JP2024088949A - Method for filtering biopolymer solutions - Google Patents

Abstract

[Problem] The present invention aims to provide a method for efficiently filtering a biopolymer solution through a 15 nm virus removal membrane. [Solution] The method includes a step of treating a biopolymer solution with a virus removal filter having a pore size of about 20 nm, and a step of treating the obtained filtrate with a virus removal filter having a pore size of about 15 nm, in which the amount of the biopolymer filtered is 20 kg/m2 or more per filter membrane area with a pore size of about 20 nm or about 15 nm. [Selected Figure] Figure 1

Description

The present invention relates to a method for filtering biopolymers, comprising the steps of treating a biopolymer solution with a virus removal filter having a pore size of about 20 nm and treating the obtained filtrate with a virus removal filter having a pore size of about 15 nm, in which the amount of the biopolymer filtered is 20 kg/ ^m2 or more per filter membrane area having a pore size of about 20 nm or about 15 nm.

Biological raw materials are raw materials derived from humans and other living organisms (excluding plants) that are used in pharmaceuticals, quasi-drugs, cosmetics, medical devices, and regenerative medicine products. When these are used in the manufacture of pharmaceuticals, etc., the necessary measures to be taken to ensure the quality, efficacy, and safety of pharmaceuticals, etc. are stipulated in Japan's "Standards for Biological Raw Materials" (Non-Patent Document 1: Ministry of Health, Labour and Welfare Notification No. 37/Established February 28, 2018).

Biological raw materials are primarily natural polymers produced by the cells of living organisms, known as biopolymers. Biopolymers are large molecules composed of covalently linked monomer units. They are classified into three main classes: polynucleotides, polypeptides, and polysaccharides, based on the constituent monomers and the structure of the resulting biopolymer. Polynucleotides, such as RNA and DNA, are long polymers composed of 13 or more nucleotide monomers. Polypeptides are polymers of amino acids and range from small molecules such as insulin to large proteins such as collagen, actin, and fibrinogen. Polysaccharides are linear or branched polymeric carbohydrates, including starch, cellulose, and alginic acid. Other examples of biopolymers include natural rubber (a polymer of isoprene), suberin and lignin (complex polymers of polyphenols), cutin and cutan (complex polymers of long-chain fatty acids), and melanin.

These biological raw materials are biopolymers derived from living organisms, and therefore carry the risk of infection with pathogens such as viruses. For this reason, in plasma fraction preparations in which proteins are purified from human plasma and used, and biological preparations in which biopolymers cultured and expressed with the addition of bovine serum, etc., are used as active ingredients, there is concern about contamination with infectious factors such as viruses derived from the raw materials or the manufacturing process, and a process for inactivating or removing viruses during the manufacturing process, i.e., a virus inactivation/removal process, is very important from the perspective of safety. In addition, infectious factors that are the subject of virus inactivation/removal are not limited to viruses. Examples include Creutzfeldt-Jakob disease caused by the transmission of proteinaceous infectious factors prions. As described above, biological polymer solutions may contain various infectious factors such as viruses. Therefore, when manufacturing products using these raw materials, it is essential to incorporate a process for sufficiently inactivating and removing viruses, etc.

Methods for inactivating these viruses include heat treatment, organic solvent/surfactant treatment (S/D treatment), chemical treatment, pH treatment, irradiation, and photosensitizer + UV treatment (Non-Patent Document 2: Biopharmaceutical Handbook, 4th Edition, 2020, Jiho Co., Ltd.). However, each of these methods has its advantages and disadvantages, and it is difficult to completely inactivate or completely remove various viruses using a single method, and these methods may also alter the biopolymers themselves in the product. Therefore, it is considered effective to use a combination of multiple virus inactivation and removal methods.

These virus removal methods physically remove viruses without chemically altering biopolymers, and include virus removal/separation methods using precipitation, chromatography, and filtration (Non-Patent Document 2: Biopharmaceutical Handbook, 4th Edition, 2020, Jiho Co., Ltd.). Among these, the filtration method using a virus removal membrane has been considered the most robust virus removal technology in recent years. Known virus removal membranes are made of natural materials such as cellulose, or synthetic polymer materials such as polyvinylidene fluoride (PVDF), polyethersulfone (PES), polysulfone (PSU), or polyethylene (PE).

In the virus removal membrane filtration method, when the molecular weight of the biopolymer in the biopolymer solution is small, the virus cannot pass through, but a virus removal membrane with a pore size that allows the biopolymer to pass through is used. Virus removal membrane filtration is a method of removing pathogens such as viruses that are larger than the pore size of the removal membrane from the product during the manufacturing process. However, one of the major problems in removing pathogens such as viruses from a biopolymer solution using a membrane is that the virus removal membrane becomes clogged, making filtration difficult or impossible. The pore size of the virus removal membrane varies depending on the manufacturer and standard, and is 10 to 70 nm. On the other hand, when a protein is used as an example of a biopolymer that is desired to pass through the pores of the removal membrane in filtration, the molecular weight of the protein is 10 to several thousand kDa, and the size of the biopolymer is at the same level as the pore size of the virus removal membrane. The smaller the pore size of the virus removal membrane, the smaller the viruses that can be captured and removed, so the greater the expected effect of the virus removal process method. However, depending on the size of the biopolymers to be passed, it may be impossible to separate the biopolymers from the viruses. Therefore, in the virus removal membrane filtration of a biopolymer solution, it is necessary to select the pore size of the removal membrane to be used, taking into consideration the size of the biopolymers to be passed. Even if a removal membrane with an appropriate pore size is selected, trace amounts of aggregates and other contaminating biopolymers present in the target biopolymer solution cannot pass through the pores of the virus removal membrane and block the pores, causing clogging of the virus removal membrane and ultimately filtration problems. In particular, in filtration with a virus removal membrane with a small pore size (10 to 20 nm), the size of the biopolymers that can pass through the pores becomes small, and preventing this clogging becomes an issue depending on the type and amount of biopolymers contained in the filtrate.

It is desirable that the filtration of a virus-containing solution using a virus removal filter device equipped with a virus removal membrane can filter a large amount of biopolymer solution in a short time and can exhibit sufficiently high virus removal performance. In order to process a large amount of biopolymer solution in a short time, the filtration of the virus-containing solution is generally performed at a high flow rate or high pressure as much as possible. If filtration is continued under such high pressure, biopolymers that should be contained in the filtrate may remain inside the membrane and lead to clogging of the removal membrane pores. In addition, in recent years, the protein concentration in pharmaceuticals that contain proteins as active ingredients, which are proteinaceous products that are a type of biopolymer, has tended to increase, and therefore in the virus removal membrane filtration process for removing viruses, it is advantageous in the manufacturing process to increase the concentration of proteins to be used in the virus removal membrane filtration. When a high-concentration protein solution is filtered through a small-pore virus removal membrane, clogging of the virus removal membrane due to protein residues inside the membrane is particularly noticeable, so filtration at high pressure may be inconvenient.

In recent years, virus removal membrane filtration has been positioned as the most robust technology, but there have been reported cases where extreme pressure conditions can affect the detectability of viruses after virus removal membrane filtration, i.e., virus leakage. As mentioned above, high transmembrane pressure (TMP) in virus removal membrane filtration is exemplified as one of the worst cases in virus removal membrane filtration (Non-Patent Document 2: Biopharmaceuticals Handbook, 4th Edition, 2020, Jiho Co., Ltd.).

Another finding is that when the TMP is low or immediately after repressurization following a temporary pressure release, virus particles are transiently detected in the filtrate, particularly in filtration experimental systems loaded with small parvoviruses or bacteriophages. This phenomenon is thought to be caused by the Brownian motion suppressed by the filtrate flow, and virus particles trapped in the virus removal membrane resume Brownian motion when the pressure is removed, detaching from the virus removal membrane wall and flowing out of the membrane through holes larger than the particle diameter at the next pressurization. This phenomenon is not specific to a particular virus removal membrane, but is common. (Non-patent Document 2: Biopharmaceutical Handbook, 4th Edition, 2020, Jiho Co., Ltd.)

Another finding is that small viruses or small virus-like particles, such as non-enveloped viruses such as minute virus of mice (MVM), can leak out of the virus removal membrane without clogging it when the load on the membrane increases (Non-Patent Document 3: Kayukawa T, Yanagibashi A, Hongo-Hirasaki T, Yanagida K. Particle-based analysis elucidates the real retention capacities of virus filters and enables optimal virus clearance study design with evaluation systems). of diverse virological characteristics. Biotechnol Prog. 2022 Mar;38(2):e3237. ).

Therefore, it is thought that the leakage of small viruses is affected by the amount of molecules of similar size to viruses present in the biopolymer sample solution filtered by the virus removal membrane. Therefore, molecules of similar size to viruses are not limited to viruses, and the coexistence and amount of biopolymers or their aggregates, or other types of biopolymers, may affect the leakage of viruses.

Therefore, when filtering using a virus removal membrane, it is necessary to set appropriate virus removal membrane filtration conditions in advance, taking into account the size, characteristics, and sample amount of the biopolymer to be filtered, as well as the size, characteristics, and amount of infectious substances that may be contaminating.

The impact of the virus removal membrane filtration process on virus removal ability can be predicted by conducting a virus clearance test that mimics the actual process.

Virus clearance tests are tests to evaluate the virus inactivation and removal capabilities required for managing the safety of, for example, biotechnology-based pharmaceuticals during process development and drug approval applications. As indicated in "Virus Safety Assessment of Biotechnology-Based Pharmaceuticals Produced Using Human or Animal Cell Lines (ICH-Q5A)" (Non-Patent Document 4), the virus inactivation and removal capabilities during the virus inactivation and removal process are quantitatively evaluated in a scaled-down production system.

In addition, in Japan, there is the "Guideline for Ensuring the Viral Safety of Plasma Fraction Preparations" (Non-Patent Document 5) for plasma fraction preparations, which requires a virus clearance test called a virus process validation test to verify or predict that known and unknown viruses that may be present in the raw plasma can be effectively removed and inactivated during the manufacturing process.

The virus clearance index obtained by the virus clearance test in the virus inactivation/removal step is generally expressed as a logarithmic value of the virus reduction value in the virus inactivation/removal step. For example, in the "Guidelines for Ensuring the Safety of Plasma Fraction Preparations Against Viruses" (Non-Patent Document 5), the virus clearance index R is expressed by the following formula:
R = log ((V1 x T1) / (V2 x T2))
where R is the logarithmic reduction, V1 is the volume of the sample before the process, T1 is the viral titer of the sample before the process, V2 is the volume of the sample after the process, and T2 is the viral titer of the sample after the process.
The virus clearance index, which is the logarithmic removal rate, is also called RF (reduction factor) or LRV (log reduction value).
In this specification, the viral clearance factor is expressed as a numerical value, but when citing the original text, a notation combining the number and log (e.g., "4 log") may be used.

The revised version of the "Guidelines for Ensuring the Viral Safety of Plasma Fraction Preparations" (Non-Patent Document 6) envisages a virus inactivation/removal process with a virus clearance capacity of 4 log or more as a "highly robust process," and states in the "Terminology" section that this target value means a process with extremely high virus clearance capacity. In addition, Non-Patent Document 7 (Masahiro Oda, Tsuyoshi Terano, Motoyoshi Okamura, Osamu Kawamata, Takeshi Arai, Sadao Ozawa, Kimihiko Kosugi, Yuanqi Hong, Hideharu Takahashi, Mina Tsukiyama, Tetsuo Sato, Tetsuya Oba, Issues and Case Studies of Viral Clearance Tests - How Far Should the Total Viral Clearance Value (LRV) Be Pursued?, PDA Journal of GMP and Validation in Japan, 2005 7(1):44-54) states that as of 2005, "the target value to be achieved is approximately 4 for the LRV," and it is believed that a consensus has been reached for many years as to the target value for the viral clearance value in one process of virus inactivation and removal.

Among the virus removal membranes, virus removal filters that are designed to remove even small viruses such as non-enveloped viruses are commercially available under product names such as Planova (trademark) 15N (Asahi Kasei), Planova (trademark) 20N (Asahi Kasei), Planova (trademark) Bio-EX (Asahi Kasei), Ultipore VF-DV20 (Pall), and Viresolve NFP (Millipore). Virus removal filters come in a variety of types, including hollow fiber, pleated, and flat membrane types.

Among these, Planova (trademark), which is commercially available as a hollow fiber membrane filter, introduces a protein solution into the hollow fiber and filters it through a membrane with a mesh-like structure in which large hollow void holes and long, thin capillary holes are three-dimensionally connected. It is explained that the membrane of the hollow fiber has a layered pore structure of several tens of micrometers, and viruses are efficiently captured and removed in stages within the membrane, while proteins pass through the membrane with minimal adsorption or deactivation. The basic structure of the filter cartridge container remains the same, and only the number of hollow fibers is increased or decreased depending on the size, making it possible to steadily scale up from experiments in the development stage to production. (Planova Filter Catalog, Asahi Kasei Medical Co., Ltd. Bioprocess Division, TAE31002J-4.1: Non-Patent Document 8)

The virus clearance index of Planova™, a virus removal membrane, increases in proportion to the size of the virus particle. The virus clearance index of PPV (porcine parvovirus, size 18-24 nm), which is known as one of the small diameter non-enveloped viruses, is less than 1.0 for Planova™ 35N (average pore size 35±2 nm), but is 4.0 or more for Planova™ 20N (average pore size 19±2 nm) and Planova™ 15N (average pore size 15±2 nm). This is because this virus removal membrane filtration is not influenced by physicochemical properties such as adsorption, and removes and separates viruses by the filtration principle of size exclusion (screening by size). Furthermore, it has been shown that as long as the size of the virus meets the conditions of the pores of the virus removal membrane, it is possible to remove it even if an unknown virus is mixed in. (Planova Filter Catalog, Asahi Kasei Medical Co., Ltd. Bioprocess Division, TAE31002J-4.1: Non-Patent Document 8)
Furthermore, the price of virus removal membranes generally increases inversely proportional to the pore size.

Non-Patent Document 9 (Modern Media. 2013 59(9):231-237) shows that filtering a prion protein solution using a virus removal membrane with a pore size of 15 nm has a certain degree of removal effect (clearance index of 3.5 log or more) on abnormal prion protein particles, and is useful for reducing the risk of infection.

Therefore, when filtering a biopolymer solution using a virus removal membrane, it is considered useful to filter the solution using a virus removal membrane with as small a pore size as possible while checking the filterability of the biopolymer, not only for removing small viruses such as known non-enveloped viruses like parvovirus, but also for removing pathogens such as unknown viruses.

Serum albumin, a type of biopolymer, is a protein with a molecular weight of 66,000 Da and is a molecule consisting of 585 amino acids. Serum albumin is the protein most abundant in plasma, which accounts for approximately 60% of all proteins, and is synthesized in the liver in the body. It is involved in maintaining plasma colloid osmotic pressure in blood, neutralizing toxic substances, and maintaining acid-base balance, and also plays a role in transporting many drugs and compounds by nonspecifically binding to them. Albumin preparations made from the above albumin are used to treat hypoalbuminemia and hemorrhagic shock caused by albumin loss due to burns and nephrotic syndrome, and reduced ability to synthesize albumin. Albumin preparations are highly concentrated protein preparations with a protein concentration of 4.4 to 25%. In addition, the volume of each preparation is large, at 20 to 250 mL, and their production requires the processing of a large volume and a large amount of protein in a short period of time.

Compared to other plasma protein preparations, the production process of albumin preparations requires a huge amount of protein to be processed, so when introducing virus removal membrane filtration, the key points to consider when introducing virus removal membrane filtration are at which stage of the production process the filtration is performed and what composition and properties of the albumin-containing aqueous solution is to be filtered. In addition, the number of virus removal membranes used, the filtration area, and the costs required for them are also key points to set. In other words, when introducing a virus removal membrane filtration process, it is important for industrial application to find conditions that allow the amount of expensive virus removal membranes used to be kept as low as possible. Typically, the pore size of the virus removal membrane introduced in the production process of albumin preparations is about 15 nm or about 20 nm, based on the filtration ability based on the molecular weight of the membrane.

There are several reports on virus removal membrane filtration of albumin preparations. Patent Document 1 (JP Patent Publication 06-279296) describes an example of filtering human serum albumin with four types of virus removal membranes of 15 nm, 35 nm, 40 nm, and 75 nm. Patent Document 2 (WO2004/089402) also reports a pretreatment consisting of treating an aqueous solution containing human serum albumin with an anion exchanger and a prefilter before filtering it with a 15 nm virus removal membrane. Furthermore, Patent Document 3 (JP2007-535495) discloses a method for purifying albumin, which includes a step of subjecting an aqueous albumin solution having a concentration of 15 g/L to 80 g/L and a pH in the range of 9 to 11.5 to a virus removal membrane at a temperature range of 15°C to 55°C. Example 2 shows a filtration example in which a filter with a pore size of 15 nm is used as the virus removal membrane, and the filtration amount per membrane area is 0.5 and 1 kg/ ^m2 . Next, in Examples 3 and 4, filtration examples with a filtration amount per membrane area of 4 kg/ ^m2 are shown. Furthermore, in Examples 5 and 6, filtration examples with a maximum filtration amount per membrane area of 8 kg/ ^m2 are shown. In such virus removal membrane filtration for albumin preparations, attempts have been made to provide a prefilter for virus removal membrane filtration with a pore size of 15 nm. However, there have been no examples of using a filter with a pore size of 20 nm or less as a prefilter.

On the other hand, Non-Patent Document 10 (Biologicals. 2012.40(6):473-481) shows the virus clearance index in the virus removal membrane filtration of the protein holotransferrin when a pore size of 15 nm or a pore size of 20 nm is used alone, or when a pore size of 15 nm or a pore size of 20 nm is connected with a different membrane area ratio, and shows an example in which the virus clearance index is increased by connecting filters of the same pore size for filtration. Also shown is an example of filtration using a virus removal membrane filter connected with the protein apotransferrin or C1-inhibitor. Furthermore, in the case of a protein prothrombin complex concentrate (PCC) preparation, an example is shown in which a virus removal membrane with a pore size of 15 nm or a pore size of 20 nm is connected using canine parvovirus (CPV, 18-26 nm). However, although the loading amount per filter membrane area is not clear, the virus removal membranes with pore sizes of 15 nm and 20 nm used have membrane areas of 0.001 ^m2 and 0.01 ^m2 , respectively, and 200 mL of 20 g/L or 100 mL of 40 g/L of PCC formulation are filtered. Therefore, the protein loading amount per filter membrane area can be estimated to be a maximum of 4 kg/ ^m2 .

Although there are previous examples of filtering biopolymers using a virus removal membrane with a pore size of 15 nm, there are still challenges to be overcome in achieving a sufficient biopolymer loading amount and the ability to remove small viruses.

JP 06-279296 WO2004/089402 Special table 2007-535495

Ministry of Health, Labour and Welfare Notification No. 37, Standards for Biologically Originated Raw Materials, enacted on February 28, 2018 Biopharmaceutical Handbook, 4th Edition, 2020, Jiho Co., Ltd. Kayukawa T, Yanagibashi A, Hongo-Hirasaki T, Yanagida K. Particle-based analysis elucidates the real retention capacities of virus filters and enables optimal virus clearance study design with evaluation systems of diverse virological characteristics. Biotechnol Prog. 2022 Mar;38(2):e3237. Viral Safety Assessment of Biotechnology-Derived Products Produced Using Human or Animal Cell Lines (ICH-Q5A) Guidelines for Ensuring the Safety of Plasma Derivatives against Viruses (Notification No. 1047 of the Pharmaceuticals and Medical Devices Agency dated August 30, 1999) Regarding the partial revision of the "Guideline for ensuring the safety of plasma fraction preparations against viruses" issued by the Blood Control Division, Pharmaceutical and Food Safety Bureau, Ministry of Health, Labour and Welfare, March 2022 Masahiro Oda, Tsuyoshi Terano, Motoyoshi Okamura, Osamu Kawamata, Takeshi Arai, Sadao Ozawa, Kimihiko Kosugi, Yunki Hong, Hideharu Takahashi, Mina Tsukiyama, Tetsuo Sato, Tetsuya Ohba, Issues and case studies of viral clearance tests: How far should we pursue the total viral clearance value (LRV)?, PDA Journal of GMP and Validation in Japan, 2005 7(1):44-54. Planova Filter Catalog, Asahi Kasei Medical Corporation Bioprocess Division, TAE31002J-4.1 Yuko Kato (Mori), Mikihiro Yuzuki, Tomoyoshi Ikuta, Katsuro Hagiwara, Safety measures for blood products: current status of prion removal, Modern Media. 2013 59(9): 231-237 Koenderman AH, ter Hart HG, Prins-de Nijs IM, Bloem J, Stoffers S, Kempers A, Derksen GJ, Al B, Dekker L, Over J. Virus safety of plasma products using 20 nm instead of 15 nm filtration as virus removing step. Biologicals. 2012 Nov;40(6):473-81.

The present invention aims to provide a method for efficiently filtering a biopolymer solution through a 15 nm virus removal membrane.

As a result of extensive investigations to achieve the above-mentioned object, the inventors have discovered a method for filtering biopolymers, comprising the steps of treating a biopolymer solution with a virus removal filter having a pore size of about 20 nm and treating the obtained filtrate with a virus removal filter having a pore size of about 15 nm, in which the amount of filtered biopolymer is 20 kg/ ^m2 or more per filter membrane area with a pore size of about 20 nm or about 15 nm, and have completed the present invention.

Therefore, the present invention includes the following:
[1]
A method for filtering biopolymers, comprising the steps of treating a biopolymer solution with a virus removal filter having a pore size of about 20 nm, and treating the obtained filtrate with a virus removal filter having a pore size of about 15 nm, wherein the amount of the biopolymer filtered is 20 kg/m2 or more per filter membrane area having a pore size of about 20 nm or a pore size of about ¹⁵ nm.
[2]
The filtration method according to [1], wherein the biopolymer is a protein.
[3]
The filtration method according to [2], wherein the protein is human serum albumin.
[4]
The filtration method described in [1], wherein the virus clearance index is 3.5 or more.
[5]
The filtration method according to [1], wherein the amount of biopolymers filtered is 40 kg/ ^m2 or more, or 60 kg/ ^m2 or more per filter membrane area having a pore size of about 20 nm or a pore size of about 15 nm.
[6]
A method for producing a formulation containing a biopolymer, comprising: a step of filtering the biopolymer, the step of filtering the biopolymer, the step of filtering the obtained filtrate, the step of filtering the biopolymer, the step of filtering the obtained filtrate, the step of formulating the obtained biopolymer, wherein the amount of the biopolymer filtered is 20 kg/m2 or more per filter membrane area having a pore size of about 20 nm or a pore size of about ¹⁵ nm.

The present invention provides a method for efficiently filtering a biopolymer solution through a virus removal membrane when filtering the solution through a virus removal filter with a pore size of approximately 15 nm.

Fig. 1 shows the filterability of a human serum albumin solution through a small-pore virus removal filter. The left figure shows the change over time in the amount of filtered protein (throughput, kg/ ^m2 ), and the right figure shows the change over time in the flow rate of filtered protein (flux, kg/ ^m2 /h). A shows the result of filtration with Planova (trademark) 15N at a pressure of 0.3 bar, B shows the result of filtration with Planova (trademark) 15N at a pressure of 0.9 bar, and C shows the result of filtration with Planova (trademark) 20N at a pressure of 0.9 bar.

The following describes in detail preferred embodiments of the present invention. However, the present invention is not limited to the following embodiments.

In one aspect, the present invention relates to a method for filtering biopolymers, comprising the steps of treating a biopolymer solution with a virus removal membrane having a pore size of about 20 nm and treating the obtained filtrate with a virus removal membrane having a pore size of about 15 nm, wherein the amount of the biopolymer filtered is 20 kg/ ^m2 or more per area of the virus removal membrane having a pore size of about 20 nm or about 15 nm.

In the present invention, a biopolymer solution refers to an aqueous solution containing a biopolymer. Preferably, the biopolymer solution contains a biopolymer that can pass through a pore size of about 15 nm. As an example of a biopolymer to be passed through the pores of the removal membrane during filtration, when the target is a protein, its molecular weight is 10 kDa, 20 kDa, 30 kDa, 40 kDa, 50 kDa, 60 kDa, 70 kDa, 80 kDa, 90 kDa, 100 kDa, 200 kDa, 250 kDa, 500 kDa, 750 kDa, 1000 kDa, 1500 kDa, 2000 kDa, 2500 kDa, 3000 kDa, 4000 kDa, 5000 kDa, 6000 kDa, 7000 kDa, 8000 kDa, 9000 kDa, or 10000 kDa or less. Preferably, the molecular weight is any value between 10 and several thousand kDa, and the size of the biopolymer is at the same level as the pore size of the virus removal membrane. The biopolymer may include polynucleotides, polypeptides, and polysaccharides. Examples of polynucleotides include RNA and DNA. Polypeptides are polymers of amino acids, and include, but are not limited to, small molecules such as insulin and large proteins such as collagen, actin, and fibrinogen. In the present invention, the biopolymer is preferably a protein. Examples of proteins include, but are not limited to, granulocyte colony-stimulating factor, interferon, interleukin, macrophage colony-stimulating factor, thrombin, erythropoietin, blood clotting factor VII, urokinase, alpha-1 protease inhibitor, blood clotting factor IX, protein C, antithrombin III, hemoglobin, albumin, osteopontin, tissue plasminogen activator, C1 inhibitor, ceruloplasmin, monoclonal IgG, blood clotting factor XI, plasma-derived mannose-binding lectin, blood clotting factor VIII, human immunoglobulin G, fibrinogen, von Willebrand factor, or immunoglobulin M. More preferably, the protein is albumin. More preferably, the protein is human serum albumin.

The virus removal membrane used in the present invention refers to a filter for removing and separating viruses by the filtration principle of size exclusion (screening by size). Virus removal membranes or virus removal filters may include membranes formed from cellulose (e.g., cuprammonium regenerated cellulose), polyvinylidene fluoride (PVDF), polyethersulfone (PES), polysulfone (PSU), polyethylene (PE), etc. Examples of virus removal filters include Planova (trademark) 15N (Asahi Kasei), Planova (trademark) 20N (Asahi Kasei), Planova (trademark) Bio-EX (Asahi Kasei), Ultipore VF-DV20 (Pall), Viasorb NFP (Millipore), etc., and those skilled in the art can select the appropriate product. Types of virus removal filters include hollow fiber type, pleated type, flat membrane type, etc. Non-limiting examples of virus removal filters include Planova (trademark) 15N or 20N filters manufactured by Asahi Kasei Medical Co., Ltd., which contain a cuprammonium regenerated cellulose hollow fiber membrane. For example, a bundle of hollow fibers is contained in a container, and the biopolymer solution is introduced into the inside of the hollow fibers and filtered by passing through a membrane with a mesh-like structure in which large hollow void holes or long, thin capillary holes are three-dimensionally connected and then exiting. The recommended operating pressure for this virus removal filter Planova (trademark) N series is 0.98 bar or less.

The virus removal filter used in the present invention has a pore size of about 15 nm or about 20 nm. The pore size of the filter is indicated as an average pore size, and the range of the pore size indicated by the word "about" is ±1 nm, ±2 nm, or ±3 nm. For example, in the present invention, a Planova (trademark) 15N filter can be used as a filter with a pore size of about 15 nm, and a Planova (trademark) 20N filter can be used as a filter with a pore size of about 20 nm.

The virus removal filter can be scaled up by, for example, increasing the number of hollow fibers. The membrane area of the virus removal filter is about 0.0001, 0.0002, 0.0003, 0.0004, 0.0005, 0.0006, 0.0007, 0.0008, 0.0009, 0.001, 0.002, 0.003, 0.004, 0.005, 0.006, 0.007, 0.008, 0.009, 0.01, 0.011, 0.012, 0.013, 0.0 14, 0.015, 0.016, 0.017, 0.018, 0.019, 0.02, 0.05, 0.10, 0.11, 0.12, 0.13, 0.14, 0.15, 0.16, 0.17, 0.18, 0.19, 0.2 ^, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, ^{5.0 m 2} or more. Preferably, the membrane area of the virus removal filter is about 0.0003, 0.001, 0.01, 0.12, 0.3, 1.0, 4.0 m 2. By increasing the membrane area, the efficiency of filtration can be increased. For example, by increasing the membrane area, the loading amount of biopolymer solution per membrane area can be reduced, and the time required for filtration can be shortened.

In the present invention, the term "filtration amount" refers to the amount of solute biopolymer in the biopolymer solution passing through the virus removal filter per membrane area, and is expressed in kg/m ^2. In one embodiment, the filtration amount is 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80 kg/m ² or more per membrane area. The filtration amount is preferably 20, 30, 40, 50, 60, 70, 80 kg/m ² or more per membrane area.

In the present invention, the effect of the virus removal membrane filtration process on the virus removal ability can be predicted by performing a virus clearance test simulating the actual process. In the present invention, the "virus clearance index" is the logarithm of the virus removal rate. The clearance index may also be used as the "RF (reduction factor)" in this specification. The virus clearance index is expressed as log((V1 x T1)/(V2 x T2)). Here, V1 is the volume of the sample before the process treatment, T1 is the virus titer of the sample before the process treatment, V2 is the volume of the sample after the process treatment, and T2 is the virus titer of the sample after the process treatment. In one embodiment, the viral clearance factor is 2.0, 2.1, 2.2, 2.3, 2.4, 2.5, 2.6, 2.7, 2.8, 2.9, 3.0, 3.1, 3.2, 3.3, 3.4, 3.5, 3.6, 3.7, 3.8, 3.9, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 7.0 or more. In another embodiment, the clearance factor is 3.5, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 7.0 or more.

In the present invention, the virus may be an enveloped or non-enveloped virus. In the present invention, the virus may be, but is not limited to, Abelson murine leukemia virus (A-MuLV), human parvovirus B19 (B19), bovine diarrhea virus (BVDV), canine parvovirus (CPV), encephalomyocarditis virus (EMCV), hepatitis A virus (HAV), human immunodeficiency virus (HIV), human immunodeficiency virus type 1 (HIV-1), herpes simplex virus (HSV), minute virus of mice (MVM), poliovirus type 1 (Polio-1), porcine parvovirus (PPV), pseudorabies virus (PRV), reovirus type 3 (Reo-3), simian virus 40 (SV40), or xenotropic murine leukemia virus (XMuLV).

In the present invention, a pretreatment step may be carried out prior to the filtration step using a virus removal filter. The pretreatment may include pretreatment with an ion exchanger, pretreatment by prefiltration, or a combination thereof. In one embodiment, pretreatment by prefiltration may be carried out, and prior to the method of the present invention, the biopolymer solution may be passed through a prefilter (removal size 35-200 nm) to remove particles of 35-200 nm or more in size, which are believed to cause clogging, by trapping them.

In the present invention, a virus removal filter with a pore size of about 20 nm and a virus removal filter with a pore size of about 15 nm may be connected or may be separated. For example, a process of treating with a virus removal filter with a pore size of about 20 nm and a process of treating the obtained filtrate with a virus removal filter with a pore size of about 15 nm may be performed in series, or the filtrate may be collected after the process of treating with a virus removal filter with a pore size of about 20 nm, and the obtained filtrate may be treated with a virus removal filter with a pore size of about 15 nm. In one embodiment, when a biopolymer solution is filtered with a virus removal filter with a pore size of about 15 nm, a filter with a pore size of about 20 nm may be provided upstream of the virus removal filter with a pore size of about 15 nm.

In one embodiment, virus leakage due to filtration is reduced or does not occur. Reducing or eliminating virus leakage increases the safety of the filtrate of the biopolymer solution, which may be advantageous for the manufacture and safety of products derived from the filtrate. Virus leakage and removal ability can be confirmed, for example, by the virus clearance index.

In one embodiment, the amount of biopolymer in the biopolymer solution applied to the filter is, but is not limited to, about 20, 30, 40, 50, 60, 70 kg/ ^m2 or more. Preferably, it is 50 kg/ ^m2 or more.

In one embodiment, the concentration of the biopolymer contained in the biopolymer solution fed to the filtration filter is about 0.001%, 0.01%, 0.1%, 1%, 8%, or 10%, but is not limited to these. Preferably, it is 1% or more.

In one embodiment, in the process of filtering, the pressure refers to the transmembrane pressure (TMP) of the virus removal filter used, which is about 0.10, 0.15, 0.20, 0.25, 0.30, 0.35, 0.40, 0.45, 0.50, 0.55, 0.60, 0.65, 0.70, 0.75, 0.80, 0.85, 0.90, or 0.98 bar. Preferably, the TMP is performed within the recommended operating pressure range of the virus removal filter used.

In one embodiment, the filtering step is performed at, but is not limited to, about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 35, 37, 40, 45, 50°C or higher. Preferably, the filtering step is performed at room temperature (1°C to 30°C).

In one embodiment, in the process of treating with a filtration filter, the pH of the biopolymer solution can be appropriately set by a person skilled in the art according to the purpose. In the process of treating with a filtration filter, the biopolymer solution can contain various substances in addition to the biopolymer, and can be appropriately set by a person skilled in the art according to the purpose. For example, the biopolymer solution contains a stabilizer or a buffer.

In the process of filtering through a virus removal filter, filtration may become difficult due to clogging of the virus removal filter. Therefore, in one embodiment, filtration through a virus removal filter can be performed with a flow rate (flux) at the end of filtration that is 1%, 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, 90% or more of the flow rate (flux) at the beginning of filtration, and can be appropriately set by a person skilled in the art according to the purpose.

In the virus clearance test of the virus removal membrane filtration process, virus leakage may occur. Therefore, in one embodiment, filtration with a virus removal filter has a virus clearance index of 3.5, 3.6, 3.7, 3.8, 3.9, 4.0, 4.5, 5.0, 5.5, 6.0, 6.5, 7.0 or more. Preferably, the clearance index is in the range of 3.5 or more.

In one aspect of the present invention, there is provided a method for filtering a biopolymer, comprising the steps of treating a biopolymer solution with a virus removal filter having a pore size of about 20 nm and treating the obtained filtrate with a virus removal filter having a pore size of about 15 nm, in which the amount of the biopolymer filtered is 20 kg/ ^m2 or more per filter membrane area having a pore size of about 20 nm or about 15 nm, as well as a method for producing a formulation containing a biopolymer, comprising the step of formulating the obtained biopolymer.

In the present invention, the terms "removal membrane" and "removal filter" are used interchangeably. Unless otherwise indicated, in this specification, "about" means within a range of ±10%, preferably ±5%.

The inventors have investigated an effective virus removal membrane filtration method using a smaller amount of filter and have obtained the following findings. The filtration method of the present invention can be applied to any biopolymer solution that can be fed to a virus removal filter with a pore size of 15 nm. An example of a biopolymer is a protein. The present invention will be described in detail using an example using an albumin solution as an example, but the present invention is not limited to these examples.

Example 1: Preparation of an aqueous solution containing human serum albumin According to the method described in Patent Document 2 (WO2004/089402), the V-fraction paste obtained by alcohol fractionation of Cohn was mixed with water for injection (JP) at least twice the weight of the paste, stirred and dissolved, and the albumin fraction dealcoholized by ultrafiltration was adjusted to pH 4.5, EC 1.5 mS/cm or less, and protein concentration 12 w/v%. Then, the albumin-containing aqueous solution was developed on an anion exchanger (Q Sepharose (registered trademark) FastFlow (Cytiva)) equilibrated with an acetate buffer of pH 4.5 and EC 5 mS/cm or less, and the non-adherent fraction was collected. The protein concentration was adjusted to 8 w/v%, pH 7.0, and filtered through a filter with a pore size of 0.1 μm to be used as the test material.

Example 2: Filtration of human serum albumin solution through small-pore virus removal membrane In order to compare and confirm the filtration flow rate due to the difference in pore size of the virus removal membrane, the above-mentioned albumin-containing aqueous solution (protein concentration 8%) was placed in a pressure vessel, and a constant pressure was applied using a nitrogen gas cylinder equipped with a pressure regulator, and the solution was passed through a virus removal membrane (manufactured by Asahi Kasei Medical), Planova (trademark) 15N (pore size 15±2 nm) or 20N (pore size 19±2 nm) (each membrane area 0.001 m ² ), to perform virus removal membrane filtration. Planova (trademark) 15N was filtered under constant pressures of 0.3 bar and 0.9 bar. Planova (trademark) 20N was filtered under constant pressure of 0.9 bar. The results are shown in Figure 1. The left figure shows the change over time in the filtration amount (throughput, kg/m ² ), and the right figure shows the change over time in the flow rate (flux, kg/m ² /h). Curve A is the result of filtration under pressure of 0.3 bar with Planova™ 15N, curve B is the result of filtration under pressure of 0.9 bar with Planova™ 15N, and curve C is the result of filtration under pressure of 0.9 bar with Planova™ 20N.

A comparison of curves A (pressure 0.3 bar) and B (pressure 0.9 bar) using Planova™ 15N confirmed that the filtration volume increases in a short time as the filtration pressure increases, and that the flow rate increases as the filtration pressure increases. A comparison of filtration with Planova™ 15N (curve B) and filtration with Planova™ 20N (curve C) at the same pressure of 0.9 bar confirmed that the filtration volume increases in a short time as the pore size of the virus removal filter increases, and that the flow rate increases as the pore size of the virus removal filter increases. In addition, the flow rate of curve B showed a greater tendency to decrease than that of curve C, indicating a tendency for clogging due to the small pore size of the virus removal membrane.

Example 3: Filterability of human serum albumin solution under the same filtration volume The same albumin-containing aqueous solution as in Example 2 was prepared, and the time required for virus removal membrane filtration at the same filtration volume (56 kg/m ² ) was compared. The results are shown in the table below.

When an aqueous albumin solution was filtered using Planova™ 15N at a constant pressure of 0.8 bar and the filtration temperature was changed to 10°C, 23°C, 30°C, and 37°C, the filtration time (h) decreased with increasing temperature, to 101/97, 64/55, 52/48/55, and 44 (Tests No. 1-8). In contrast, when Planova™ 20N of the same membrane area was added before Planova™ 15N and connected for filtration, Planova™ 15N was filtered at a constant pressure of 80 kPa, and the filtration time was 52 h (Test No. 9). This shows that even when Planova 20N was placed in the front stage as a prefilter, the filtration time was equivalent to that of Tests No. 5-7 without such a prefilter. In addition, the behavior of the connected continuous filtration with a configuration of Planova (trademark) 15N + Planova (trademark) 15N is considered to be mainly governed by the filterability of the Planova 15N filtration in the previous stage, and the filtration time can be predicted from the test results of Test Nos. 1 to 8.

Example 4: Filterability and virus capture ability of human serum albumin solution with added model virus The filtration conditions of the virus removal membrane Planova (trademark) 15N (manufactured by Asahi Kasei Medical) were changed to compare the filtration amount (protein load). In addition, a porcine parvovirus (PPV, size 18-24 nm) solution was added as a model virus to the above-mentioned albumin-containing aqueous solution at an amount of 1 v/v% or 0.1 v/v%, and this was filtered through Planova (trademark) 15N, and the virus infectivity of the virus mixed solution before and after filtration was measured to confirm the virus clearance index (RF). In test No. 16, in which Planova (trademark) 20N was installed in front of Planova (trademark) 15N, both virus removal filters with the same membrane area were connected and filtered, and the TMP of Planova (trademark) 15N was adjusted to 0.8 bar for filtration. The results are shown in the table below.

A process with a virus clearance index of 4 or more is assumed as a highly robust virus inactivation/removal process. From the results of Test No. 11, when an albumin solution loaded with 1 v/v% PPV was subjected to filtration with Planova (trademark) 15N, the virus removal filter became clogged at about half the filtration, and the virus clearance index could not be evaluated. Next, from the results of Test No. 12 to 15, in filtration using only Planova (trademark) 15N, PPV was detected in the filtrate even when the protein load was reduced from 62 kg/m ² to 56 kg/m ² , 45 kg/m ² , or 23 kg/m ² , and the virus clearance index remained at about 3, which did not result in a highly robust virus removal process, and there was no effect of the filtration temperature. This shows that even if the amount of filters with a pore size of about 15 nm is simply increased several times, it is difficult to achieve a highly robust virus removal process with a protein load of several tens of kg/m ^2. In addition, these Test Nos. Although the total PPV load in Test No. 12 to 15 was less than the total PPV load in Test No. 11, no significant blockage due to clogging of the virus removal filter occurred, and virus leakage was observed. In Test No. 16, when filtration was performed using Planova™ 20N with the same membrane area provided in front of Planova™ 15N, PPV was not detected in the filtrate, and the virus clearance index was 3.7 or more, indicating a highly robust virus removal process.

Example 5: Comparison of aggregate content before and after virus removal membrane filtration The albumin solution before and after the filtration in Example 2 was analyzed by size-exclusion high performance liquid chromatography (SE-HPLC). The analysis conditions for SE-HPLC are as follows.
Column: TSKgel G3000SWXL, 7.8 mm x 30 cm
Mobile phase: 0.1 mol/L sodium phosphate + 0.1 mol/L sodium sulfate, 0.05% sodium azide, pH 6.7
Column temperature: constant temperature around 25°C Flow rate: 0.5 mL per minute
Detection: 280 nm
The results of comparing the amount of aggregates contained in each sample are shown in the table below.

In SE-HPLC, molecules are eluted in order from larger to smaller molecules due to the principle of size exclusion. In this SE-HPLC analysis, the molecules are fractionated and eluted between retention times of 10 and 20 minutes. As a result of the analysis, the monomer of human serum albumin was eluted at a retention time of about 18 minutes, and the dimer was eluted at a retention time of about 16 minutes. Although a small peak was detected with a content of at least 0.1%, there was no peak between retention times of about 10 and 16 minutes, and no aggregates larger than the dimer of human serum albumin were detected. And, as shown in Table 3, there was no significant change in the content of aggregates, dimers and monomers of human serum albumin before and after virus removal membrane filtration. It is considered that the amount of molecules of the same size as viruses present in the biopolymer sample solution filtered by the virus removal membrane affects the leakage of small-diameter viruses. Therefore, it is preferable to have fewer protein aggregates. The molar concentration of an 8% solution of human serum albumin with a molecular weight of 66,000 Da is about 1.2 mmol/L. The amount of aggregates can be evaluated at 1/1000 of that amount by SE-HPLC. However, the amount of PPV added to the liquid before filtration in Example 4 is at the level of ^107. This virus amount is at the amol to fmol level, which is a very small amount. It can be seen that in virus removal membrane filtration of a high-concentration protein solution, even if aggregates are evaluated by an analytical method such as SE-HPLC, it is difficult to estimate the effect on the filterability of the virus removal membrane filtration.

For example, a prefilter with a pore size of 35 nm or more has been provided upstream of Planova (trademark) 15N. However, when a prefilter with such a pore size is applied to a small non-enveloped virus such as PPV (porcine parvovirus, size 18-24 nm), the PPV easily passes through the prefilter. Therefore, the addition of a known prefilter to Planova (trademark) 15N cannot be expected to enhance the virus clearance ability of small viruses such as non-enveloped viruses. Therefore, for example, a highly robust virus removal process can be easily performed by a person skilled in the art by repeatedly filtering through a virus removal filter with a pore size of about 15 nm and then filtering through a virus removal filter with a pore size of about 15 nm, or by connecting and filtering a virus removal filter with a pore size of about 15 nm. However, in this case, the load of proteins and viruses on the front-stage virus removal filter with a pore size of approximately 15 nm increases, and effective distribution of the load of proteins and viruses to both successive filters cannot be expected, and there is a potential risk of clogging on the front-stage virus removal filter with a pore size of approximately 15 nm. Furthermore, the cost of the virus removal filter is twice that of a virus removal filter with a pore size of approximately 15 nm. To date, there have been no examples of using a filter with a pore size of 20 nm or less as a prefilter in virus removal membrane filtration of a biopolymer solution of several tens of kg/ ^m2 . The inventors have clarified in Example 1 the difference in the clogging characteristics of biopolymers between filters with a pore size of approximately 20 nm and filters with a pore size of approximately 15 nm, and have newly discovered in the present invention that a filter with a pore size of 20 nm or less is newly used in front of a virus removal filter with a pore size of 15 nm, thereby distributing the load of biopolymers and viruses over both the virus removal filter with a pore size of approximately 20 nm and the virus removal filter with a pore size of approximately 15 nm, thereby providing a method for efficient, low-cost virus removal membrane filtration of biopolymers of 20 kg/ ^m2 or more. For example, when an albumin-containing aqueous solution with a filtration rate of biopolymers of 20 kg/m2 or ^more is filtered with a virus removal membrane with a pore size of about 15 nm, it is effective to provide a membrane with a pore size of about 20 nm upstream of the virus removal membrane with a pore size of about 15 nm, and this results in a robust virus removal membrane filtration method even when the filtration rate of biopolymers is at the level of several tens of kg/ ^m2 per membrane area with a pore size of about 20 nm or a pore size of about 15 nm, or even at 60 kg/ ^m2 or more.

The present invention provides a method for efficiently filtering a biopolymer solution through a virus removal membrane with a pore size of approximately 15 nm.

Claims (6)

A method for filtering biopolymers, comprising the steps of treating a biopolymer solution with a virus removal filter having a pore size of about 20 nm, and treating the obtained filtrate with a virus removal filter having a pore size of about 15 nm, wherein the amount of the biopolymer filtered is 20 kg/m2 or more per filter membrane area having a pore size of about 20 nm or a pore size of about ¹⁵ nm. The filtration method according to claim 1, wherein the biopolymer is a protein. The filtration method according to claim 2, wherein the protein is human serum albumin. The filtration method according to claim 1, wherein the virus clearance index is 3.5 or more. 2. The filtration method according to claim 1, wherein the amount of biopolymers filtered is 40 kg/ ^m2 or more, or 60 kg/m2 or more per area of a filter membrane having a pore size of about 20 nm or a pore size of about ¹⁵ nm. A method for producing a formulation containing a biopolymer, comprising: a step of filtering the biopolymer, the step of filtering the biopolymer, the step of filtering the obtained filtrate, the step of filtering the biopolymer, the step of filtering the obtained filtrate, the step of formulating the obtained biopolymer, wherein the amount of the biopolymer filtered is 20 kg/m2 or more per filter membrane area having a pore size of about 20 nm or a pore size of about ¹⁵ nm.

Images