable regions are derived from organisms other than mice, while other constant regions are replaced with immunoglobulins derived from humans. Humanized antibody proteins, on the other hand, refer to antibody proteins in which the complementarity-determining region (CDR) of the variable region is derived from an organism other than humans, while other framework regions (FR) are derived from humans. Humanized antibody proteins exhibit further reduced immunogenicity compared to chimeric antibody proteins.

[0109] The categories (isotypes) and subclasses of antibody proteins are not particularly limited. For example, antibody proteins are classified into five categories—IgG, IgA, IgM, IgD, and IgE—based on differences in the structure of their constant regions. However, the filtering method of this embodiment is designed to purify antibody proteins of any of these five categories. Furthermore, among human antibody proteins, IgG has four subclasses—IgG1 to IgG4—and IgA has two subclasses—IgA1 and IgA2. However, the filtering method of this embodiment is designed to purify antibody proteins of any subclass. It should be noted that antibody-associated proteins, such as Fc fusion proteins formed by proteins binding to the Fc region, also exist.

[0110] Furthermore, antibody proteins can also be classified according to their origin. However, the antibody protein that the filter membrane in this embodiment is intended to purify can be any one of natural human antibody proteins, recombinant human antibody proteins manufactured using gene recombination technology, monoclonal antibody proteins, and polyclonal antibody proteins. Among these antibody proteins, from the viewpoint of the need and importance of the filtration method in this embodiment, human IgG and monoclonal antibodies are preferred for the antibody protein that is intended to be purified, but it is not limited to them.

[0111] In this embodiment, the protein-containing solution refers to a solution containing proteins like those in the examples above. During the manufacturing and purification of these proteins, the solution contains a large number of various impurities, so membrane separation technology is used for purification.

[0112] In this embodiment, the protein-containing solution is not particularly limited in that the protein is dissolved in the solution. The type of buffer solution that can be used is not particularly limited; for example, it can be a solution containing tris, acetic acid, histidine, glycine, phosphoric acid, or citric acid.

[0113] In order to stabilize proteins, solutions containing proteins may contain surfactants such as Tween, sugars such as sorbitol, maltose, sucrose and trehalose, inorganic salts such as sodium chloride and sulfonates, and amino acids such as arginine and lysine.

[0114] In this embodiment, the concentration of the protein-containing solution is not particularly limited if the protein is dissolved in the solution. Examples of lower limits for the concentration of the protein-containing solution include 0.01 mg / mL or more; other examples include 0.05 mg / mL or more; other examples include 0.1 mg / mL or more; other examples include 0.5 mg / mL or more; other examples include 1.0 mg / mL or more; and further, other examples include 5.0 mg / mL or more. Examples of upper limits for the concentration of the protein-containing solution include 200 mg / mL or less; other examples include 150 mg / mL or less; other examples include 100 mg / mL or less; other examples include 50 mg / mL or less; and other examples include 25 mg / mL or less.

[0115] In this embodiment, the concentration of the buffer solution is not particularly limited as long as the aforementioned dissolved substance dissolves. Examples of lower limits for the buffer solution concentration include 0.1 mmol / L or higher; other examples include 1 mmol / L or higher; other examples include 10 mmol / L or higher; other examples include 50 mmol / L or higher; other examples include 100 mmol / L or higher; and other examples include 300 mmol / L or higher. On the other hand, sometimes the buffer solution is not used.

[0116] Furthermore, the concentration of the stabilizer added to the protein-containing solution can be determined by using the protein stability of the protein solution as an indicator. Preferably, it is 500 mmol / L or less, or 300 mmol / L or less, and more preferably 250 mmol / L or less. If the protein is sufficiently stable, the stabilizer may not be required.

[0117] In this embodiment, the pH of the protein-containing solution or buffer solution is not particularly limited. Examples of lower pH values ​​include 3.5 and above; other examples include 4.5 and above; further examples include 5.0 and above; further examples include 5.5 and above; further examples include 6.0 and above. Examples of upper pH values ​​include 10.0 and below; other examples include 9.0 and below; further examples include 8.0 and below; further examples include 8.5 and below; further examples include 8.0 and below; further examples include 7.5 and below; further examples include 7.0 and below. The method for measuring pH is not particularly limited, and methods using hydrogen electrodes, hydroquinone electrodes, antimony electrodes, and glass electrodes are examples. Methods using glass electrodes are preferred.

[0118] In this embodiment, the filter membrane is not particularly limited to the filter membrane used in the process of purifying proteins from the protein-containing solution described above. The filter membrane separates impurities in various sizes. In membrane separation technology, a protein-containing solution is passed through the membrane from one surface to the other, capturing impurities inside the membrane. Additionally, the protein, as the target substance, is passed through together with the liquid, which serves as the medium, thereby removing impurities such as viruses. The shape of the filter membrane is not particularly limited; hollow fiber membranes and flat membranes are examples, with hollow fiber membranes being a preferred example. Alternatively, flat membranes are also a preferred example. The raw material of the filter membrane is not particularly limited; it can be a non-synthetic polymer filter membrane such as regenerated cellulose, or a filter membrane formed from synthetic polymers. Examples of raw materials for the filter membrane include polyethersulfone, polysulfone, polyvinylidene fluoride, cellulose, cellulose derivatives, or mixtures thereof. These raw materials can be layered in the filter membrane, or other raw materials can be coated onto one raw material. Furthermore, the surface of the filter membrane can be modified using graft polymerization or coating techniques.

[0119] Filter membranes are typically shaped to allow protein-containing solutions to pass through from one surface to the other. Hollow fiber membranes are formed, for example, by bundling multiple membranes together and placing them inside a cylindrical shell called a sheath, with both ends reinforced using adhesives such as polyurethane, leaving the hollow fiber ends open. Flat membranes are formed, for example, by clamping the membrane between a component having an inlet for the protein-containing solution and a component having an outlet, with the ends of the two components fused together. Filter membranes shaped into appropriate forms, such as folded membranes or stacked membranes, can also be used as filters.

[0120] In this embodiment, the filter membrane is not particularly limited. However, from the viewpoint that the size of the protein, which is the target substance for purification, is very close to the size of the pores of the filter membrane, and that the membrane is easily affected by aggregates, the membrane used in the virus removal filter is preferred. The membrane used in the virus removal filter is referred to as a virus removal membrane.

[0121] In this embodiment, the virus removal membrane is not particularly limited to any membrane that can substantially separate the virus from the target protein. For example, a membrane with a nominal pore size of approximately 15 nm or more and approximately 75 nm or less is suitable. For instance, a membrane with a nominal pore size of 20 nm allows proteins of 150 kDa, such as monoclonal antibodies, to pass through, but not, for example, parvoviruses. On the other hand, if a membrane with a nominal pore size of 35 nm or more is used, larger proteins, such as fibrinogen of 340 kDa, can pass through, and for example, simian virus 40 can be removed.

[0122] In this embodiment, the virus removal membrane has, for example, a molecular weight cutoff of 100 kD or more and 1000 kD or less. Relatedly, the membrane's molecular weight cutoff (MWCO) refers to the nominal molecular weight of molecules and particles that can pass through the membrane at a rate of 90%.

[0123] In this embodiment, the raw material for the virus removal membrane is not particularly limited to any material that can remove viruses, and examples include polyethersulfone, polysulfone, polyvinylidene fluoride, cellulose, cellulose derivatives, or mixtures thereof.

[0124] a) The process of passing a protein-containing solution through a filtration membrane

[0125] In this embodiment, the specific method of passing the protein-containing solution through the filter membrane is not particularly limited. Examples include, for instance, pressing the protein-containing solution through one surface of the filter membrane, or using a pump to pass the protein-containing solution through the filter membrane. Alternatively, dead-end filtration can be used, allowing the total amount of protein-containing solution sent to the filter membrane to pass through. Tangential flow filtration, where a portion passes through and the remainder returns to the original solution, is also possible. Protein purification processes often involve multiple stages and can be connected to processes before and after the filtration membrane for continuous processing.

[0126] Before and after passing the protein-containing solution through the filter membrane, a buffer solution can be passed through the filter membrane separately from the protein-containing solution. This buffer solution can be of the same properties as the buffer solution constituting the protein-containing solution, or it can be of different properties; preferably, it is a buffer solution of the same properties. Examples of buffer solutions are described above. When recovering the protein-containing solution, all the protein-containing solution that has passed through the filter membrane can be recovered. Alternatively, multiple filters can be arranged side-by-side for filtration, or filtration can be performed while changing the filter membrane midway through the process.

[0127] There is no particular limit to the amount of protein-containing solution that passes through the filtration membrane; however, a lower limit of 10 L / m³ can be cited as an example. 2 As an upper limit, 1000 L / m can be cited as an example. 2 .

[0128] As described above, the method of passing a protein-containing solution through a filter membrane can be either pressure filtration or filtration using a pump or similar device to deliver the protein solution to the filter membrane. When applying pressure, constant-pressure filtration, where a constant pressure is always applied, is preferred, but the pressure can be varied depending on the situation. The intermembrane differential pressure can be determined by factors such as the time required for filtration, and it is preferable to operate below the recommended upper limit pressure for the filter membrane used. Exceeding the upper limit pressure can deform the filter membrane structure due to the pressure, potentially allowing impurities such as viruses that should be removed to pass through. When using a pump or similar device for filtration, constant-rate filtration, where the intermembrane differential pressure applied to the filter membrane is constant, is preferred, but the pump flow rate can also be adjusted to maintain a constant intermembrane differential pressure. The pump's set flow rate is not limited, similar to the pressure filtration method, but the intermembrane differential pressure is preferably set in a way that does not exceed the recommended upper limit pressure of the filter membrane.

[0129] a-1) The process of confirming filter membrane blockage in step a).

[0130] The method of this embodiment can include the step of confirming filter membrane blockage in step a). To determine whether the filter membrane is blocked, the filtration rate when a protein-containing solution is passed through the filter membrane can be measured. An example of an indicator of filtration rate is the volume of liquid passing through the filter membrane per unit time, expressed in L / m³. 2 / hour (hour). The filtration rate can also be calculated by dividing by the applied pressure to obtain the filtration rate per unit pressure. There are no particular limitations on the method for measuring the filtration rate. For example, when a protein-containing solution is pumped at a constant pressure, a method can be used to measure the volume of liquid passing through per unit time. The volume can be measured using a flow meter, or by measuring the weight or volume of liquid collected at constant intervals. In the case of dead-end filtration, the introduced flow rate and the passing volume are equal, so the introduced flow rate can be measured. Alternatively, when a constant flow rate is introduced into the filter membrane using a pump, the differential pressure between the two surfaces of the filter membrane (membrane differential pressure) can be measured. If the filter membrane becomes clogged, a higher intermembrane differential pressure is required for the same flow rate; therefore, converting this to the above indicators confirms a decrease in filtration rate. There are no particular limitations on the threshold filtration rate for determining membrane clogging; a filtration rate that is practically impossible to filter at is not defined. For example, it could be 70% or less compared to the initial filtration rate, or 50% or less in other cases. For example, if the measured filtration rate is below the threshold filtration rate, the filter membrane is considered clogged.

[0131] In addition, to determine whether the filter membrane has become clogged, the filtration pressure when a protein-containing solution is passed through the filter membrane can be measured. An increase in filtration pressure can be confirmed by measuring the inter-membrane pressure difference as described above.

[0132] a-2) The process of determining the filtration rate in step a).

[0133] The method of this embodiment can include a step of measuring the filtration rate in step a). To determine whether the filter membrane is clogged, the filtration rate is measured when a protein-containing solution is passed through the filter membrane. An example of an indicator of filtration rate is the volume of liquid passing through the filter membrane per unit time, expressed in L / m³. 2 / hour. The filtration rate can also be calculated by dividing by the applied pressure to obtain the filtration rate per unit pressure. There are no particular limitations on the method for measuring the filtration rate. For example, when a protein-containing solution is pumped at a constant pressure, a method can be used to measure the volume of liquid passing through per unit time. The volume can be measured using a flow meter, or by measuring the weight or volume of liquid collected at constant intervals. In the case of dead-end filtration, the introduced flow rate and the passing volume are equal, so the introduced flow rate can be measured. Alternatively, when a constant flow rate is introduced into the filter membrane using a pump, the differential pressure between the two surfaces of the filter membrane (membrane differential pressure) can be measured. If the filter membrane becomes clogged, a higher intermembrane differential pressure is required for the same flow rate; therefore, converting this to the above indicators confirms a decrease in filtration rate. There are no particular limitations on the threshold filtration rate for determining membrane clogging; a filtration rate that is practically impossible to filter at is not defined. For example, it could be 70% or less compared to the initial filtration rate, or 50% or less in other cases. For example, if the measured filtration rate is below the threshold filtration rate, the filter membrane is considered clogged.

[0134] b) The process of obtaining the filter membrane cross-section from the filter membrane after process a).

[0135] After step a), to confirm whether the filter membrane blockage is due to aggregates, the blocked filter membrane is recovered, and the presence or absence of aggregates inside the filter membrane is confirmed. By observing the inside of the filter membrane, a fragment of the filter membrane can be obtained, and the cross-section of the filter membrane can be observed. To cut the filter membrane fragment, a razor or other tool can be used to cut the filter membrane; there are no particular limitations, but a sharp tool is preferred to maintain the cross-sectional structure as much as possible, or to preserve the state of the particles remaining inside. If the filter membrane is a hollow fiber, it can be cut perpendicular to the fiber length direction to obtain a fragment of the filter membrane. If the filter membrane is a flat membrane, it can be cut perpendicular to the membrane surface to obtain a fragment of the filter membrane. If the filter membrane is difficult to cut due to its very small thickness, it can be embedded in an embedding agent such as alkanes or resin, and then sliced ​​using a slicer or similar instrument. The thickness of the slice is preferably 2 μm or more and 20 μm or less, more preferably 4 μm or more and 10 μm or less. Alternatively, liquid nitrogen or similar substances can be used to cool the filter membrane to extremely low temperatures, and then the filter membrane can be cut to obtain fragments of the filter membrane.

[0136] b') The process of obtaining the cross-section of the filter membrane from the filter membrane when the filtration rate decreases.

[0137] If the filtration speed decreases in step a-2), indicating membrane clogging, the clogged membrane is recovered to confirm whether the clogging is due to aggregates, and the presence of aggregates within the membrane is checked. Observing the interior of the membrane allows for the acquisition of membrane fragments and the observation of its cross-section. To cut out membrane fragments, a razor or similar tool can be used; there are no particular limitations, but a sharp tool is preferred to maintain the cross-sectional structure and the state of any remaining particles. If the membrane is a hollow fiber, it can be cut perpendicular to the fiber length to obtain membrane fragments. If the membrane is a flat membrane, it can be cut perpendicular to the membrane surface to obtain membrane fragments. If the membrane is very thin and difficult to cut, it can be embedded in an encapsulating agent such as alkanes or resin, and then sliced ​​using a slicer or similar tool. The thickness of the slice is preferably 2 μm or more and 20 μm or less, more preferably 4 μm or more and 10 μm or less. Alternatively, the filter membrane can be cooled to an extremely low temperature using liquid nitrogen or the like, and then the filter membrane can be cut to obtain a fragment of the filter membrane.

[0138] c) A process of treating the protein-containing solution before step a), the filter membrane before step b), or the cross-section of the filter membrane after step b) with at least one staining agent specific to protein aggregates.

[0139] To confirm the presence of aggregates within the filter membrane, the following staining treatment is performed. For example, before step a), the protein aggregates in the solution can be stained by adding a staining agent specific to protein aggregates to the protein-containing solution. An appropriate amount of staining agent can be added to the protein-containing solution before filtration using the filter membrane. Alternatively, before step b), the protein aggregates captured by the filter membrane can be stained by pre-adding a staining agent specific to protein aggregates to the filter membrane. Or, after step b), the protein aggregates captured by the filter membrane can be stained by adding a staining agent specific to protein aggregates to the filter membrane cross-section. Adding staining agent to the filter membrane cross-section simultaneously during the fragmentation of the filter membrane also includes treating the filter membrane cross-section with staining agent after step b).

[0140] c') The step of adding at least one staining agent specific to protein aggregates to the filter membrane.

[0141] To confirm the presence of aggregates within the filter membrane, the filter membrane is stained using a staining agent specific to protein aggregates. Step c') is performed, for example, after step b'). In this case, the cross-section of the filter membrane fragment obtained in step b') is stained using the aggregate-specific staining agent, and the staining status is observed. Alternatively, step c') can be performed before step b'). In this case, an aggregate-specific staining agent is added to the filter membrane prior to obtaining the fragment. Then, step b') is performed to obtain a filter membrane fragment, and the staining status of the filter membrane cross-section is observed.

[0142] In step c) or c'), a staining agent specific to protein aggregates will specifically stain the protein aggregates, but will not stain proteins that are not aggregated. Examples of such staining agents include common reagents such as thiosulfate T and Nile Red, or the detection reagent contained in the PROTEOSTAT aggregate assay kit sold by ENZO LIFESCIENCES. For example, protein aggregate specific staining agents described in WO03 / 000853 or WO2011 / 065980 can also be used. These staining agents are compounds that recognize and bind to specific hydrophobic regions on the surface of aggregates as aggregation sites, emitting fluorescence. The amount of staining agent used and the conditions vary depending on the reactivity of each product, and also on the amount of aggregates contained in the filter membrane. Preferably, the amount of each staining agent is an amount that forms a large excess of aggregates and is an amount that will not cause problems with background fluorescence. Preferably, the amount and conditions of the staining agent are set in a manner that produces a clear difference between the signal intensity when staining a negative control sample that does not contain aggregates and the signal intensity when staining a positive control sample that contains aggregates. A clear difference refers to confirming that the brightness of the obtained image is different in a manner that the signal intensity of the positive control sample is, for example, 5 times or more, preferably 10 times or more, than the signal intensity of the negative control sample, but there is no particular limitation.

[0143] Other examples of staining agents specific to aggregates include antibodies, antibody fragments, or peptides that have specific affinity for aggregates and bind to them. For example, protein-specific staining agents described in Anal. Chem. 2016, 88, 10095-10101 can also be used. Antibodies, antibody fragments, and peptides do not emit signals, so they can be directly labeled using fluorescent substances. Alternatively, secondary binding substances such as fluorescently labeled antibodies that specifically bind to primary binding substances such as antibodies, antibody fragments, or peptides with specific affinity for aggregates can be used. Furthermore, other examples of staining agents specific to aggregates include antibodies labeled with enzymes such as alkaline phosphatase or peroxidase. This enzyme can be added to the cross-section of the filter membrane to react with the aggregates, and then a substance that emits a signal through the enzyme reaction can be added to obtain a signal. When using any staining agent, it is preferable to observe the cross-section of the filter membrane after washing away excess staining agent, as no excess signal is detected. In addition, to suppress non-specific reactions between the dye and the filter membrane cross-section, a blocking agent is used to cover the exposed filter membrane cross-section before staining. This inhibits the adsorption of dye onto the filter membrane cross-section, thus preventing non-specific staining of the filter membrane. Any substance can be used as long as it can adsorb onto the filter membrane cross-section to cover it and does not react with the dye. Examples include proteins such as BSA, serum, and skim milk powder; protein-free products such as BlockPro and Stabilblot; and blocking agents combining both, such as BlokingOne. In particular, non-protein blocking agents, developed for sealing membranes used in protein blotting, are sometimes suitable for sealing filter membranes that are also polymeric membranes.

[0144] To improve the accuracy of staining specific to aggregates, a process can be performed to wash the filtered membrane and remove excess proteins from it. Examples of washing solutions include protein-free water and various buffer solutions. Since the membrane's state can be maintained and adsorbates can be retained on it, a buffer solution with the same solvent as the protein-containing solution is preferred as the washing solution. The membrane can be washed by filtering with the washing solution, by immersing the membrane in the washing solution, or by directly pouring the washing solution onto a cross-section of the membrane. The method of washing the membrane is not particularly limited; since the washing solution can be passed through the pores of the membrane, filtration washing is preferred.

[0145] To improve the accuracy of staining aggregates specifically, the filtered membrane can be fixed to retain residual proteins and particles. Fixation can be performed using techniques commonly used in biological experiments, such as methods using aldehydes like formaldehyde or glutaraldehyde, or methods using organic solvents like ethanol or acetone. When performing fixation, it is preferable to wash the fixative in a manner that does not affect subsequent staining. Furthermore, when using aldehyde fixation, activation treatment can be performed before staining to moderate the fixation and improve reactivity with the staining agent. The solvents used for fixation, subsequent washing, and activation are not limited; common buffers such as phosphate-buffered saline (PBS) are preferred. Additionally, to improve washing efficiency, surfactants such as Tween 20 or Tween 80 can be added in trace amounts to the buffer.

[0146] d) The process of confirming the presence of aggregates in the filter membrane fragment.

[0147] If a staining agent specific to aggregates is used for treatment, and a staining signal is confirmed at the observation section of the filter membrane, then aggregates are present in the filter membrane. Therefore, it can be determined that the protein-containing solution contains protein aggregates, and the resulting blockage is likely the cause of the blockage. Conversely, if no staining signal is confirmed at the observation section of the filter membrane, it can be determined that there are no protein aggregates in the filter membrane, and no blockage caused by aggregates. If blockage caused by aggregates is confirmed, pre-filtration treatment to remove aggregates can be performed at the front end of the filter membrane, or the stability of the target protein can be investigated, and a buffer solution that is less prone to aggregation can be selected. Thus, in the filtration process (especially the virus removal process) of biopharmaceutical manufacturing, choosing countermeasures that do not affect the filter membrane due to aggregates can prevent blockage when filtering protein-containing liquids using the filter membrane. On the other hand, if blockage is confirmed, but the cause is not aggregates, and the target protein adsorbed on the pore walls of the filter membrane is considered the cause, countermeasures such as choosing a buffer solution that inhibits the adsorption of the target protein or changing the raw materials of the filter membrane can be implemented.

[0148] Alternatively, a staining agent specific to aggregates can be added to the filter membrane, and another staining agent capable of staining both aggregated and non-aggregated proteins as a whole, allowing simultaneous observation of the protein and its aggregates. This additional step confirms that, in the absence of aggregates, there are no denatured or aggregated proteins adsorbed onto the filter membrane, narrowing the pores. Staining with a protein-staining agent can be performed before, after, or simultaneously with step c) or c'). It should be noted that the two staining agents are selected in a way that allows for mutual recognition. For example, when using fluorescent staining agents, they are selected with different excitation and fluorescence wavelengths. There are no particular limitations on the staining agent capable of staining the entire protein; examples include antibodies that specifically bind to the target protein and bind with pigment. Furthermore, when a large amount of protein adsorbed onto the filter membrane, fluorescence originating from aromatic amino acids in the adsorbed proteins can be detected.

[0149] Therefore, according to the method of this embodiment, it can be confirmed that aggregates exist in protein-containing liquids, which can improve the efficiency of biopharmaceutical manufacturing processes, for example.

[0150] If filtration is not actually performed, the extent to which the filter membrane is clogged will not be known if aggregates exist in the protein-containing liquid. Therefore, the method of this embodiment can be performed on a smaller scale than intended beforehand, and then filtration can be performed on a larger scale. In the case of prior testing, a protein-containing solution pre-mixed with a staining agent specific to aggregates can be filtered through a filter membrane, and then the cross-section of the filter membrane can be observed. Even if a staining agent specific to aggregates is mixed into the protein-containing solution, a signal cannot be detected at the cross-section of the filter membrane if the amount of aggregates is small. However, if the aggregates accumulate in the filter membrane to the extent of causing filter membrane clogging, a signal can be detected at the cross-section of the filter membrane. By pre-mixing a staining agent specific to aggregates into the protein-containing solution, the staining process after filtration can be saved. Since staining agent that does not react with aggregates may remain in the filter membrane, it is preferable to thoroughly wash the filter membrane. From the viewpoint of preventing non-specific reactions between the staining agent and the filter membrane, it is preferable to add the staining agent to the protein-containing solution, remove the staining agent that does not bind to aggregates from the protein-containing solution using dialysis or the like, and then filter the protein-containing solution.

[0151] Example

[0152] The following describes the embodiments in detail based on examples and comparative examples, but the present invention is not limited to the following embodiments and can be implemented in any manner without departing from the spirit of the present invention.

[0153] 1. Preparation of protein-containing solutions containing aggregates

[0154] To confirm whether protein aggregates can be specifically stained with dyes, protein aggregates were artificially created by acid treatment and used in a model filtration experiment.

[0155] Using a human immunoglobulin preparation (Venoglobulin IH 5% intravenous injection from blood donation, Japan Blood Products Corporation), 200 mL of a solution with a final immunoglobulin concentration of 2% and a sodium chloride concentration of 100 mmol / L was prepared. The pH of this solution was 4.5, and the average diameter of the particles contained in the solution, according to the dynamic light scattering method, was 11.8 nm. The pH of this solution was lowered to 2.5 at room temperature using a 1 mol / L hydrochloric acid aqueous solution, and after standing for 1 hour, it was returned to pH 4.5 using a 1 mol / L sodium hydroxide aqueous solution and allowed to stand for 24 hours, thereby obtaining a solution containing human immunoglobulin aggregates (hereinafter sometimes referred to as "solution A").

[0156] The average diameter of the particles in solution A was determined using dynamic light scattering, and the result was an average diameter of 74.3 nm. The determination was performed using a Malvern ZetaSizer Nano. The results are as follows: Figure 1 As shown. The light intensity (attenuation conditions of the irradiated light), distance from the light source, and total measurement time (cumulative number of measurements and measurement time per cumulative measurement) are automatically set using the device's settings. Additionally, the measurement temperature is 25.0°C and the measurement angle is 173°.

[0157] Solution A was subjected to size exclusion chromatography analysis. The results, calculated from the peak areas of the obtained chromatograms, showed that it contained 16.5% monomers, 5.6% dimers, and 77.8% polymers (trimers or larger). It should be noted that in this embodiment, to prepare a model experiment containing polymers and adding them to a protein solution, the average diameter of the fractionated sample analyzed by size exclusion chromatography was not determined. Size exclusion chromatography was performed using a high-performance liquid chromatograph (Prominence, Shimadzu Corporation) and a column (TSK gel G3000SWXL, TOSOH CORPORATION, size exclusion limit molecular weight: 500,000 Da). The mobile phase contained 0.3 mol / L phosphate buffer (pH 6.9), 0.2 mol / L arginine-HCl, and 0.1 mol / L NaCl. The measurement temperature was 25°C, the measurement time was 20 minutes, and the flow rate was 1.0 mL / min. An example of the measurement results is provided. Figure 2 As shown. Figure 2Peak 1 represents a multimer of globulin (trimer or larger), peak 2 represents a dimer of globulin, and peak 3 represents a monomer. This confirms that solution A is a protein-containing solution containing aggregates. The ratio of each monomer to multimer is calculated from the peak area of ​​the chromatogram and used as the relative area ratio. Similarly, in the following examples, the concentration of monomers and multimers, etc., is expressed using the relative area ratio % (%).

[0158] 2. Confirmation of clogging caused by filtering protein-containing solutions containing aggregates using a filter membrane.

[0159] 2-1 Confirmation of membrane clogging

[0160] Human immunoglobulin preparation (Venoglobulin IH 5% IV, Japan Blood Products Corporation) was used to form a 300 mL solution with a globulin concentration of 1% and a sodium chloride concentration of 100 mmol / L. The solution was diluted with distilled water for injection (Otsuka Pharmaceutical) and sodium chloride was added. This solution was divided into two portions, and 0.29 mL of solution A was added to one portion to form a polymer containing 0.3% or more trimers of globulin. These solutions were then subjected to constant pressure dead-end filtration at 98 kPa using a Planova 20N (registered trademark, manufactured by Asahi Kasei Medical Co., Ltd.) with a hollow fiber membrane used as a virus removal filter. The membrane area of ​​the filter used was 0.001 m². 2 The weight of the filtrate was measured, and the filtration rate was calculated over time. To accurately compare the degree of clogging, the ratio of the initial filtration rate to the filtration rate at each measurement point was determined. The results are as follows: Figure 3 As shown, the sample without added aggregates did not cause clogging. In contrast, the sample with added aggregates caused clogging.

[0161] 2-2 Confirmation of the presence of aggregates in a clogged filter using a staining agent specific to the aggregates.

[0162] The filtered virus-removing filter was washed with water, and the hollow fiber membrane was removed from the filter and fixed by immersion in 4% paraformaldehyde phosphate buffer (FUJIFILM Wako Pure Chemical Corporation). To wash away the paraformaldehyde, the hollow fiber membrane was washed in PBS, cut into pieces approximately 5 mm in diameter, and embedded at -20°C using an OCT complex (Sakura Finetek). Thin sections of the hollow fiber membrane (8 μm) were obtained using a cryostat (Leica, CM1950). After washing the complex with water, the sections were washed with PBS and then blocked by immersion in 2% BSA in PBS-T solution at room temperature for 10 minutes. After washing, the stock solution of the staining agent contained in the PROTEOSTAT protein aggregation assay kit (Enzo Lifesciences) was diluted 1000-fold with the accompanying buffer and added to the hollow fiber membrane sections. After reacting at room temperature for 40 minutes, the sections were washed to remove excess staining agent. Each wash was performed four times using PBS containing 0.05% Tween 20 (hereinafter PBS-T). As a blank experiment, the same treatment was performed using Planova 20N (registered trademark, manufactured by Asahi Kasei Medical Co., Ltd.) without any filtration, and slides were prepared. Observation of the slides was performed using a Leica system (fluorescence microscope DMI8, LASX software). Fluorescence observation used an excitation filter of 472 / 30nm, a fluorescence filter of 598 / 25nm, and a dichroic mirror of 598nm. Exposure time and other observation conditions were adjusted to provide sufficient brightness for samples containing aggregates, and all samples were observed. Microscopic images were captured as 16-bit TIFF files using a CMOS monochrome microscope camera (Leica, DFC9000). The images were acquired using ImageJ (ver 1.52a), and the observation results are presented without brightness contrast adjustment. Figures 4-6 As shown. For filters that do not filter any substances, such as Figure 4 As shown, no fluorescence signal stronger than the background fluorescence was observed. For filters that filtered samples without added aggregates, such as... Figure 5 As shown, no fluorescence signal stronger than the background fluorescence was observed. For filters that filtered samples containing aggregates, such as... Figure 6 As shown, a fluorescence signal stronger than the background fluorescence was observed. Therefore, the method of this embodiment confirms that non-aggregated globulins do not react with the staining agent, and only globulin aggregates can be detected.

[0163] Additionally, using ImageJ, the results of representing the line profiles of brightness from the inner surface to the outer surface at four locations for each image are as follows: Figures 7-9 As shown. For filters that do not filter any substances, such as Figure 7 The brightness is low, as shown. Additionally, for filters that filter samples without added aggregates, such as... Figure 8 The brightness is low, as shown. For filters that filter samples containing aggregates, such as... Figure 9 As shown, the brightness is high, and characteristic peaks appear. The location of the detected peak can be regarded as the region where aggregates are captured. As shown above, even if the amount of aggregates of trimer or higher is as small as 0.3%, the aggregates can be detected by using the method of this embodiment, and the presence of aggregates can be distinguished from the presence of aggregates in the membrane where no aggregates are present.

[0164] [Example 1]

[0165] Using a human immunoglobulin preparation (Venoglobulin IH 5% intravenous injection from blood donation, sourced from the Japan Blood Products Agency), a protein-containing solution of 3% concentration was prepared by diluting it with citrate-phosphate buffer at pH 4.0 to a concentration of 3%. As the citrate-phosphate buffer, a solution prepared by adjusting McIlvaine buffer (10-fold concentration) (manufactured by NACALAI TESQUE, INC.) to a final concentration of 20-fold was used. This solution was divided into two portions: one portion with a sodium chloride concentration of 30 mmol / L (Example 1-1), and the other portion with a sodium chloride concentration of 190 mmol / L (Example 1-2). A membrane with a membrane area of ​​0.001 m² was used. 2 Hollow fiber membranes (Planova 20N, registered trademark, manufactured by Asahi Kasei Medical Co., Ltd.) were used to perform dead-end filtration of these solutions at a constant pressure of 98 kPa for 5 hours. The filtration rate per unit area and per unit time was calculated 10 minutes after the start of filtration, and the ratio of this ratio to the filtration rate in the last 5 minutes (hereinafter referred to as the "filtration rate ratio") is shown in Table 1. For Example 1-1, the filtration rate was slightly lower, and the filtration rate ratio was 0.87. In contrast, for Example 1-2, the filtration rate ratio was 0.71, and the degree of clogging of the hollow fiber membrane was greater.

[0166] The filtered hollow fiber membrane was recovered and its cross-section was stained using the same steps as described in step 2-2 above. Images of the membrane cross-section observed using a 100x oil immersion objective lens are shown below. Figure 10 and Figure 11 As shown. Additionally, the line contour obtained from the image is shown in the diagram. Figure 12 and Figure 13 As shown. Figure 10 and Figure 12 As shown, for Example 1-1, with Figure 5Similarly, only signals considered as background fluorescence were detected, whereas, as Figure 11 As shown, in the fluorescence microscopy images of Examples 1-2, regions emitting strong signals in a striped pattern were observed, particularly near the outer surface. Additionally, as... Figure 13 As shown, in the line profile, a higher intensity can be confirmed in the region near the outer surface compared to Example 1-1. As demonstrated above, the reduced filtration rate in Example 1-2, which resulted in clogging, is due to IgG aggregates generated by the high sodium chloride concentration.

[0167] [Table 1]

[0168]

[0169] [Example 2]

[0170] The pH of the buffer solution was set to 7.0. Otherwise, a protein-containing solution was prepared in the same manner as in Example 1, with sodium chloride added to form a 30 mmol / L solution, as the solution for Example 2. The solution for Example 2 was tested using a membrane with a surface area of ​​0.001 m². 2 Planova 20N (registered trademark, manufactured by Asahi Kasei Medical Co., Ltd.) was used for dead-end filtration at a constant pressure of 98 kPa for 5 hours. The filtration rate of the solution in Example 2, calculated similarly to Example 1, was 0.31, as shown in Table 1, indicating significant clogging of the hollow fiber membrane. The cross-section of the hollow fiber membrane used was stained and observed in the same manner as in Example 1, as shown in Table 1. Figure 14 As shown. Additionally, the line contours obtained from the image are as follows: Figure 15 As shown, no areas with fluorescence intensity stronger than the background fluorescence were observed in the fluorescence image, and no strong signal was detected even after confirming the line outline. This result indicates that the clogging in the filter of Example 2 was not caused by aggregates.

[0171] [Comparative Example 1]

[0172] For each section obtained using the same virus removal filter as used in Examples 1-1 and 1-2 (Comparative Examples 1-1 and 1-2, respectively), staining was performed using a goat polyclonal antibody against human immunoglobulins modified with FITC as a fluorescent agent. This antibody stained both aggregated and non-aggregated proteins. The sections were blocked by immersion in PBS-T solution of 2% BSA at room temperature for 10 minutes. After washing, antibody diluted 100-fold with Dako antibody diluent buffer (Agilent) was added to the sections. The sections were reacted at room temperature under light for 60 minutes, and then washed to remove excess staining agent. Each wash was performed four times with PBS-T. The stained sections were then observed under a fluorescence microscope using the same system as in Example 1, except for an excitation light filter of 480 / 40 nm, a fluorescence filter of 527 / 30 nm, and a dichroic mirror of 505 nm. The results are as follows. Figure 16 and Figure 17 As shown. No major differences were found between Comparative Example 1-1 and Comparative Example 1-2. However, the fluorescence signal of Comparative Example 1-1 was slightly higher than that of Comparative Example 1-2, suggesting the possibility of a higher amount of globulin. Figure 18 and Figure 19 They represent Figure 16 and Figure 17 The images shown depict the fluorescence intensity profiles along the film thickness direction. No difference in fluorescence intensity was found between the two. This illustrates why specific film blockage cannot be addressed when using staining agents that are not specific to aggregates.

[0173] [Comparative Example 2]

[0174] The slides obtained using the same filter as the virus removal filter used in Example 2 (Comparative Example 2) were stained and observed using the same procedure as in Comparative Example 1. The resulting images are shown below. Figure 20 As shown. Additionally, by Figure 20 The image obtained shows the contour of the fluorescence intensity line along the film thickness direction, as shown in the image. Figure 21 As shown. This observation detects fluorescence along the entire membrane thickness direction, staining the protein as a whole; therefore, this result cannot be used to determine the presence of aggregates.

Claims

1. A method for evaluating the clogging of a filter membrane for a protein-containing solution, comprising the following steps: Step a) involves passing the protein-containing solution through a filter membrane; Step b), after step a), obtaining the filter membrane cross-section from the filter membrane; Step c) involves treating the protein-containing solution before step a) using at least one staining agent that is specific to the aggregates of the protein, or treating the filter membrane before step b), or treating the cross-section of the filter membrane after step b). and Step d) confirms the presence of the protein aggregates in the cross-section of the filter membrane; The staining agents that are specific to the aggregates are the following (I), (II), or (III): (I) Selected from at least one of Nile Red, Thiosulfate T and PROTEOSTAT; (II) A staining agent containing a binding substance that specifically binds to the aggregate portion of the protein aggregate, the binding substance being labeled with a fluorescent or luminescent substance, the binding substance being at least one selected from antibodies, antibody fragments and peptides. (III) Contains the following (A) and (B): (A) A primary binding substance that specifically binds to the aggregate portion of the protein aggregate; (B) A secondary binding substance that specifically binds to the primary binding substance and contains a detectable substance, wherein... The primary binding substance is at least one selected from antibodies, antibody fragments, and peptides, or The secondary binding substance is an antibody labeled with a fluorescent substance, or The secondary binding substance is an enzyme-modified antibody. After step c), the process further includes adding a substrate that will be decomposed by the enzyme into a fluorescent or luminescent substance to the filter membrane and detecting fluorescence or luminescence.

2. The method according to claim 1, wherein step c) is performed prior to step b).

3. The method according to claim 1, further comprising a step of washing the filter membrane after step a).

4. The method according to claim 1, further comprising a step of washing the filter membrane after step c).

5. The method of claim 1, further comprising the step of adding a staining agent to the filter membrane to stain both the non-aggregated protein and the aggregated protein.

6. The method according to claim 1, further comprising the step of fixing the cross-section of the filter membrane.

7. The method according to claim 1, wherein, The cross-section of the filter membrane is the cross-section of a slice cut from the filter membrane.

8. The method according to claim 1, wherein, The enzyme is a peroxidase or an alkaline phosphatase.

9. The method according to claim 1, wherein, The filter membrane is a virus removal membrane.

10. The method of claim 1, further comprising, after step c), a step of removing unreacted staining agent specific to the aggregate.

11. The method according to claim 10, wherein, The unreacted staining agent specific to the aggregates is removed by dialysis, ultrafiltration, or gel filtration.

Images