System and methods for counter-current filtration

Abstract

The invention relates to systems and methods for the removal or reduction of impurities from a recombinantly produced protein or polypeptide preparation by implementing counter- current filtration.

Description

[0001] SYSTEM AND METHODS FOR COUNTER-CURRENT FILTRATION

[0002] TECHNICAL FIELD

[0003] The present disclosure relates to the field of biotechnology, and in particular to recombinant protein manufacturing and aims to address the challenge of economically processing large quantities of biomolecules from crude production solutions. Disclosed herein are systems, methods, and uses of counter-current filtration systems for the removal or reduction of impurities in a protein preparation. More particularly, the filtration systems and methods operate to reduce impurity (e.g., host cell proteins (HCP)) content in a protein preparation recombinantly produced in a host cell in the manufacturing process of proteins intended for administration to a patient. The disclosed methods may be performed to produce therapeutic compositions having reduced host cell protein content.

[0004] BACKGROUND

[0005] The listing and discussion of an apparently prior-published document in this specification should not necessarily be taken as an acknowledgement that the document is part of the state of the art or is common general knowledge.

[0006] The field of biotechnology has seen significant advancements in recent years, resulting in the production of large quantities of biomolecules. These biomolecules include enzymes, antibodies, proteins, and peptides, which are typically produced using recombinant DNA techniques in various biological systems. However, the economic processing of these biomolecules from crude production solutions remains a challenge for the biopharmaceutical industry. The crude production solutions include inter alia conditioned cell culture media, clarified homogenised / lysed conditioned cell cultures, and in vitro production solutions that include starting materials and components in addition to the biomolecule of interest. The recovery of these biomolecules from crude production solutions or production materials must meet regulatory purity and safety standards while also being economically feasible. WO 2007 / 120449 describes systems and counter-current diafiltration methods for the separation, recovery, and / or purification of proteins, peptides, nucleic acids, biologically produced polymers and other compounds from aqueous fluids. In counter-current diafiltration, the sample is typically separated in retentate and permeate, where the retentate contains the desired biomolecule and the permeate contains the impurities. The retentate and permeate are circulated in opposite directions, with the retentate flowing counter-currently to the permeate. This creates a concentration gradient across the membrane, which helps to drive the separation of the desired biomolecule from the impurities. As the retentate is circulated in a counter-current direction to the permeate, fresh buffer solution is continuously added to the retentate to help remove impurities and maintain a constant volume.

[0007] Host cell proteins (HCP) are endogenous proteins from host cells that are involved in cell maintenance and growth, and protein synthesis and processing. There is however a particular challenge with such proteins in the field of therapeutic or diagnostic proteins. The presence of HCP, in particular immunogenic HCP, negatively influences product quality and patient safety by posing concerns such as inducing aggregation, product fragmentation by catalytic activity and / or eliciting an immune response due to their foreign nature (i.e., nonhuman). HCP have therefore been identified as critical quality attribute (CQA) of protein formulations. Due to the quality and safety concerns, several purification steps are often required to remove HCP below the limit set by the regulatory bodies / the arbitrarily but commonly accepted amounts of residual HCP in final bulk material in the range of 1-100 ppm (F. Wang, D. Richardson, M. Shameem, Host-cell protein measurement and control, Biopharm. Int. 28 (2015) 32-38). For instance, despite the high selectivity of protein A chromatography, HCP can persist after this step and pose a significant clearance burden to subsequent polishing steps (e.g., anion exchange chromatography; cation exchange chromatography; hydrophobic interaction chromatography; mixed-mode chromatography). A recent article by Zydney et al., described achieving a 2-log HCP removal by high- performance counter-current membrane purification (HPCMP), which, however, still requires additional polishing operations to reduce to <100 ppm (A. Mohammadzadehmarandi, A. Determan, C. Krumm, L. D. McIntosh, A. L. Zydney, High-performance counter-current membrane purification for host cell protein removal from monoclonal antibody products). WO 2022 / 072934 describes a process of reducing HCP in the purification process of therapeutic or diagnostic antibodies or antigen-binding fragments thereof. WO 2018 / 047080 describes a process of purifying a recombinant polypeptide from host cell proteins by applying a solution comprising the recombinant polypeptide and HCP to a superantigen chromatography solid support, washing the superantigen chromatography solid support with a wash buffer, and eluting the recombinant polypeptide.

[0008] As such, despite the aforementioned methods, there remains a need for improved or novel systems for the recovery of desired biomolecules from crude production solutions or production materials and / or purification processes for recombinantly produced biomolecules from impurities, such as, e.g., host cell proteins. In particular, there remains a need for alternative methods of reducing HCP in the purification process of therapeutic or diagnostic proteins. Such alternative methods reduce HCP preferably without affecting product stability, yield, or cost to ultimately maintain product quality and / or is amenable to large scale manufacturing and ensuring patient safety.

[0009] SUMMARY

[0010] It is an object of the present disclosure to provide a method for removing or reducing an impurity from a preparation comprising a molecule of interest and an impurity(ies) (e.g., endogenous proteins, such as HCP).

[0011] In particular, the present disclosure aims to provide a filtration method for reducing the content of HCP in a recombinantly produced protein preparation. It is a further objective of the present disclosure to provide a system for implementing such a process. Also, or alternatively, it is an object of the present disclosure to address one or more of the aforementioned problem(s). The disclosure may also solve further problems that will be apparent from the disclosure of the exemplary embodiments.

[0012] In a first aspect, the present disclosure relates to a filtration unit for counter-current filtration. In some embodiments, the filtration unit comprises a receptacle (e.g., a mixing unit, a feed tank), a pump, and a filter membrane. In some embodiments, the filtration unit further comprise a plurality of inlets, conduits, and a plurality of outlets. In some embodiments, the filtration unit comprises a device for regulating the re- circulation of retentate to the feed tank within the same unit. In some embodiments, the filtration unit comprises a device to generate a counter-current capable of re-circulating part of the fluid contained in the receptacle of said filtration unit to a receptacle of a preceding filtration unit.

[0013] In a second aspect, the present disclosure relates to a system for counter-current filtration comprises at least two, such as at least three, such as at least four, such as at least five filtration units according to the first aspect of the disclosure.

[0014] In some embodiments, the system is for removing or reducing an impurity in a protein preparation comprising a molecule of interest. In some embodiments, the system comprises a plurality of fluidly connected filtration units according to the first aspect of the disclosure including a first filtration unit and multiple subsequent filtration units, wherein each unit is configured to route received fluid across a membrane to provide a permeate and a retentate. The first filtration unit is capable of receiving a protein preparation as fluid at an inlet side from outside the system and the subsequent modules receive permeate from a preceding filtration unit as fluid and optionally fluid from a subsequent unit. At least one of the units, has / have a counter-current line or channel configured for returning part of the fluid contained in the receptacle of said unit back to the receptacle of a preceding filtration unit. In some embodiments, the filtration system exemplified above comprises a last module that has a permeate line or channel for recycling permeate to the inlet side of the first filtration module, e.g., as a wash fluid. Also, or alternatively, in some embodiments, the filtration system exemplified above has an additional line for wash liquid to wash out the system of any remaining molecule of interest and / or impurity. The permeate obtainable from the last filtration unit may be used as a wash liquid.

[0015] In a third aspect, the present disclosure relates to a method for removing or reducing an impurity from a protein preparation by subjecting the protein preparation a molecule of interest to a counter-current filtration, optionally using a system according to the second aspect of the disclosure. The method may be described as a multi-stage countercurrent filtration, where multi-stage refers to the presence of at least two filtration units, such as filtration units according to the first aspect of the disclosure. The method comprises a step of creating at least one counter-current by re-circulating part of the fluid contained in a receptacle of a (subsequent) filtration unit (e.g., such as, a second filtration unit or a third filtration unit or a fourth filtration unit) into a receptacle of a filtration unit preceding said filtration unit.

[0016] BRIEF DESCRIPTION OF DRAWINGS

[0017] The embodiments of the disclosure, together with its advantages, may be best understood from the following description taken in conjunction with the accompanying figures.

[0018] Fig. 1 shows an overlay of chromatograms of a protein preparation comprising polypeptide 1 (tR~ 11 min) and HOP (tR~ 15-18 min) (1) before filtration (“Unfiltered”), (2) after a first filtration step (“Filtered once”), and (3) after a second filtration step (“Filtered twice”) using UPLC method 1.

[0019] Fig. 2 shows an exemplary system for counter-current filtration system (1) according to the second aspect of the disclosure. The system shown in Fig. 2 comprises four filtration units (10, 20, 30, 40). The dotted square indicates an exemplary filtration unit for counter-current filtration according to the first aspect of the disclosure. The figure also shows the back flow (QB) and the transmission of the product (TR,P) as well as the transmission of the impurity (TR,I) as used in the “proof of concept calculation” described below.

[0020] Figs. 3A-D shows a plot of simulated / calculated retentate concentrations in four feed tanks (110, 210, 310, 410) of four filtration units (10, 20, 30, 40) making up the filtration system (1) of Fig. 2, as a function of dimensionless wash volume for a feed fluid comprising a molecule of interest (e.g., polypeptide 1) and HCP without a counter-current flow (y = 0). Fig. 3B and Fig. 3D are in semi-log plots.

[0021] Figs. 4A-D correspond to Figs. 3A-D but with an added counter-current flow (y = 0.25) from filtration unit 3 to filtration unit 2 and from filtration unit 2 to filtration unit 1 using a system as shown in Fig. 2.

[0022] Figs. 5A-D shows a series of chromatograms of a retentate and a permeate at different time points and from different filtration units (10, 20, 30, 40). Fig. 5A shows the time points 9:30 and 10:05. Fig. 5B shows the time points 10:55 and 11 :30. Fig. 5C shows the time points 12:15 and 13:00. Fig. 5B shows the time point 13:45.

[0023] Fig. 6 shows the integrated areas of the LIPLC peaks belonging to the molecule of interest and HCP, respectively, as a function of time.

[0024] Fig. 7 shows a chromatogram of a sample taken from a feed tank (110) from a first filtration unit (10) at the starting point of a batch process and a chromatogram of an HCP enriched sample taken from the feed tank (110) of the first filtration unit (10) at the end of the batch process and a product-enriched sample taken from the feed tank (410) of the last filtration unit (40) at the end of the batch process.

[0025] Fig. 8 shows an exemplary mass balance setup for a system run in counter-current mode, where “i” represents a filtration unit and where “i-1” represents a filtration unit which is upstream from a filtration unit “i" and where “i+1” represents a filtration unit downstream from the filtration unit “i". Put differently, filtration unit “i-1” is a preceding filtration unit of filtration unit “i” and filtration unit “i+1” and filtration unit “i” is a subsequent filtration unit of filtration unit “i-1” and a preceding filtration unit of filtration unit “i+1” and filtration unit “i+1” is a subsequent filtration unit of filtration unit “i” and filtration unit “i-1”.

[0026] Figs. 9A and 9B show two different types of exemplary filtration units according to the first aspect of the disclosure for use in a counter-current filtration process or method. In both figures the liquid is circulated over / through the filter membrane by a feed pump and the pressure on the filter is controlled by a back-pressure valve (241). Fig. 9A shows a setup with a feed tank and using a pump (251) to control the counter-current flow from one unit to the preceding unit. Fig. 9B shows a setup where the feed tank has been replaced by a mixing unit. Put differently, the streams are returned and mixed in the piping (e.g., a mixing unit 210). The setup shown in Fig. 9B comprises a flow measurement device (FC) for measuring the flow (in e.g. L / h) and a control valve (252) for adjusting the counter-current flow. The control valve can be opened or closed to match the desired flow (“set-point flow”). The counter-current flow can be controlled by a control valve utilising the pressure difference between the higher pressure on the outlet of the filter membrane and the lower pressure on the pumps suction side in the preceding unit.

[0027] Fig. 10A shows an exemplary system (1) for counter-current batch filtration according to the second aspect of the disclosure similar to the one shown in Fig. 2, but with an added receptacle (510) where the permeate from the last filtration unit (40) can be collected or recycled. In Fig. 10A, the intermediate filtration units (20, 30) are only indicated schematically for clarity. Figs. 10B-E show the individual filtration units.

[0028] The figures are schematic and simplified for clarity. Throughout, the same reference numerals are used for identical or corresponding parts.

[0029] DESCRIPTION

[0030] In the following detailed description, numerous specific details are set forth to provide a full understanding of the subject technology. It will be apparent to one of ordinary skill in the art that the subject technology may be practised without some of these specific details. Well- known structures and techniques may not be shown in detail so as to not obscure the subject technology.

[0031] Documents are cited throughout the text of this specification. Each of the documents cited herein (including all patents, patent applications, scientific publications, manufacturer’s specifications, instructions etc.), whether supra or infra, is hereby incorporated by reference in its entirety. In the event of a conflict between the definitions or teachings of such incorporated references and definitions or teachings recited in the present specification, the text of the present specification takes precedent.

[0032] In the following, the elements of the present disclosure will be described. These elements may be listed with specific elements, but it should be understood that they may be combined in any manner and in any number to create additional embodiments. The various described examples and preferred embodiments should not be construed to limit the present disclosure to only the explicitly described embodiments. This description should be understood to support and encompass embodiments which combine the explicitly described embodiments with any number of the disclosed and / or preferred elements. Furthermore, any permutations and combinations of all described elements in this application should be considered disclosed by the description of the present application unless the context indicates otherwise. To facilitate an understanding of the present subject technology, a number of terms and phrases are defined below. Unless defined otherwise herein, all technical and scientific terms used herein have the meaning commonly understood by one of ordinary skill in the art to which the disclosure pertains.

[0033] Disclosed herein are systems and methods for counter-current filtration of biologic products. More particularly, the filtration systems and methods operate to reduce impurity content (e.g., host cell proteins content, target protein aggregates) of a protein preparation comprising recombinantly produced molecule(s) of interest, such as, e.g., polypeptides. Such molecules of interest are typically intended for administration to a patient. The disclosed methods can be performed to produce therapeutic compositions having reduced host cell protein content.

[0034] Surprisingly, the methods described herein can achieve HCP counts, such as immunogenic HCP counts, well below the industry acceptable standard of < 100 ppm. Advantageously, the current method does not require the addition of chemicals and / or the use of additional chromatography steps to achieve an HCP count well below < 100 ppm. Additional purification methods may of course be combined with the method disclosed herein to further reduce the impurity content. For instance, the systems described herein may be connected to an ion exchanger such as, e.g., an anion exchanger.

[0035] Conveniently, the need for solid-liquid separation can be avoided because both the molecule of interest and the impurity(ies) are kept in solution. The use of a liquid protein preparation and the option of connecting several re-usable filtration units after each other allows for an efficient and fast separation and thereby facilitating a shorter overall processing time.

[0036] At the outset an experiment was set up to investigate if repeated filtrations of a mixture of molecule of interest and impurity could lead to further reductions of impurity content compared to a single filtration. This experiment is described under the heading “proof of concept” below. As can be seen in Fig. 1, the chromatogram of the unfiltered protein preparation shows a ratio of 0.87 between the impurity (HCP) and a molecule of interest (polypeptide 1). After one filtration through a filter membrane having a molecular cut-off weight of 30 kDa, the ratio is reduced to 0.18, which shows that the transmission of polypeptide 1 is higher than for HCP. As explained further below, HCP are a heterogenous and complex protein group produced by host cells, comprising both target proteins and proteins native to the host cell itself. HCP differ significantly in molecular mass, isoelectric points, hydrophobic properties, and structures. Hence, one could assume that a further filtration of the so-obtained material through a filter membrane having the same molecular cut-off weight of 30 kDa would not lead to a further reduction of HCP. However, taking into account the Renkin equation, it was hypothesized that a further filtration through a filter membrane having the same molecular cut-off weight of 30 kDa could still lead to a further reduction of HCP content. Indeed, as can be seen in Fig. 1 , “filtered twice”, the second filtration led to a further reduction of HCP content.

[0037] Taking the Renkin equation a step further, it was then hypothesised that adding a counter-current flow may improve the separation between the molecule(s) of interest and the impurity(ies) even further based on the assumption that both small and large molecules will diffuse through a membrane of an appropriate size eventually, but with a different transmission. It was envisioned that a counter-current flow could push the impurity towards the first filtration unit and push the molecule of interest to the last filtration unit. Conversely, without an added counter-current flow both the impurity(ies) and the molecules of interest would eventually end up in the last filtration unit (see Modelling Example B and Figs. 3a-d, Filter 4).

[0038] At the outset, the hypothesis was confirmed by a “proof-of-concept calculation” as set out below.

[0039] Proof of concept calculation

[0040] A tank filled with a protein preparation comprising two components with a concentration of 10 g / L was considered, where one component has a transmission through the filter set to 5% and the other component has a transmission through the filter set to 1%. The permeate flow through the filter was set to 10 L / h.

[0041] The transport of the two components (component 0 and component 1) through the filter can then be calculated according to equations (1) and (2): (1) (2)

[0042] A counter-current flow can then be added, which was set to 0.3 L / h resulting in equation (3). m0,left= mUeft= QB- CF= 0.3

[0043] The net transport of the components can then be calculated as shown in equations (4) and (5).

[0044] This leads to a net transport of 2 g / h of component 0 in one direction and a net transport of 2 g / h of component 1 in the opposite direction (see Fig. 2).

[0045] Definitions

[0046] Percentages, concentrations, amounts, and other numerical data may be expressed or presented herein in a "range” format. It is to be understood that such a range format is used merely for convenience and brevity and thus should be interpreted flexibly to include not only the numerical values explicitly recited as the limits of the range, but also to include all the individual numerical values or sub-ranges encompassed within that range as if each numerical value and sub-range is explicitly recited. As an illustration, a numerical range of "4% to 20%" should be interpreted to include not only the explicitly recited values of 4% to 20%, but to also include individual values and sub-ranges within the indicated range Thus, included in this numerical range are individual values such as 4, 5, 6, 7, 8, 9, 10, ... 18, 19, 20% and sub-ranges such as from 4-10 %, 5-15 %, 10-20%, etc. This same principle applies to ranges reciting minimal or maximal values. Furthermore, such an interpretation should apply regardless of the breadth of the range, or the characteristics being described.

[0047] With reference to HCP content, “ppm” is to be understood as ng(HCP) / mg(molecule of interest). It is to be noted that contrary to the concentration measurement of the molecule of interest, the HCP signal is only reflective of antibody binding and does not strictly reflect the mass of HCP.

[0048] Polypeptide 1 is an extended semaglutide precursor having a molecular weight of 4712 g / mol. Extended semaglutide precursors are described in WO 2015 / 091613. Polypeptide 1 may be prepared according to the methods described in WO 2015 / 091613 and references to methods disclosed therein. An example of a semaglutide precursor is [Arg34]GLP- 1(9-37) peptide. An example of an extended semaglutide precursor is [Arg34]GLP- 1(9-37) peptide fused to an N-terminal extension. Suitable N-terminal extensions are disclosed in WO 2015 / 091613.

[0049] Polypeptide 2 is an extended cagrilintide precursor having a molecular weight of 4555 g / mol. In the methods / processes and systems described herein, it should be understood that the inlet side refers to the retentate side of the filtration unit and the outlet side refers to the permeate side of a filtration unit.

[0050] If no specific temperature is indicated, then it should be assumed that standard room temperature applies. As used herein, “standard room temperature” refers to a temperature interval of 19 °C to 24 °C.

[0051] It must be noted that as used herein, the singular forms "a", "an", and "the", include plural references unless the context clearly indicates otherwise. Thus, for example, reference to "a filter" includes one or more of such different filters and reference to "the method" includes reference to equivalent steps and methods known to those of ordinary skill in the art that could be modified or substituted for the methods described herein. Unless otherwise indicated, the term "at least" preceding a series of elements is to be understood to refer to every element in the series. Those skilled in the art will recognise or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the disclosure described herein. Such equivalents are intended to be encompassed by the present disclosure.

[0052] The term “and / or” wherever used herein includes the meaning of "and", "or" and "all or any other combination of the elements connected by said term.

[0053] The term "comprise", and variations thereof such as "comprises" and "comprising", will be understood to imply the inclusion of a stated integer or element or step or group of integer or element or step, but not the exclusion of any other integer or element or step or group of integer or element or step. When used herein the term “comprise” can be substituted with the terms “contain” or “include” or sometimes when “have”.

[0054] The term “obtainable” can be used interchangeably with “obtained”.

[0055] The term "about" is used herein to mean approximately, roughly, or around. When the term "about" is used in conjunction with a numerical range, it modifies that range by extending the boundaries above and below the numerical values set forth. In general, the term "about" can modify a numerical value above and below the stated value by 20 percent, up or down (higher or lower), preferably 15 percent, up or down (higher or lower), more preferably within 10 percent, up or down (higher or lower), and most preferably by 5 percent, up or down (higher or lower).

[0056] The term “counter-current” indicates the direction of a flow and refers to the flow of a fluid that flows in opposite direction compared to a fluid that flows in concurrent flow. With reference to Fig. 2 and Figs. 10A-E, fluid moving to the right, i.e. moving from feed tank (110) towards feed tank (410) flows concurrently while fluid moving to the left, i.e. moving from feed tank (410) towards feed tank (110) flows counter-currently. More specifically, herein the term “counter-current” specifically refers to the feed flow that is created by back-feeding fluid from a subsequent filtration unit into a preceding filtration unit, such as, e.g., back-fed fluid from filtration tank (310) to filtration tank (210). The skilled person will also understand that this also covers back-feeding fluid from, e.g. filtration tank (310) directly to filtration tank (110) as long as the fluid moves counter-currently from a downstream filtration unit (“subsequent” filtration unit) to a filtration unit further upstream (“preceding” filtration unit). Again, with reference to Fig. 2 and Figs. 10A-E, even though the retentate is re-circulated to feed tank (210) via conduit (231) in a counter-current direction, the retentate does not flow from a subsequent filtration unit to a preceding filtration unit but moves within the same filtration. Hence, the retentate flow is not considered a counter-current flow herein. As the skilled person will appreciate the retentate flow (“QR”) is driven by the same pump as the (feed) fluid flow (“QF”) and not a separate flow contrary to the counter-current flow (“QB”).

[0057] The term “concurrent flow” as used herein refers to the flow of a fluid that flows in opposite direction compared to a fluid that flows in counter-current flow. For instance, taking Fig. 2 or Fig. 10A and Fig. 10B and Fig. 10E, a fluid that flows concurrently moves from the left side towards the right side of the figure, e.g., from feed tank (110) towards feed tank (410).

[0058] The expression “enriched in impurity” means that the ratio between impurity and molecule of interest increased after passing the fluid over a filter membrane. For instance, if the ratio between impurity and molecule of interest was 2:3 before passing the fluid over a filter membrane and the ratio after filtration is 2:2 in the retentate then the retentate is enriched in impurity (compared to the fluid before filtration).

[0059] The expression “enriched in molecule of interest” means that the ratio between molecule of interest and impurity increased after passing the fluid over a filter membrane. For instance, if the ratio between molecule of interest and impurity was 2:2 before passing the fluid over a filter membrane and the ratio after filtration is 2:1 in the permeate then the permeate is enriched in molecule of interest (compared to the fluid before filtration).

[0060] The term “fluid” as used herein generally refers to the fluid comprising the molecule of interest and the impurity(ies). Put differently, the protein preparation comprising the molecule of interest that is subjected to a method of removing or reducing an impurity and introduced as a fluid into a first filtration unit. The fluid may have been centrifuged to remove cells, such as e.g., yeast cells, before entering the filtration system described herein. The pH of the fluid may also have been adjusted before subjecting the feed to the filtration process described herein. Fluid is used in a general sense, and unless indicated otherwise in a particular context, can encompass liquid materials containing dispersed and / or solubilised species, pure liquids, or other flowable materials. Fluid is also used to refer to the content in a receptacle of a filtration unit, because skilled person will understand from the context that e.g. the permeate from e.g. a filtration unit “i-1” will be fed into e.g. the receptacle of the subsequent filtration unit “i” where it may also mix with part of the fluid that is fed back from e.g. a further subsequent filtration unit “i+1”. Looking at, e.g., Fig. 5, the skilled person will understand the concentration of the impurity and the molecule of interest will vary in the various receptables / filtration units during the course of the filtration. At the end of the filtration, most of the molecule of interest will be in the last filtration unit while most of the impurity will be in the first filtration unit.

[0061] The term “protein preparation” or “preparation” as used herein refers to a mixture comprising recombinantly produced molecule(s) of interest, such as, e.g., polypeptides. Such protein preparation may have been centrifuged before subjecting such a protein preparation as feed fluid to a method as disclosed herein.

[0062] The term “protein” as used herein includes e.g., peptides and polypeptides.

[0063] The term “flux” or “overall flux” is the total volumetric rate of permeation divided by the total area of membrane in operation, usually expressed in litres per square meter per hour, e.g. Iitres / m2 / hour (LMH).

[0064] “Fouling” should be understood to mean obstruction of pores in a membrane by a gel layer, by a cell cake or cell paste, or by internal binding of molecules to the membrane pores.

[0065] The term “fermentation broth” as used herein refers to fluid produced by culture or fermentation of biological organism, such a bacteria, fungi, mammalian, insect, or plant cells. The fermentation broth contains inter alia a molecule of interest and impurities. The fermentation broth may have been centrifuged to remove most of the cell debris and / or filtered before subjecting the fermentation broth in form of a protein preparation to the filtration process described herein.

[0066] The term “contaminant” or “impurity” as used herein refers to any foreign or undesirable molecule that is present in the feed stream at the beginning of the filtration process. There may be “process impurities” present. These are impurities that are present as a result of the process in which the molecule of interest was produced. For instance, impurity includes, without limitations, DNA, RNA, host cell materials, such as host cell proteins (HCP), CHOP, leached Protein A, nucleic acids, a variant, size variant, fragment, aggregate or derivative of the molecule of interest, endotoxin, cell culture media component(s), etc.

[0067] A “host cell” refers to cells stably or transiently transfected, transformed, transduced or infected with one or more expression vectors expressing one or more protein of the present disclosure. Creation and isolation of host cell lines producing proteins of the present disclosure can be accomplished using standard techniques known in the art.

[0068] The expression “host cell proteins” refers to proteins of the host cells that are involved in cell maintenance and growth, and protein synthesis and processing. HCP are identified during the manufacturing of biopharmaceuticals as part of the quality control process. Enzyme linked immunosorbent assay (ELISA) is the predominant method for HCP analysis in pharmaceutical products due to its high sensitivity to proteins, which allows it to detect the low levels of HCP in produced drugs. An exemplary enzyme linked immunosorbent assay is described in the “examples” section (see “HCP quantification using ELISA”). Methods such as the combination of mass spectrometry (MS) and liquid chromatography (LC-MS) may also be used to allow for more efficient and effective HCP analysis and purification. Certain HCP have been associated with immunogenicity concerns in patients and there is a desire by regulators to reduce HCP in order to minimize immunogenicity concerns. An example of a host cell protein is a protease, which can cause damage to the protein of interest if still present during and after purification. For example, if a protease remains in the sample comprising the protein of interest, it can create product-related substances or impurities which were not originally present. The presence of proteases can cause decay, e.g., fragmentation, of the protein of interest over time during the purification process, and / or in the final formulation. In some embodiments, the HCP are produced / derived from a mammalian cell, a bacterial cell, or a yeast cell. Examples of yeast cells are komagataella pastoris, Asperigillus sp., Saccharomyces cerevisiae, Schizosaccharomyces pombe, and Neurospora crassa.

[0069] The term "measurement", "measuring" or "determining" preferably comprises a qualitative, a semi-quantitative or a quantitative measurement.

[0070] The term “filter membrane” refers to a semi-permeable layer of material for separating particles or molecules from a fluid based on their size and / or other physical properties.

[0071] The term “molecule of interest” refers to a recombinantly produced molecule to be purified, such as, e.g., a polypeptide, a protein. The “molecular weight cut-off (MWCO)” as used herein refers to the size (kilodaltons) designation for the ultra or nanofiltration membrane and may be calculated using US standard ASTM E1343-90 (“Standard test method for molecular weight cut-off evaluation of flat sheet ultrafiltration membranes"). Typically, the MWCO is defined as the molecular weight of the globular protein that is 90% retained by the membrane.

[0072] A “filter membrane having medium pores” as used herein means that the separation factor is above 1.2, such as above 1.5, such as above 2.0, such as above 2.5. Preferably the separation factor is in the range of 2-100. A filter membrane having medium pores should be selected such that the transmission of the product is higher than the transmission of the molecule of interest.

[0073] A “filter membrane having large pores” as used herein means that the transmission for both, the molecule of interest and the impurity is about 1. A filter membrane having large pores should essentially let both the molecule of interest and the impurity pass through and only function as a first filtration step to e.g. reduce the turbidity of the feed fluid.

[0074] A “filter membrane having small pores” as used herein means that the transmission for both, the molecule of interest and the impurity is close to 0. A filter membrane having small pores should essentially retain both the molecule of interest and the impurity.

[0075] “Nephelometric Turbidity Units” (NTU) is a measure of the turbidity or cloudiness of a liquid sample. It is used to quantify the amount of light that is scattered by particles suspended in the liquid. The higher the concentration of suspended particles, the higher the turbidity reading. NTU is typically measured using a nephelometer, an instrument that measures the amount of scattered light at a specific angle (usually 90 degrees) from a light source. The amount of light scattered is proportional to the concentration of suspended particles in the sample. The nephelometer converts the amount of scattered light into an NTU reading, which is displayed on the instrument's screen.

[0076] The term “permeate” as used herein refers to the portion of the fluid stream that passes through the filter membrane. The skilled person will understand that once the permeate will enters into a subsequent filtration unit it will mix with fluid of said filtration unit and hence be referred to as fluid.

[0077] A “pump” is a device that moves fluid(s) by mechanical action from one location, a first location, along the flow path to another location; a second location, along the flow path positionally different from the first location. The term “recombinantly produced molecule” as used herein refers to a molecule, such as, e.g. a polypeptide, which was produced by a host organism, such as, e.g. a yeast organism.

[0078] The term “retentate” as used herein refers to the portion of the fluid stream that does not pass through the filter membrane. Retentate can be re-circulated and is the mixture which stays on the feed side (inlet side) of the membrane. The skilled person will understand that for clarity the retentate will be referred to as fluid once it is re-circulated into the receptacle of the filtration unit.

[0079] The term “net permeation rate” is the flow rate of permeate leaving the system, typically expressed in litres per minute. The next permeation rate is a measure of the system throughput or capacity.

[0080] “Product Purity” or “Purity” is the degree of isolation of the product in the product stream. It can be understood to mean the amount of desired compound isolated compared to the sum amount of the other components in the stream and can be expressed as a weight percentage. Alternatively, it can be understood to mean the ratio of the concentration of the product relative to that of another selected component in the product stream and can be expressed as a number having the units of concentration divided by concentration. In various embodiments, purity is measured directly or indirectly instrumentally or manually, such as by determination of enzymatic activity (e.g., as determined colourimetrically); and or by product colour determination by absorbance measurement, CIELAB formula, or US Pharmacopeia (USP) Monographs, and so forth to measure product colour; and or by impurity level measurement (e.g., measurement of microbial impurities in fresh product or as part of shelf life studies); and or total protein content or other product component; and or organoleptically by odour, taste, texture, visual colour, and so forth (e.g., in fresh product or as part of shelf life studies).

[0081] The term “receptacle” relates to a container, mixing unit, tank or any means that can hold and / or mix fluid.

[0082] The terms “polypeptide” and “protein” (and / or their respective plural forms), as used herein, are used interchangeably to refer to polymers of any length comprising amino acid residues linked by peptide bonds. The conventional one-letter or three-letter codes for amino acid residues are used herein. The polymer can be linear or branched. It can comprise modified amino acids, and it can be interrupted by non-amino acids. The terms also encompass an amino acid polymer that has been modified naturally or by intervention: for example, disulfide bond formation, glycosylation, lipidation. acetylation, phosphorylation, or any other manipulation or modification, such as conjugation with a labelling component. Also included within the definition are, for example, polypeptides containing one or more analogs of an amino acid (including, for example, unnatural ammo acids, etc,), as well as other modifications known in the art.

[0083] The term “ultrafiltration” relates to a form of membrane filtration in which hydrostatic pressure forces a liquid against a semipermeable membrane. Suspended solids and solutes of high molecular weight are retained, while water and low molecular weight solutes pass through the membrane. In some examples, ultrafiltration membranes have pore sizes in the range of 1 pm to 100 pm. The terms "ultrafiltration membrane" "ultrafiltration filter" "filtration membrane" and "filtration filter" may be used interchangeably. Examples of filtration membranes include but are not limited to polyvinylidene difluoride (PVDF), polysulfone, polyethersulfone, polyarylsulfone, regenerated cellulose, polyamide, polypropylene, polyethylene, polytetrafluoroethylene (PTFE), cellulose acetate, polyacrylonitrile, vinyl copolymer, polyamides, polycarbonates, or blends thereof or the like. In some embodiments, the filter membrane comprises or consists of ceramics, such as alumina, zirconia, titanium oxide, silicon carbide or other filter materials suitable for use in a cGMP manufacturing environment.

[0084] “Product Yield” or “Yield” is the total amount of product collected in the product stream, usually expressed as a percentage of the total amount in the feed stream. In various embodiments, one can obtain a high yield of a high purity product. In other embodiments, one can obtain a high yield of a product, however, the product may have a lower level of purity, for example, as in producing a crude extract from the retentate side of the filtration module.

[0085] The term "vibrating filter-plate assembly device" as used herein refers to the vibration of the filter-plate assembly device during filtration operation to create a dynamic cross-flow filtration or dynamic shear-enhanced filtration. The vibrating motion of the device of the invention increases permeate flux or membrane selectivity or both.

[0086] The term "vibrating motion" as used herein means movement (vibrating movement) of all of the filter surface in relation to the surrounding liquid with a frequency. Thus, the inertia of the liquid overcomes the drag of the surface resistance.

[0087] The term “protein preparation” refers to a mixture comprising a molecule of interest and an impurity. The protein preparation may be introduced as a fluid into the first filtration unit. The terms “pipe”, “conduit”, “tube”, and “channel” are used interchangeably.

[0088] According to various embodiments, methods are provided that refer to processes or actions involved in sample preparation or other procedures. It will be understood that in various embodiments a method or process can be performed in the order of processes as presented, however, in related embodiments the order can be altered as deemed appropriate by one of skill in the art in order to accomplish a desired result.

[0089] Filtration unit

[0090] In a first aspect, the disclosure relates to a filtration unit for counter-current filtration.

[0091] A filtration unit comprises a receptacle, a filter membrane, a pump, a plurality of inlets, a plurality of outlets, and a plurality of channels. In some embodiments, the receptacle is a feed tank. In some embodiments, the receptacle is omitted and the fluids (e.g., feed fluid, back-fed fluid, retentate, etc.) are mixed in a pipe or conduit, optionally by a static mixer. In some embodiments, the channels are conduits. In some embodiments, the channels are tubes. In some embodiments, the channels are pipes. In some embodiments, the pump is a feed pump. In some embodiments, the pump or feed pump is controlled (either mechanically or through automation) to operate the same flow rates. The individual pumps (for the same purpose e.g. for pumping the feed fluid, or alternatively, for pumping the counter-current feed fluid) of each filtration unit will have closely-matched pumping rates. In some embodiments, the pump is a membrane pump.

[0092] The filtration unit further comprises a device for creating and / or controlling a counter-current flow. In some embodiments, the device for creating and / or controlling a counter-current flow is a pump or a control valve. In some embodiments, the device for creating and / or controlling a counter-current flow is a positive displacement pump.

[0093] The dotted line in Fig. 2 indicates an exemplary filtration unit (20). As shown in Fig. 10C, the exemplary filtration unit (20) comprises a feed tank (210), a filter membrane (220), and a feed pump (230). The feed tank (210) comprises an inlet (211) such that feed fluid from a subsequent filtration unit can enter the feed tank (210), an inlet (213) such that countercurrent feed fluid can enter the feed tank (210), and an inlet (214) such that retentate can enter the feed tank (210). The feed tank (210) further comprises a feed outlet (221). The filter membrane (220) comprises an inlet (212) on a first side of the membrane. The filter membrane (220) further comprises a retentate outlet (223) on a first side of the membrane and a permeate outlet (222) on the other side of the membrane. The filtration unit (20) further comprises a recirculation channel (231) for circulating a retentate from a first or entrance side of the membrane (220) to the feed tank (210). The filtration unit (20) also comprises a feed channel (232) for transporting the feed fluid from the outlet of the feed tank (210) to the inlet of the filter membrane (220). The filtration unit (20) further comprises a permeate channel

[0094] (233) for transporting the permeate from the other side of the filter membrane (220) to the subsequent filtration unit. The filtration unit further comprises a counter-current feed channel

[0095] (234) for transporting the feed fluid from a subsequent filtration unit back to the feed tank (210).

[0096] In some embodiments, the filter membrane is an ultrafiltration membrane. In some embodiments, the filter membrane is a TFF membrane. In some embodiments, the TFF membrane is selected from a hollow fibre membrane, a cassette membrane, a ceramics membrane, and a rotating disc membrane. In some embodiments, the membrane can have a MWCO in the range of 1-500 kDa, 1-200 kDa, 1-150 kDa, 1-100 kDa, such as 1-95 kDa, 1- 90 kDa, such as 1-85 kDa, such as 1-80 kDa, such as 1-75 kDa, such as 1-70 kDa, such as

[0097] 1-65 kDa, such as 1-60 kDa, such as 1-55 kDa, such as 1-50 kDa, such as 1-45 kDa, such as 1-40 kDa, such as 1-35 kDa, such as 1-30 kDa, such as 1-25 kDa, such as 1-20 kDa, such as 1-15 kDa, such as 1-10 kDa. In some embodiments, the membrane can have a MWCO in the range of 2-100 kDa, such as 2-95 kDa, 2-90 kDa, such as 2-85 kDa, such as

[0098] 2-80 kDa, such as 2-75 kDa, such as 2-70 kDa, such as 2-65 kDa, such as 2-60 kDa, such as 2-55 kDa, such as 2-50 kDa, such as 2-45 kDa, such as 2-40 kDa, such as 2-35 kDa, such as 2-30 kDa, such as 2-25 kDa, such as 2-20 kDa, such as 2-15 kDa, such as 2-10 kDa. In some embodiments, the membrane can have a MWCO in the range of 3-100 kDa, such as 3-95 kDa, 3-90 kDa, such as 3-85 kDa, such as 3-80 kDa, such as 3-75 kDa, such as 3-70 kDa, such as 3-65 kDa, such as 3-60 kDa, such as 3-55 kDa, such as 3-50 kDa, such as 3-45 kDa, such as 3-40 kDa, such as 3-35 kDa, such as 3-30 kDa, such as 3-25 kDa, such as 3-20 kDa, such as 3-15 kDa, such as 3-10 kDa. In some embodiments, the membrane can have a MWCO in the range of 4-100 kDa, such as 4-95 kDa, 4-90 kDa, such as 4-85 kDa, such as 4-80 kDa, such as 4-75 kDa, such as 4-70 kDa, such as 4-65 kDa, such as 4-60 kDa, such as 4-55 kDa, such as 4-50 kDa, such as 4-45 kDa, such as 4-40 kDa, such as 4-35 kDa, such as 4-30 kDa, such as 4-25 kDa, such as 4-20 kDa, such as 4- 15 kDa, such as 4-10 kDa. In some embodiments, the membrane can have a MWCO in the range of 5-100 kDa, such as 5-95 kDa, 5-90 kDa, such as 5-85 kDa, such as 5-80 kDa, such as 5-75 kDa, such as 5-70 kDa, such as 5-65 kDa, such as 5-60 kDa, such as 5-55 kDa, such as 5-50 kDa, such as 5-45 kDa, such as 5-40 kDa, such as 5-35 kDa, such as 5-30 kDa, such as 5-25 kDa, such as 5-20 kDa, such as 5-15 kDa, such as 5-10 kDa. In some embodiments, the membrane can have a MWCO in the range of 10-100 kDa, such as 10-95 kDa, 10-90 kDa, such as 10-85 kDa, such as 10-80 kDa, such as 10-75 kDa, such as 10-70 kDa, such as 10-65 kDa, such as 10-60 kDa, such as 10-55 kDa, such as 10-50 kDa, such as 10-45 kDa, such as 10-40 kDa, such as 10-35 kDa, such as 10-30 kDa, such as 10-25 kDa, such as 10-20 kDa, such as 10-15 kDa. In some embodiments, each filter membrane can have a MWCO selected from 1 kDa, 5 kDa, 10 kDa, 20 kDa, 30 kDa, 50 kDa, 100 kDa, and 150 kDa. In some embodiments, each filter membrane can have a MWCO in the range of 5 kDa to 150 kDa, such as from 10 kDa to 150 kDa, such as from 15 to 150 kDa. In some embodiments, each filter membrane can have a MWCO in the range of 5 kDa to 150 kDa, such as from 5 kDa to 130 kDa, such as from 5 to 120 kDa, such as from 5 to 110 kDa, such as from 5 to 100 kDa.

[0099] In some embodiment, the filtration unit comprises a temperature controlling device for heating and / or cooling as shown, in e.g., in Fig. 2. In some embodiments, the feed tank and / or the membrane can be heated and / or cooled. In some embodiments, at least one of the feed tank is kept at a temperature in the range of 15 °C to 35 °C. In some embodiments, at least one of the feed tank is cooled to a temperature in the range of -5 °C to 15 °C, such as 0 °C to 10 °C. In some embodiments, at least one of the feed tanks is heated to a temperature in the range of 35 °C to 85 °C, such as in the range of 40 °C to 50 °C. Conveniently, at least the first feed tank may be heated. Advantageously, the last feed tank may be cooled. In some embodiments, at least the first feed tank may be heated, and the last feed tank may be cooled. In some embodiments, at least the first feed tank may be kept at a temperature in the range of 40 °C to 50 °C and / or the last feed tank may be kept at a temperature in the range of 0 °C to 10 °C.

[0100] In some embodiments, the filter membrane comprises or consists of a material selected from the group consisting of polyvinylidene difluoride (PVDF), polysulfone, polyethersulfone, polyarylsulfone, regenerated cellulose, polyamide, polypropylene, polyethylene, polytetrafluoroethylene (PTFE), cellulose acetate, polyacrylonitrile, vinyl copolymer, polyamides, polycarbonates, or blends thereof or the like. In some embodiments, the filter membrane comprises or consists of ceramics, such as alumina, zirconia, titanium oxide, silicon carbide.

[0101] In some embodiments, the filter membrane is a cross flow filtration unit, wherein the cross flow filtration unit provides a filter plate for filtration of fluids, the plane having a membrane, which is fluid tight bonded at its edges to the surface of a partly hollow supporting plate comprising exit openings, internal flow channels, and flow areas for a first liquid medium, and the membrane being in fluid contact with said first liquid medium at its internal surface and being in fluid contact with a second liquid medium at its external surface. Conveniently, the membrane may be selected such that it is based on the cross-flow filtration principle, but the turbulence at the membrane surface is created by vibrating the membrane relative to the feed instead of applying a fast cross flow allowing the transmembrane pressure (TMP) to become close to uniform throughout the system. Such membrane may be selected from the membranes disclosed in WO 2018 / 145714. In particular embodiments, the cross-flow filtration unit is selected from a cross flow filtration unit described in WO 2018 / 145714. In some embodiments, the / each filter membrane is a vibrating filter-plate assembly device comprising a vessel housing for a filter-plate assembly and one or more flexible volume chamber(s) being filled with gas, where the flexible volume chamber(s) is / are adapted to expand and / or compress the volume(s) of the flexible volume chamber(s) inside the vessel housing allowing the liquid in the vessel chamber to move across the surface of said filter-plates when said vessel housing comprising said filter-plate assembly is subjected to a vibrating motion, wherein the vibrating device comprises at least one flexible support or suspension, where the vessel housing is supported by said at least one flexible support allowing vibrating motion of the vessel housing. Advantageously, using a vibrating filter-plate assembly can reduce build-up and fouling on the membrane.

[0102] Filtration system

[0103] In a second aspect, the disclosure relates to a filtration system for counter-current filtration of a protein preparation comprising an impurity and a molecule of interest. The filtration system comprises a plurality, e.g., at least two filtration units according to the first aspect of the disclosure. Conveniently, the plurality of filtration units may be fluidly connected such that fluid can move serially from the first to subsequent filtration units until the fluid reaches the last filtration stage, and also that the counter-current fluid moves counter-currently from a later or subsequent to an earlier or preceding filtration unit.

[0104] In some embodiments, the system comprises three or more filtration units. In some embodiments, the system comprises four or more filtration units. In some embodiments, the system comprises five or more filtration units. In some embodiments, the system comprises six or more filtration units. In some embodiments, the system comprises seven or more filtration units. Generally, there is no particular limit on the number of filtration units that could be used in the system. However, connecting more than six filtration units may not provide any significant further advantage in terms of cost and / or separation efficiency.

[0105] Conveniently, the first filtration unit is configured to receive a protein preparation as fluid from outside of the system via an inlet (111). In some embodiments, the filtration system exemplified above comprises a permeate channel connectable to the last filtration unit for recycling permeate to the inlet side of the first filtration unit. Also, or alternatively, in some embodiments, the filtration system exemplified above has an additional line for wash liquid to wash out the system of any remaining molecule of interest and / or impurity.

[0106] Each filtration unit may comprise a filter membrane having the same or a different molecular cut-off weight as exemplified under the heading “filtration unit”. In preferred embodiments, the last filtration unit comprises a filter membrane having a smaller molecular weight cut-off compared to the molecular cut-off weight of the filter membrane of the preceding filtration units, e.g. 5 kDa (last filtration unit) and 30 kDa (preceding filtration units). Conveniently, the filtration system may comprise a pre-filter having a molecular cut-off weight that is significantly larger than the molecular cut-off weight of filter membranes of the respective filtration units.

[0107] An exemplary system (1) is shown in Figure 2 and Figure 10A. Referring to Fig. 2 and Fig 10A, the system (1) comprises four filtration units (10, 20, 30, 40) according to the first aspect of the disclosure. As can be seen in Fig. 2 and Fig. 10A, the first filtration unit (10) and the fourth filtration unit (40) do not comprise a counter-current channel (134, 434) for transporting the fluid from the receptacle (310, 210) of a subsequent filtration unit to a receptacle (210, 110) of the preceding filtration unit. The third receptacle (310) and the fourth receptacle (410) do not comprise an inlet for receiving a counter-current fluid from a subsequent filtration unit. Sending fluid from the last filtration unit back to the preceding filtration unit would defeat the purpose of achieving a good separation since the last filtration unit comprises the purified molecule of interest and back-feeding part of the fluid would reduce the overall process yield.

[0108] In particular, as can be seen in e.g., Fig. 10B, the first filtration unit (10) comprises a feed tank (110), a filter membrane (120), and a feed pump (130). The feed tank (110) comprises an inlet (111) such that fluid from outside the system can enter the feed tank (110), an inlet (113) such that counter-current fluid can enter the feed tank (110), and an inlet (114) such that retentate can enter the feed tank (110). The feed tank (110) further comprises an outlet (121). The filter membrane (120) comprises an inlet (112) on a first side of the membrane. The filter membrane (120) further comprises a retentate outlet (123) on a first side of the membrane and a permeate outlet (122) on the other side of the membrane. The filtration unit (20) further comprises a re-circulation channel (131) for circulating a retentate from a first or entrance side of the membrane (120) to the feed tank (110). The filtration unit (10) also comprises a feed channel (132) for transporting the feed fluid from the outlet of the feed tank (110) to the inlet of the filter membrane (220). The filtration unit (10) further comprises a permeate channel (133) for transporting the permeate from the other side of the filter membrane (120) to the subsequent filtration unit. The first filtration unit (10) is fluidly connected to a second filtration unit (20) via the permeate channel (133) for transporting the permeate from the other side of the filter membrane (120) to the second filtration unit (20). As can be seen in e.g. Fig. 10C, the second filtration unit (20) comprises a feed tank (210), a filter membrane (220), and a feed pump (230). The feed tank (210) comprises an inlet (211) such that fluid from a subsequent filtration unit can enter the feed tank (210), an inlet (213) such that counter-current fluid can enter the feed tank (210), and an inlet (214) such that retentate can enter the feed tank (210). The feed tank (210) further comprises a feed outlet (221). The filter membrane (220) comprises an inlet (212) on a first side of the membrane. The filter membrane (220) further comprises a retentate outlet (223) on a first side of the membrane and a permeate outlet (222) on the other side of the membrane. The filtration unit (20) further comprises a re-circulation channel (231) for circulating a retentate from a first or entrance side of the membrane (220) to the feed tank (210). The filtration unit (20) also comprises a feed channel (232) for transporting the fluid from the outlet of the feed tank (210) to the inlet of the filter membrane (220). The filtration unit (20) further comprises a permeate channel (332) for transporting the permeate from the other side of the filter membrane (220) to the subsequent filtration unit. The filtration unit further comprises a counter-current feed channel (234) for transporting the fluid back to the feed tank (110) of the first filtration unit. The second filtration unit (20) is fluidly connected to a third filtration unit (30) via the permeate channel (233) for transporting the permeate from the other side of the filter membrane (220) to the third filtration unit (30). As can be seen in e.g., Fig. 10D, the third filtration unit comprises a feed tank (310), a filter membrane (320), and a feed pump (330). The feed tank (310) comprises an inlet (311) such that feed fluid from a subsequent filtration unit can enter the feed tank (310), and an inlet (314) such that retentate can enter the feed tank (310). The feed tank (310) further comprises a feed outlet (321). The filter membrane (320) comprises an inlet (312) on a first side of the membrane. The filter membrane (320) further comprises a retentate outlet (323) on a first side of the membrane and a permeate outlet (322) on the other side of the membrane. The filtration unit (30) further comprises a recirculation channel (331) for circulating a retentate from a first or entrance side of the membrane (320) to the feed tank (310). The filtration unit (30) also comprises a feed channel (232) for transporting the fluid from the outlet of the feed tank (310) to the inlet of the filter membrane (320). The filtration unit (30) further comprises a permeate channel (333) for transporting the permeate from the other side of the filter membrane (320) to the subsequent filtration unit. The filtration unit further comprises a counter-current feed channel (334) for transporting the fluid back to the feed tank (210) of the second filtration unit. The third filtration unit (30) is fluidly connected to a fourth filtration unit (40) via the permeate channel (433) for transporting the permeate from the other side of the filter membrane (320) to the fourth filtration unit (40). As can be seen in e.g., Fig. 2 and Fig. 10E (where an additional feed tank (510) is shown), the fourth filtration unit comprises a feed tank (410), a filter membrane (420), and a feed pump (430). The feed tank (410) comprises an inlet (411) such that fluid from a subsequent filtration unit can enter the feed tank (410), and an inlet (414) such that retentate can enter the feed tank (410). The feed tank (410) further comprises a feed outlet (421). The filter membrane (420) comprises an inlet (412) on a first side of the membrane. The filter membrane (420) further comprises a retentate outlet (423) on a first side of the membrane and a permeate outlet (422) on the other side of the membrane. The filtration unit (40) further comprises a recirculation channel (431) for circulating a retentate from a first or entrance side of the membrane (420) to the feed tank (410). The filtration unit (40) also comprises a feed channel (432) for transporting the fluid from the outlet of the feed tank (410) to the inlet of the filter membrane (420). The filtration unit (40) further comprises a permeate channel (433) for transporting the permeate from the other side of the filter membrane (420) to the first filtration unit (10). Also, or alternatively to the last filter membrane a chromatographic column may be used such as a reversed phase column or an ion exchanger may be used. In some embodiments, the ion exchanger is a cation exchanger. In some embodiments, the ion exchanger is an anion exchanger. Also, or alternatively to the last filter membrane a hydrophobic interaction chromatography column may be used. Also, or alternatively to the last filter membrane an affinity column may be used. Also, or alternatively to the last filter membrane a size-exclusion or gel-filtration column may be used.

[0109] It will be understood, e.g., by looking at Fig. 2 or Fig. 10A that the fluid in the respective receptacles may be a mixture of retentate and counter-current fluid or retentate, permeate, and counter-current fluid, or permeate and retentate.

[0110] Methods for reducing or removing an impurity

[0111] In a third aspect, the disclosure relates to a method for removing or reducing an impurity from a protein preparation comprising a molecule of interest recombinantly produced in a host cell by subjecting the protein preparation to a counter-current filtration, optionally using the system according to the second aspect of the disclosure. Conveniently, the method separates the protein preparation which is a fluid into a permeate portion and a retentate portion when passed across / through a filter membrane, wherein the permeate portion is typically enriched in molecule of interest and the retentate portion is enriched in impurity. It will be understood from the context that the permeate / retentate is enriched in molecule of interest / impurity compared to the respective concentration of molecule of interest / impurity before being passed through / across the filter membrane. It will also be understood that this typically will not apply to the last filtration unit which comprises a filter membrane having a molecular cut-off weight that keeps the molecule of interest at the retentate side of the filter membrane. The receptacle of the last filtration may be cooled to decrease the transmission of the molecule of interest through the filter membrane.

[0112] In particular embodiments, the method is for removing or reducing HCP from / in a protein preparation comprising a molecule of interest by subjecting the protein preparation as a fluid to a counter-current filtration. The pH of the fluid may be adjusted before subjecting the fluid to the method as described herein. The pH of the fluid may be adjusted to a pH in the range of about 3-10, such as about 4-9, such as about 5-9, such as about 6-9, such as about 7-9, such as about 7.0-8.9. The pH of the fluid may be adjusted to a pH in the range of 7.2-7.7. The pH of the fluid may be adjusted to a pH of about 7.4. The pH of the fluid may be adjusted to a pH in the range of 8-9. The pH of the fluid may be adjusted to a pH of about 8.5. The fluid may also have been subjected to a pre-filtration and / or pre-concentration step to reduce NTU and / or increase the initial concentration. In some embodiments, the feed fluid will pass several filtration units, such as at least two, at least three, at least four, or at least five filtration units before the fluid comprising the molecule of interest will be collected from the feed tank of the last filtration unit. Conveniently, the amount of impurity in the last filtration unit may be below 100 ppm. In some embodiments, fluid of at least one subsequent filtration unit, such as a second filtration unit, re-circulates / is back-fed to a preceding filtration unit, such as a first filtration unit, by passing part of the fluid from the receptacle of the subsequent filtration unit to the receptacle of the preceding filtration unit via a counter-current channel.

[0113] In some embodiments, the method is for removing or reducing an impurity, the method comprising the steps of: a. introducing protein preparation comprising a molecule of interest and an impurity as a fluid into a first filtration unit of a filtration system (1), wherein the filtration system comprises at least two filtration units (20, 30), wherein each filtration unit comprises a filter membrane (120, 130); b. passing the fluid through the filter membrane (120) of the first filtration unit (20) to obtain a retentate and a permeate; c. re-circulating the so-obtained retentate to the first filtration unit (20), and passing the so-obtained permeate to the second filtration unit (30) as a fluid and passing the fluid through the filter membrane (220) to obtain a retentate and a permeate; and d. creating at least one counter-current flow by passing a portion of the feed fluid from the subsequent filtration unit (30) to the preceding filtration unit (20); and e. optionally recovering the permeate obtainable from the subsequent filtration unit (30).

[0114] In some embodiments, the method for removing or reducing an impurity from / in a protein preparation comprises the steps of: a. introducing a protein preparation comprising a molecule of interest and an impurity as a fluid into a first filtration unit (10), wherein the first filtration unit comprises a receptacle (110) and a filter membrane (120); b. passing the fluid through and / or across the filter membrane (120) of the first filtration unit (10) to obtain a retentate and a permeate; c. re-circulating the retentate obtained in step b to the receptacle (110) of the first filtration unit (10), and passing the permeate obtained in step b to a second filtration unit (20), wherein the second filtration unit comprises a receptacle (210), and a filter membrane (220), and wherein the permeate obtained in step b enters the receptacle (210) of the second filtration unit; d. passing fluid from the receptacle (210) of the second filtration unit (20) through the filter membrane (220) of the second filtration unit (20) to obtain a retentate and a permeate; e. re- circulating the retentate obtained in step d to the receptacle (210) of the second filtration unit (20), and f. collecting the permeate obtained in step d or passing the permeate obtained in step d to a third filtration unit (30); and g. creating a first counter-current flow by re-circulating a fraction of the fluid from the receptacle (210) of the second filtration unit (20) to the receptacle (110) of the first filtration unit (10). In some embodiments, step f. consists of passing the permeate obtained in step d to the third filtration unit (30), wherein the filtration unit comprises a receptacle (310) and a filter membrane (320), wherein the permeate obtained in step d enters the receptacle (310) of the third filtration unit; and wherein the method further comprises h. passing fluid from the receptacle (310) of the third filtration unit through the filter membrane (320) of the third filtration unit to obtain a retentate and a permeate; i. re-circulating the retentate obtained in step h to the receptacle (310) of the third filtration unit, and passing the permeate obtained in step h to a fourth filtration unit (40), wherein the fourth filtration unit (40) comprises a receptacle (410) and a filter membrane (420), wherein the permeate obtained in step h enters the receptacle (410) of the fourth filtration unit (40); j. passing fluid from the receptacle (410) of the fourth filtration unit through the filter membrane (420) of the fourth filtration unit to obtain a retentate and a permeate; and i. creating a second counter-current flow by re-circulating a fraction of the fluid from the receptacle (310) of the third filtration unit (30) to the receptacle of the second filtration unit; and k. recovering the molecule of interest from the fourth filtration unit; and optionally l. re-circulating the permeate obtained in the fourth filtration unit to the first filtration unit as a wash liquid.

[0115] Without wishing to be bound by theory, it will be understood that, in the course of the filtration method, the content of the first feed tank will comprise a mixture of the fluid fed back from the second feed tank and the retentate from the first filter membrane, and optionally permeate fed back from the last filter membrane. Similarly, the content of the second feed tank will comprise permeate from the first filtration membrane, fluid fed back from the third feed tank, and retentate fed back from the second filter membrane; and so forth with respect to any further filtration unit.

[0116] The skilled person will understand that at the end of the method, the impurity will have accumulated in the first filtration unit and the molecule of interest will have accumulated in the last filtration unit when operated in batch mode. The impurity concentration will reach a pseudo-steady state in filtration units 10, 20, and 30. This can be seen from modelling in Fig. 4C, where the concentrations of the impurities after the initial change reaches an almost steady state. Experimental verification of this can be seen when looking at e.g., Fig. 5 or Fig. 6. After an initial increase in the filtration units 2 and 3 (see first row of Fig. 5), the area of the impurities stays almost constant in the following rows with time points from 10:05 to 13:45. With regard to the first filtration unit (Fig. 5, first column) it should be noted that the volume was reduced as can be seen in the header. However also here the area of the impurities is almost constant for the different time points.

[0117] The filter membranes are selected such that more molecule of interest can pass through than impurity. Put differently, the transmission of the molecule of interest is higher than the transmission of the impurity. The filtration system may comprise more than four filtration units. For instance, the filtration system may comprise five filtration units or six filtration units or seven filtration units. The skilled person will understand that the method will change accordingly by simply repeating e.g., steps d and e with every filtration unit added with the proviso that the fluid or a portion thereof in the last receptacle will not be transferred back to the feed tank of a preceding filtration unit to create a counter-current. In some embodiments, water or buffer is added to wash out molecule of interest and / or impurities. In some embodiment, there is one or more concentration step(s). The fluid may also be subjected to a chromatographic purification, such as, e.g., an ion exchanger after the last filtration unit or instead of the last filtration step.

[0118] Surprisingly, adding a counter-current flow and selecting a filter membrane that provides for a different transmission of impurity compared to molecule of interest allows for better separation and purification of the molecule of interest compared to regular cross-flow tangential filtration, which in turn typically results in higher purities and yields of the molecule of interest. Referring to Figs. 3A and C, one can see that without the added counter-current flow, the impurity starts accumulating in the last filtration unit (Fig. 3C, “Filter no. 4”) as well as the product (Fig. 3A, “Filter no. 4”). As can be seen, the impurity starts to accumulate very quickly with increased wash volume (around 1.5) while most of the product only reaches the last filtration unit at around 0 = 6 (see Fig. 3A). Hence, while good separation may be achieved with a low 0, it will not be possible to achieve a high yield. Conversely, looking at Fig. 4C one can see that the impurity does not accumulate to the same extend in the last filtration unit with an added counter-current flow. Hence, a process with added countercurrent flow allows for efficient separation and / or high yield. The skilled person will understand that by adding more filtration units or increasing the counter-current flow the amount of impurity (such as, e.g., HCP) ending up in the last filtration unit can be further reduced.

[0119] In some embodiment, each receptacle individually is a feed tank. In some embodiments, each receptacle individually is a mixing unit configured for mixing incoming fluids. In some embodiments, each receptacle individually is a pipe or channel configured for mixing incoming fluids.

[0120] In particular embodiments, the impurity is HCP. In some embodiments, the HCP is immunogenic HCP.

[0121] Surprisingly, the methods described herein achieve HCP counts well below the industry acceptable standard of < 100 ppm. The HCP content can be measured by methods known in the art. For instance, the immunogenic HCP content may be measured according to the method as set out in the examples. The HCP content may be measured by quantitative HPLC-MS.

[0122] Surprisingly, some embodiments of third aspect achieve HCP content of < 50 ppm whilst preserving stability of the molecule of interest, reducing aggregation, and / or maintaining the yield of the molecule of interest. More surprisingly, some embodiments of third aspect achieve HCP content of < 40 ppm, such as < 30 ppm, such as < 20 ppm, whilst preserving stability of the molecule of interest, reducing aggregation, and / or maintaining the yield of the molecule of interest. HCP content may be measured by ELISA.

[0123] The methods described herein are broadly applicable for the separation of impurity and molecule of interest. The methods described herein may be performed in order to produce a composition comprising a molecule of interest having reduced host cell protein content (such as e.g., immunogenic HCP content), wherein the HCP content of said composition is <100 ppm, < 50 ppm, < 40 ppm, < 35, ppm, <30 ppm, < 25 ppm, < 20 ppm, optionally wherein the immunogenic HCP content can be measured by ELISA. Accordingly, there is provided methods of reducing host cell protein content in a peptide preparation such as a GLP-1 peptide preparation such as a semaglutide or extended semaglutide preparation. Conveniently, subjecting a protein preparation comprising a molecule of interest and HCP, to the methods disclosed herein may reduce the HCP content by a factor of 100, such as by a factor 1000, such as by a factor of 10000. The methods disclosed herein may results in a log 4 reduction of HCP.

[0124] In some embodiments, the molecule of interest is recombinantly produced in a mammalian host cell, such as a Chinese hamster ovary cell host cell. In preferred embodiments, the molecule of interest is recombinantly produced in a yeast host cell, such as an s. cerevisiae host cell. In some embodiments, the molecule of interest is recombinantly produced in a bacterial host cell, such as an E. coli host cell.

[0125] Accordingly, in particular embodiments, there is provided a method of reducing host cell protein content in a peptide preparation comprising a peptide recombinantly produced in a yeast host cell to a method according to the third aspect of the disclosure. The pH of the peptide preparation may be adjusted to a pH of about 3, about 4, about 5, about 6, about 7, about 8, about 8.5, or about 9. In some embodiments, the pH of the peptide preparation (or fluid) is in the range of 6.5 to 8.5. In some embodiments, the pH of the peptide preparation (or fluid) is in the range of 7.0 to 8.0. In some embodiments, the pH of the peptide preparation (or fluid) is in the range of 7.8 to 8.8.

[0126] Based on an exemplary embodiment of the second aspect of the disclosure as shown in Figs. 10A-E, an exemplary method according to the third aspect will be described. The protein preparation in form of a fluid comprising a molecule of interest (e.g., a recombinantly produced peptide having a molecular weight in the range of 1-7 kDa, such as in the range of 2.5-5.0) and an impurity (e.g., HCP having a molecular weight in the range of about 7-130 kDa, e.g., having a median molecular weight of around 40 kDa) is provided and enters a first receptacle (e.g., feed tank 110) via an inlet (111). The fluid then exists the feed tank (110) via an outlet (121) and passes via a conduit (123) through a first filtration membrane (120) having large pores (e.g. a membrane having a separation factor essentially 1) or alternatively having medium pores (e.g. wherein the transmission of the molecule of interest is higher than the impurity. For instance, 5% for the molecule of interest and 1% for the impurity) to obtain a first retentate and a first permeate. The first retentate is enriched in impurity while the first permeate is enriched in molecule of interest. The first retentate circulates back via a conduit (131) and enters the first feed tank (110) via an inlet (114). A pump (130) controls the flow of the fluid such that membrane fouling is avoided or reduced and / or build-up is avoided or reduced. A flow meter may be used to measure the flow rate. The so-obtained permeate passes through a conduit (122) and enters a second feed tank (210) via an inlet (211). A portion of the fluid in the second feed tank (210) is fed back via a conduit (134) entering the first feed tank (120) via an inlet (213) thereby creating a countercurrent. One or more device(s) (251, 252) control the flow of the counter-current. The device may be a pump (251) or a control valve (252). Next, the feed fluid exits the feed tank (210) via an outlet (221) and passes through a second filter membrane (220) having medium pores (e.g. wherein the transmission of the molecule of interest is higher than the impurity. For instance, 5% for the molecule of interest and 1% for the impurity) to obtain a retentate and a permeate. The so-obtained retentate is further reduced in impurities compared to the retentate obtained in the preceding filtration unit and the permeate is further enriched in molecule of interest compared to the permeate obtained in the subsequent filtration unit. The retentate circulates back via a conduit (231) and enters the second feed tank (210) via an inlet (214). A pump (230) controls the feed flow such that membrane fouling is avoided or reduced and / or build-up is avoided or reduced. The permeate passes through a conduit (233) and enters a feed tank (310) of a subsequent filtration unit via an inlet (311). A portion of the feed fluid in the feed tank (310) is fed back via a conduit (234) entering the second feed tank (210) via an inlet (213) thereby creating a counter-current. A device (251 , 252) controls the counter-current flow. Next, the feed fluid exits the feed tank (310) via an outlet (321) and passes through a membrane (320) having medium pore size to obtain a retentate and a permeate. The impurity content in the so-obtained retentate is further reduced compared to the retentate obtained in the preceding filtration unit and the so-obtained permeate is further enriched in molecule of interest compared to the permeate obtained in the preceding filtration unit. The retentate is circulated back via a conduit (331) and enters the feed tank (310) via an inlet (314). A pump (330) controls the feed flow such that membrane fouling is avoided or reduced and / or build-up is avoided or reduced. The third permeate passes through a conduit (333) and enters a feed tank (410) of a subsequent filtration unit via an inlet (411). Next, the feed fluid exits the feed tank (410) via an outlet (421) and passes through a filter membrane (420) having small pores (e.g. a membrane having a transmission of essentially 0 with regard to molecule of interest and impurity) to obtain a retentate and a permeate. The so-obtained retentate comprises the molecule of interest. The content of impurity in the so-obtained retentate is reduced by a factor of at least 1000, such as 5000, such as 10000 compared to the content of impurity in the feed fluid at the start of the method. The so-obtained permeate may be circled back to the first filtration unit to be used as a wash liquid or alternatively the so-obtained permeate may be collected in e.g., a receptacle. Also, or alternatively the fourth permeate may be passed into a receptacle to be disposed of and / or circulates back to the first receptacle or feed tank to serve as a wash liquid.

[0127] Conveniently, the molecule of interest may be obtained in a yield of > 85%, such as above 90%, such as above 95%, such as above 96%, such as above 97%, such as above 98%, such as above 99%. In some embodiments, the molecule of interest is obtainable in a yield of 85-98%, such as 90-96%.

[0128] The molecule of interest can be a recombinantly produced protein, polypeptide, nucleic acid, or glycoprotein. In some embodiments, the molecule of interest is a recombinantly produced peptide. In some embodiments, the molecule of interest is a peptide obtained by recombinant production in yeast, such as in s. cerevisiae. In some embodiments, the molecule of interest is a recombinantly produced GLP-1 peptide or a precursor thereof. In some embodiments, the molecule of interest is polypeptide 1 obtained by recombinant production in yeast, such as in s. cerevisiae. In some embodiments, the molecule of interest is polypeptide 2 obtained by recombinant production in yeast, such as in s. cerevisiae.

[0129] The relative (immunogenic) HCP to product ratio in the feed fluid after can be below 20 ppm.

[0130] In some embodiments, each filter membrane has a different MWCO. In particular embodiments, the last filter membrane has small pores and the one or more preceding filter membrane have medium pores and optionally the first filter membrane has large pores.

[0131] In some embodiments, the first filter membrane has a transmission in the range of about 0.8 to 1.

[0132] In some embodiments, the last filter membrane has a transmission in the range of about 0.0 to 0.1. A filter membrane having a transmission in the range of about 0.0 to 0.1 may be referred to herein as a filter membrane having small pores. Conveniently, a filter membrane having small pores may be part of the last filtration unit to obtain a permeate and a retentate, wherein the retentate comprises the molecule of interest and the permeate may be used a wash liquid.

[0133] In some embodiments, at least one filter membrane has a separation factor in the range of 2 to 100. In some embodiments, at least one filter membrane has a separation factor in the range of 5 to 100. In some embodiments, at least one filter membrane has a separation factor in the range of 10 to 100. In some embodiments, at least one filter membrane has a separation factor in the range of 15 to 100. In some embodiments, at least one filter membrane has a separation factor in the range of 20 to 100. In some embodiments, at least one filter membrane has a separation factor in the range of 25 to 100. In some embodiments, at least one filter membrane has a separation factor in the range of 30 to 100. In some embodiments, at least one filter membrane has a separation factor in the range of 35 to 100. In some embodiments, at least one filter membrane has a separation factor in the range of 40 to 100. In particular embodiments, the first filter membrane has a transmission in the range of about 0.8 to 1 , the last filter membrane has a transmission in the range of about 0.0 to 0.1 , and at least one filter membrane has a separation factor in the range of 2 to 100.

[0134] The filter membrane described herein may have a MWCO in the range of 1-100 kDa, such as in the range of 1-90 kDa, such as in the range of 1-90 kDa, such as in the range of 1-85 kDa, such as in the range of 1-80 kDa, such as in the range of 1-75 kDa, such as in the range of 1-70 kDa, such as in the range of 1-65 kDa, such as in the range of 1-60 kDa, such as in the range of 1-55 kDa, such as in the range of 1-50 kDa, such as in the range of 1-45 kDa, such as in the range of 1-40 kDa, such as in the range of 1-35 kDa, such as in the range of 1-30 kD, such as in the range of 1-25 kDa, such as in the range of 1-20 kDa, such as in the range of 1-15 kDa, such as in the range of 1- 10 kDa, such as in the range of 1-5 kDa. The filter membrane described herein may have a MWCO in the range of 5-100 kDa, such as in the range of 5-90 kDa, such as in the range of 5-90 kDa, such as in the range of 5-85 kDa, such as in the range of 5-80 kDa, such as in the range of 5-75 kDa, such as in the range of 5-70 kDa, such as in the range of 5-65 kDa, such as in the range of 5-60 kDa, such as in the range of 5-55 kDa, such as in the range of 5-50 kDa, such as in the range of 5-45 kDa, such as in the range of 5-40 kDa, such as in the range of 5-35 kDa, such as in the range of 5-30 kD, such as in the range of 5-25 kDa, such as in the range of 5-20 kDa, such as in the range of 5-15 kDa, such as in the range of 5-10 kDa. In some embodiments, each filter membrane individually is selected from the group of membranes having a MWCO of 1 kDa, 5 kDa, 10 kDa, 15 kDa, 20 kDa, 25 kDa, 30 kDa, 40 kDa, 45 kDa, 50 kDa, 55 kDa, 60 kDa, 65 kDa, 70 kDa, 75 kDa, 80 kDa, 85 kDa, 90 kDa, 95 kDa, and 100 kDa.

[0135] In particular embodiments, the system comprises a membrane having a “medium pore size” in at least the first two filtration units and a membrane having a “small pore size” in the last filtration unit. In further particular embodiments, the system comprises a membrane having a “large pore size” either replacing the membrane having a “medium pore size” in the first filtration unit or set up at the beginning of the system. In some embodiments, a filter membrane having a large pore size is added upstream of the first filtration unit comprising a filter membrane having a medium pore size.

[0136] In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about or above 2 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about or above 5 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about or above 10 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about or above 20 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about or above 30 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about or above 40 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about or above 50 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about or above 60 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about or above 70 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about or above 80 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about or above 90 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about 100 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about 110 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about 120 over each step. In some embodiments, the ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about 130 over each step. “Over each step” means passing from one filtration unit to the next. For instance, ratio between the impurity, such as HCP, and the molecule of interest is reduced by a factor of about or above 2 by passing the feed fluid from the first filtration unit to the second filtration unit. The ratio between the impurity, such as HCP, and the molecule of interest is then further reduced by a factor of about or above 2 by passing the feed fluid further from the second filtration unit to third filtration unit and so on.

[0137] It should be understood to those skilled in the art that the optimal operation of the system relies on knowledge of how a particular feed material and product compound behaves under various operational conditions and that this knowledge is normally gathered through pilot and production-scale studies.

[0138] List of exemplary embodiments

[0139] Embodiment 1. A method for removing or reducing an impurity, the method comprising the steps of: a. subjecting a fluid comprising a molecule of interest and an impurity to a countercurrent filtration by feeding the fluid into a first receptacle (110) of a counter-current filtration system comprising three or more filtration units (10, 20, 30); b. passing the fluid through a first filter membrane to obtain a retentate and a permeate; c. re-circulating the so-obtained retentate to the first receptacle (110), and passing the so-obtained permeate into a second feed tank (210) of a subsequent or second filtration unit (200); d. passing the fluid from the second feed receptacle (210) through a second filter membrane to obtain a retentate and a permeate; e. re-circulating the so-obtained retentate to the second receptacle (210), and passing the so-obtained permeate into a third or last receptacle (310) of a subsequent or third filtration unit (300); f. passing the fluid from the third receptacle (310) through a third or last filter membrane to obtain a retentate comprising the molecule of interest and a permeate; and g. creating a counter-current feed flow by passing a portion of the fluid from the second receptacle (210) to the first receptacle (110), and optionally controlling said portion via a pump or a control valve; and h. recovering the retentate obtained in the third or last filtration unit or recovering the fluid from the third or last receptacle (310); and optionally i. re-circulating the permeate obtained in the third or last filtration unit to the first filtration unit as a wash liquid.

[0140] Embodiment 2. A method for removing or reducing an impurity, the method comprising the steps of: a. subjecting a fluid comprising a molecule of interest and an impurity to a countercurrent filtration by feeding the fluid into a first receptacle (110) of a counter-current filtration system comprising four or more filtration units (10, 20, 30, 40); b. passing the fluid through a first filter membrane to obtain a retentate and a permeate; c. re-circulating the so-obtained retentate to the first receptacle (110), and passing the so-obtained permeate into a second receptacle (210) of a subsequent or second filtration unit (20); d. passing the fluid from the second receptacle (210) through a second filter membrane to obtain a retentate and a permeate; e. re-circulating the so-obtained retentate to the second receptacle (210), and passing the so-obtained permeate into a third receptacle (310) of a subsequent or third filtration unit (30); f. passing the fluid from the third receptacle (310) through a third filter membrane to obtain a retentate comprising the molecule of interest and a permeate; g. re-circulating the so-obtained retentate to the third receptacle (310), and passing the so-obtained permeate into a fourth or last receptacle (410) of a subsequent or fourth filtration unit (400); h. passing the fluid from the fourth or last receptacle (410) through a fourth or last filter membrane to obtain a retentate comprising the molecule of interest and a permeate; and i. creating a first counter-current feed flow by passing a portion of the fluid from the second receptacle (210) to the first receptacle (110); and j. creating a second counter-current feed flow by passing a portion of the fluid from the third receptacle (310) to the second receptacle (210); and k. recovering the retentate obtained in the fourth or last filtration unit or recovering the fluid from the fourth or last receptacle (410); and optionally l. re-circulating the permeate obtained in the fourth or last filtration unit to the first filtration unit as a wash liquid.

[0141] Embodiment 3. A method of removing or reducing an impurity in a protein preparation comprising a molecule of interest recombinantly produced in a host cell, the method comprising the steps of: a. introducing the protein preparation as a fluid into a first filtration unit (10), wherein the first filtration unit comprises a receptacle (110) and a filter membrane (120); b. passing the fluid through the filter membrane (120) of the first filtration unit (10) to obtain a retentate and a permeate; c. re-circulating the retentate obtainable in step b to the receptacle (110) of the first filtration unit (10), and passing the permeate obtainable in step b to a second filtration unit (20), wherein the second filtration unit comprises a receptacle (210), and a filter membrane (220), and wherein the permeate obtainable in step b moves into the receptacle (210) of the second filtration unit; d. passing fluid from the receptacle (210) of the second filtration unit (20) through the filter membrane (220) of the second filtration unit (20) to obtain a retentate and a permeate; e. re-circulating the retentate obtainable in step d to the receptacle (210) of the second filtration unit (20), and f. collecting the molecule of interest from the second filtration unit or passing the permeate obtainable in step d to a third filtration unit (30); and g. creating a first counter-current flow by re-circulating a fraction of the fluid from the receptacle (210) of the second filtration unit (20) to the receptacle (110) of the first filtration unit (10).

[0142] Embodiment 4. The method according to claim 1, wherein step f. consists of passing the permeate obtainable in step d to the third filtration unit (30), and wherein the third filtration unit comprises a receptacle (310) and a filter membrane (320), and wherein the permeate obtainable in step d enters the receptacle (310) of the third filtration unit; and wherein the method further comprises h. passing fluid from the receptacle (310) of the third filtration unit through the filter membrane (320) of the third filtration unit to obtain a retentate and a permeate; i. re-circulating the retentate obtainable in step h to the receptacle (310) of the third filtration unit, and passing the permeate obtainable in step h to a fourth filtration unit (40), wherein the fourth filtration unit (40) comprises a receptacle (410) and a filter membrane (420), wherein the permeate obtainable in step h enters the receptacle (410) of the fourth filtration unit (40); j. passing fluid from the receptacle (410) of the fourth filtration unit through the filter membrane (420) of the fourth filtration unit to obtain a retentate and a permeate; and i. creating a second counter-current flow by re-circulating a fraction of the fluid from the receptacle (310) of the third filtration unit (30) to the receptacle of the second filtration unit; and k. collecting the molecule of interest from the fourth filtration unit or passing the permeate obtainable in step j to a subsequent filtration unit; and optionally l. re-circulating the permeate obtainable in the fourth filtration unit to the first filtration unit as a wash liquid.

[0143] Embodiment 5. The method according to any one of embodiments 1 to 4, wherein the method is a tangential flow filtration method.

[0144] Embodiment 6. The method according to any one of embodiments 1 to 5, wherein the fluid circulates across each filter membrane in tangential flow filtration mode. Embodiment 7. The method according to any one of embodiments 1 to 6, wherein the fluid is circulated using tangential flow filtration mode at a loading of between 1 to 100 litres of bioreactor harvest square metre of surface area of each filter membrane.

[0145] Embodiment 8. The method according to any one of embodiments 1 to 7, wherein each counter-current flow individually is in the range of

[0146] 0.01 < — < 0.99. l+y

[0147] Embodiment 9. The method according to any one of embodiments 1 to 7, wherein each counter-current flow individually is in the range of

[0148] 0.05 < — < 0.95, such as 0.1 < — < 0.90. l+y l+y

[0149] Embodiment 10. The method according to any one of embodiments 1 to 9, wherein each counter-current flow rate is constant.

[0150] Embodiment 11. The method according to any one of embodiments 1 to 10, wherein the flow rate of the first permeate being passed into the second receptacle is in the range of 1 litres / m2 / hour (LMH) to 300 LMH.

[0151] Embodiment 12. The method according to any one of embodiments 1 to 11, wherein the flow rate of the first permeate being passed into the second receptacle is in the range of 1 litres / m2 / hour (LMH) to 200 LMH.

[0152] Embodiment 13. The method according to any one of embodiments 1 to 12, wherein the flow rate of the second permeate being passed into the third receptacle is in the range of 1 litres / m2 / hour (LMH) to 100 LMH.

[0153] Embodiment 14. The method according to any one of embodiments 1 to 13, wherein the flow rate of the third permeate being passed into the fourth receptacle is in the range of 1 litres / m2 / hour (LMH) to 50 LMH.

[0154] Embodiment 15. The method according to any one of embodiments 1 to 14, wherein the flow rate of a further permeate being passed into a further receptacle is in the range of 1 litres / m2 / hour (LMH) to 25 LMH.

[0155] Embodiment 16. The method according to any one of embodiments 1 to 14, wherein the impurity is host cell protein (HCP), optionally immunogenic HCP. Embodiment 17. The method according to any one of embodiments 1 to 16, wherein the molecule of interest has been recombinantly produced, optionally wherein the molecule of interest has been produced in yeast, such as s. cerevisiae.

[0156] Embodiment 18. The method according to any one of embodiments 1 to 17, wherein the molecule of interest is a protein or a peptide, such as a recombinantly produced protein or peptide.

[0157] Embodiment 19. The method according to any one of embodiments 1 to 18, wherein the molecule of interest is a GLP-1 peptide or a precursor thereof, such as a recombinantly produced GLP-1 peptide or precursor thereof, such as an N-terminally extended [Arg34]GLP- 1(9-37) peptide.

[0158] Embodiment 20. The method according to any one of embodiments 1 to 18, wherein the molecule of interest is polypeptide 1 or polypeptide 2.

[0159] Embodiment 21. The method according to any one of embodiments 1 to 20, wherein each receptacle is a tank such as a feed tank or a mixing unit such as a pipe.

[0160] Embodiment 22. The method according to any one of embodiments 1 to

[0161] 21 , wherein each receptacle individually has a capacity in the range of 0.01 to 5 m3.

[0162] Embodiment 23. The method according to any one of embodiments 1 to

[0163] 21 , wherein each receptacle individually has a capacity in the range of 0.1 to 5 m3.

[0164] Embodiment 24. The method according to any one of embodiments 1 to

[0165] 21 , wherein each receptacle individually has a capacity in the range of 1 to 5 m3.

[0166] Embodiment 25. The method according to any one of embodiments 1 to

[0167] 21 , wherein each receptacle individually has a capacity in the range of 5 to 10 m3.

[0168] Embodiment 26. The method according to any one of embodiments 1 to 25, wherein each receptacle individually is kept at a temperature in the range of -5 °C to 85 °C, such as in the range of 15 °C to 35 °C.

[0169] Embodiment 27. The method according to any one of embodiments 1 to 25, wherein each receptacle individually is kept at a temperature of 35 °C to 85 °C, such as 45 °C to 50 °C or 35 °C to 40 °C. Embodiment 28. The method according to any one of embodiments 1 to 27, wherein the fourth or last receptacle is kept at a temperature in the range of -5 °C to 15 °C, such as 0 °C to 10 °C.

[0170] Embodiment 29. The method according to any one of embodiments 1 to 28, wherein each filter membrane is an ultrafiltration membrane.

[0171] Embodiment 30. The method according to any one of embodiments 1 to 29, wherein at least one filter membrane has a separation factor in the range of 2-100.

[0172] Embodiment 31. The method according to any one of embodiments 1 to 30, wherein at least two filter membranes individually have a separation factor in the range of 2-100.

[0173] Embodiment 32. The method according to any one of embodiments 1 to 31 , wherein at least one filter membrane has a separation factor in the range of 5-80.

[0174] Embodiment 33. The method according to any one of embodiments 1 to 31 , wherein at least one filter membrane has a separation factor in the range of 10-70.

[0175] Embodiment 34. The method according to any one of embodiments 1 to 33, wherein the separation factor of the first filter membrane differs from the separation factor of the last filter membrane.

[0176] Embodiment 35. The method according to any one of embodiments 1 to 34, wherein the first filter membrane has a molecular weight cut-off selected from the range of 30-100 kDa.

[0177] Embodiment 36. The method according to any one of embodiments 1 to 34, wherein the first filter membrane has a molecular weight cut-off of about 30 kDa.

[0178] Embodiment 37. The method according to any one of embodiments 1 to 34, wherein the first filter membrane has a molecular weight cut-off of about 35 kDa.

[0179] Embodiment 38. The method according to any one of embodiments 1 to 34, wherein the first filter membrane has a molecular weight cut-off of about 40 kDa.

[0180] Embodiment 39. The method according to any one of embodiments 1 to 34, wherein the first filter membrane has a molecular weight cut-off of about 45 kDa or about 50 kDa or about 55.

[0181] Embodiment 40. The method according to any one of embodiments 1 to 39, wherein the second filter membrane has the same molecular weight cut-off as the first filter membrane.

[0182] Embodiment 41. The method according to any one of embodiments 1 to 40, wherein the second filter membrane has a molecular weight cut-off selected from the range of 20-100 kDa. Embodiment 42. The method according to any one of embodiments 1 to 40, wherein the second filter membrane has a molecular weight cut-off of about 20 kDa.

[0183] Embodiment 43. The method according to any one of embodiments 1 to 40, wherein the second filter membrane has a molecular weight cut-off of about 30 kDa.

[0184] Embodiment 44. The method according to any one of embodiments 1 to 40, wherein the second filter membrane has a molecular weight cut-off of about 35 kDa.

[0185] Embodiment 45. The method according to any one of embodiments 1 to 40, wherein the second filter membrane has a molecular weight cut-off of about 40 kDa or of about 50 kDa.

[0186] Embodiment 46. The method according to any one of embodiments 1 to 45, wherein the third filter membrane has the same molecular weight cut-off as the first filter membrane and / or the second filter membrane.

[0187] Embodiment 47. The method according to any one of embodiments 1 to 45, wherein the third filter membrane has a molecular weight cut-off selected from the range of 20-100 kDa.

[0188] Embodiment 48. The method according to any one of embodiments 1 to 45, wherein the third filter membrane has a molecular weight cut-off of about 20 kDa.

[0189] Embodiment 49. The method according to any one of embodiments 1 to 45, wherein the third filter membrane has a molecular weight cut-off of about 30 kDa.

[0190] Embodiment 50. The method according to any one of embodiments 1 to 45, wherein the third filter membrane has a molecular weight cut-off of about 35 kDa.

[0191] Embodiment 51. The method according to any one of embodiments 1 to 45, wherein the third filter membrane has a molecular weight cut-off of about 40 kDa.

[0192] Embodiment 52. The method according to any one of embodiments 1 to 51 , wherein the first receptacle is capable of receiving the first retentate, the feed solution, a portion of the feed fluid re-circulatable from the second receptacle, and optionally the permeate obtainable from the last filtration membrane.

[0193] Embodiment 53. The method according to any one of embodiments 1 to 52, wherein the second receptacle is capable of receiving the first permeate, the second retentate, and a portion of the feed fluid re-circulatable from the third receptacle.

[0194] Embodiment 54. The method according to any one of embodiments 1 to 53, wherein method comprises an additional step of re-circulating the fourth or last permeate to the first receptacle for use as wash fluid. Embodiment 55. The method according to any one of embodiment 1 to 53, accumulating the molecule of interest in the fourth or last receptacle.

[0195] Embodiment 56. The method according to any one of embodiment 1 to 55, adding a wash liquid to the first tank to wash out any remaining molecule of interest.

[0196] Embodiment 57. The method according to 56, wherein the wash liquid is selected from the group consisting of a solvent, water, buffer, and a mixture thereof.

[0197] Embodiment 58. The method according to any one of embodiments 1 to 57, wherein the feed fluid passes through a filter membrane to reduce NTU before entering the first filtration unit.

[0198] Embodiment 59. The method according to 58, wherein the filter membrane to reduce NTU has a molecular weight cut-off in the range of 100-800 kDa, such as in the range of 100-600 kDa.

[0199] Embodiment 60. The method according to 58, wherein the filter membrane to reduce NTU has a molecular weight cut-off in the range of 120-600 kDa, such as in the range of 120-500 kDa.

[0200] Embodiment 61. The method according to any one of embodiments 1 to 60, wherein each receptacle individually is kept at a temperature in the range of -5 °C to 15 °C, such as 0 °C to 10 °C or at a range of 30°C to 85 °C, such as 35 °C to 55 °C or 35 °C to 45 °C.

[0201] Embodiment 62. The method according to any one of embodiments 1 to 61, wherein the ratio of impurity to molecule of interest in the fourth or last receptacle is reduced by a factor of about or above 10, such as about or above 100, such as about or above 1000, such as about 10000 compared to the ratio of impurity to molecule of interest in the feed solution in the first or initial receptacle, at the start of the method.

[0202] Embodiment 63. A method for optimising counter-current filtration for removing host cell protein from a feed fluid comprising host cell protein and a molecule of interest, the method comprising:

[0203] (i) selecting a membrane having a separation factor a in the range on 2-200;

[0204] (ii) applying a mathematical formula (6) (iii) adjusting the counter-current flow and / or the number of filtration units to achieve the desired reduction in host cell protein.

[0205] Embodiment 64. A filtration unit for counter-current filtration, wherein the filtration unit comprises a receptacle, a filter membrane, a pump, a plurality of inlets, a plurality of outlets, a plurality of channels, and a device for creating and / or controlling a counter-current flow.

[0206] Embodiment 65. A filtration unit (10, 20, 30, 40) comprising a receptacle (110, 210, 310, 410), a filter membrane (120, 220, 320, 420), and a feed pump (130, 230, 330, 430), wherein the receptacle (110, 210, 310, 410) comprises an inlet (111, 211 , 311, 411) such that feed fluid from a subsequent filtration unit can enter the receptacle (110, 210, 310, 410), an inlet (113, 213, 313, 413) such that counter-current feed fluid can enter the receptacle (110, 210, 310, 410), and an inlet (114, 214, 314, 414) such that retentate can enter the receptacle (110, 210, 310, 410), the receptacle (110, 210, 310, 410) further comprises a feed outlet (121 , 221 , 321, 421), and wherein the filter membrane (120, 220, 320, 420) comprises an inlet (112, 212, 312, 412) on a first side of the membrane and further comprises a retentate outlet (123, 223, 323, 423) on a first side of the membrane and a permeate outlet (122, 222, 322, 422) on the other side of the membrane, optionally wherein each receptacle is a feed tank.

[0207] Embodiment 66. The filtration unit according to embodiment 65, wherein the filtration unit (20) further comprises a recirculation channel (131 , 231, 331 , 431) for circulating a retentate from a first or entrance side of the membrane (120, 220, 320, 420) to the receptacle (110, 210, 310, 410); a feed channel (132, 232, 332, 432) for transporting the feed fluid from the outlet of the receptacle to the inlet of the filter membrane (120, 220, 320, 420); a permeate channel (133, 233, 333, 433) for transporting the permeate from the other side of the filter membrane (120, 220, 320, 420) to the subsequent filtration unit; and a counter-current feed channel (234, 334) for transporting the feed fluid from a subsequent filtration unit back to the feed tank (120, 210).

[0208] Embodiment 67. The filtration unit according to any one of embodiments 64 to 66, wherein the filter membrane is an ultrafiltration membrane.

[0209] Embodiment 68. The filtration unit according to any one of embodiments 64 to 67, wherein each filter membrane individually has a molecular weight cut-off in the range of 1-800 kDa, such as in the range of 1-500 kDa, such as in the range of 1-200 kDa, such as in the range of 1-150 kDa, such as in the range of 5-100 kDa, such as in the range of 5-50 kDa, such as in the range of 5-30 kDa. Embodiment 69. The filtration unit according to any one of embodiments 64 to 68, wherein the filtration unit is connected or connectable to a subsequent filtration unit according to 64, optionally, wherein the subsequent filtration unit does not comprise an outlet for the feed solution to create a counter-current.

[0210] Embodiment 70. The filtration unit according to any one of embodiment 64 to 69, wherein the filtration unit is connected or connectable to a preceding filtration unit according to embodiment 64 or 65, optionally, wherein the preceding filtration unit does not comprise an outlet for the feed solution to create a counter-current.

[0211] Embodiment 71. A system for counter-current filtration of a feed fluid comprising an impurity and a molecule of interest, wherein the system comprises at least two filtration units according to any one of embodiments 64 to 70.

[0212] Embodiment 72. A system for counter-current filtration of a feed fluid comprising an impurity and a molecule of interest, wherein the system comprises two or more filtration units, wherein each filtration comprises (i) a receptacle having two or more inlet(s), wherein the inlets are for receiving a feed solution, a permeate or a retentate; and (ii) two or more outlets, wherein each outlet individually is for carrying the feed fluid or for returning the feed fluid to a preceding filtration unit to create a counter-current; (iii) a filter membrane; (iv) a pump to transfer feed fluid from the receptacle across the filter membrane to produce a retentate and a permeate; optionally (v) a device to regulate the re-circulating of the retentate to the receptable within each filtration unit; (vi) a device to create a counter-current flow from a subsequent filtration unit to a preceding filtration unit; and (vii) a plurality conduits.

[0213] Embodiment 73. The system for counter-current filtration according to embodiment 71 or embodiment 72, wherein the first filtration unit does not comprise an outlet for returning the feed fluid to a preceding filtration unit to create a counter-current.

[0214] Embodiment 74. The system for counter-current filtration according to any one of 71 to 73, wherein the last filtration unit does not comprise an outlet for returning the feed fluid to a preceding filtration unit to create a counter-current.

[0215] Embodiment 75. The system for counter-current filtration according to any one of embodiments 71 to 74, wherein one or more receptacles have an outlet to re-circulate a portion of the feed fluid to the preceding filtration unit, wherein said outlet is connected or connectable to a conduit connected or connectable to a receptacle of the preceding filtration unit, wherein the receptacle of the preceding filtration unit has an inlet to receive a portion of the feed fluid from the one or more receptacles having an outlet such that a counter-current flow can be created.

[0216] Embodiment 76. The system for counter-current filtration according to any one of embodiments 71 to 75, wherein the system comprises three or more filtration units.

[0217] Embodiment 77. The system for counter-current filtration according to embodiment 76, wherein the first filtration unit is connected or connectable to the second filtration unit via a conduit for carrying a permeate from the first filtration unit to the receptacle of the second filtration unit, wherein the second filtration unit is connected or connectable to the third filtration unit via a conduit for carrying a permeate from the second filtration unit to the receptacle of the third filtration unit, and wherein the receptacle of the third filtration unit comprises an outlet to collect the purified molecule of interest.

[0218] Embodiment 78. The system for counter-current filtration according to 77, wherein the third filtration unit is connected or connectable to the first filtration unit.

[0219] Embodiment 79. The system for counter-current filtration according to embodiment 77, wherein the system further comprises a means to introduce wash liquid.

[0220] Embodiment 80. The system for counter-current filtration according to any one of embodiments 71 to 74, wherein the system comprises four filtration units.

[0221] Embodiment 81. The system for counter-current filtration according to embodiment 76, wherein the first filtration unit is connected or connectable to the second filtration unit via a conduit for carrying a permeate from the first filtration unit to the receptacle of the second filtration unit, wherein the second filtration unit is connected or connectable to the third filtration unit via a conduit for carrying a permeate from the second filtration unit to the receptacle of the third filtration unit, wherein the third filtration unit is connected or connectable to the fourth filtration unit via a conduit for carrying a permeate from the third filtration unit to the receptacle of the fourth filtration unit, and wherein the receptacle of the fourth filtration unit comprises an outlet to collect the purified molecule of interest.

[0222] Embodiment 82. The system for counter-current filtration according to embodiment 81, wherein the fourth filtration unit is connected or connectable to the first filtration unit.

[0223] Embodiment 83. The system for counter-current filtration according to 81 , wherein the system further comprises a means to introduce wash liquid.

[0224] Embodiment 84. The system for counter-current filtration according to any one of embodiments 71 to 74, wherein the system comprises five or more filtration units. Embodiment 85. The system for counter-current filtration according to 84, wherein the first filtration unit is connected or connectable to the second filtration unit via a conduit for carrying a permeate from the first filtration unit to the receptacle of the second filtration unit, wherein the second filtration unit is connected or connectable to the third filtration unit via a conduit for carrying a permeate from the second filtration unit to the receptacle of the third filtration unit, wherein the third filtration unit is connected or connectable to the fourth filtration unit via a conduit for carrying a permeate from the third filtration unit to the receptacle of the fourth filtration unit, wherein the fourth filtration unit is connected or connectable to the fifth filtration unit via a conduit for carrying a permeate from the fourth filtration unit to the receptacle of the fifth filtration unit, wherein the receptacle of the second filtration unit comprises an inlet to receive the permeate from the first filtration unit, and wherein the receptacle of the fifth filtration unit comprises an outlet to collect the purified molecule of interest.

[0225] Embodiment 86. The system for counter-current filtration according to embodiment 81 , wherein the fourth filtration unit is connected or connectable to the first filtration unit.

[0226] Embodiment 87. The system for counter-current filtration according to embodiment 81 , wherein the system further comprises a means to introduce wash liquid.

[0227] Embodiment 88. The system for counter-current filtration according to any one of embodiments 71 to 87, where the receptacle of the second filtration unit has an outlet and a conduit connected or connectable to the receptacle of the first filtration unit, wherein the receptacle of the first filtration unit has an inlet such that a back flow can be created.

[0228] Embodiment 89. The system for counter-current filtration according to any one of embodiments 71 to 88, wherein each receptacle individually is a feed tank.

[0229] Embodiment 90. The system for counter-current filtration according to any one of embodiments 71 to 89, wherein at least one filter membrane, preferably the last filter membrane, has small pores.

[0230] Embodiment 91. The system for counter-current filtration according to any one of embodiments 71 to 89, wherein at least one filter membrane, preferably the last filter membrane, has a MWCO in the range of about 1- 10 kD, such as about 1 kDa, about 2 kDa, about 3 kDa, about 4 kDa, about 5 kDa, about 6 kDa, about 7 kDa, about 8 kDa, about 9 kDa, or about 10 kD, preferably about 5 kDa.

[0231] Embodiment 92. The system for counter-current filtration according to any one of embodiments 71 to 91, wherein at least one filter membrane, preferably the first filter membrane, has large pores. Embodiment 93. The system for counter-current filtration according to any one of embodiments 71 to 91, wherein at least one filter membrane, preferably the first filter membrane, has a MWCO in the range of about 100 to 800 kDa, such as about 100 kDa, about 120 kDa or about 150 kDa.

[0232] Embodiment 94. The system for counter-current filtration according to any one of embodiments 71 to 93, wherein one or more filter membrane(s) have medium pores.

[0233] Embodiment 95. The system for counter-current filtration according to any one of embodiments 71 to 93, wherein one or more filter membrane(s) have a MWCO of about 10 to 80 kDa, such as about 20 kDa, such as about 25 kDa, such as about 30 kDa, such as about 35 kDa, such as about 40 kDa, such as about 45 kDa, such as about 50 kDa, such as about 55 kDa, such as about 60 kDa, such as about 65 kDa, such as about 70 kDa, such as about 75 kDa, such as about 80 kDa, preferably such as about 30 kDa or 50 kDa.

[0234] Embodiment 96. The method according to any one of embodiments 1 to 62, wherein one or more filter membrane(s) have medium pores.

[0235] Embodiment 97. The method according to any one of embodiments 1 to 62, and 96, wherein a filter membrane(s) has small pores, preferably wherein the last filter membrane has small pores.

[0236] Embodiment 98. The method according to any one of embodiments 1 to 62, 96, and 97, wherein a filter membrane(s) has large pores, preferably wherein the first filter membrane has small pores.

[0237] Embodiment 99. The method according to any one of embodiments 1 to 62, 96, 97, and 98, wherein the impurity is HCP.

[0238] Embodiment 100. The method according to embodiment 63, wherein the mathematical formula (6) is extendable to N steps adding element(s) having formula (7) to the diagonal of formula and adding elements below the diagonal having formula (8); and adding elements above the diagonal having formula (9)

[0239] Embodiment 101. A method for removing or reducing an impurity, the method comprising the steps of: a. introducing a feed fluid comprising a molecule of interest and an impurity into a first filtration unit (10), wherein the filtration unit comprises a receptacle (110) and a filter membrane (120); b. passing the feed fluid through the filter membrane (120) of the first filtration unit (10) to obtain a retentate and a permeate; c. re-circulating the retentate obtained in step b to the receptacle (110) of the first filtration unit (10), and passing the permeate obtained in step b to a second filtration unit (20), wherein the second filtration unit comprises a receptacle (210), and a filter membrane (220), and wherein the permeate obtained in step b enters the receptacle (210) of the second filtration unit; d. passing feed fluid from the receptacle (210) of the second filtration unit (20) through the filter membrane (220) of the second filtration unit (20) to obtain a retentate and a permeate; e. re-circulating the retentate obtained in step d to the receptacle (210) of the second filtration unit (20), and f. collecting the permeate obtained in step d or optionally passing the permeate obtained in step d to a third filtration unit (30); and g. creating a first one counter-current flow by re-circulating a fraction of the fluid from the receptacle (210) of the second filtration unit (20) to the receptacle (110) of the first filtration unit (10).

[0240] Embodiment 102. The method according to embodiment 101 or embodiment 102, wherein step f. consists of passing the permeate obtained in step d to the third filtration unit (30), wherein the filtration unit comprises a receptacle (310) and a filter membrane (320), wherein the permeate obtained in step d enters the receptacle (310) of the third filtration unit; and wherein the method further comprises h. passing fluid from the receptacle (310) of the third filtration unit through the filter membrane (320) of the third filtration unit to obtain a retentate and a permeate; i. re-circulating the retentate obtained in step h to the receptacle (310) of the third filtration unit, and passing the permeate obtained in step h to a fourth filtration unit (40), wherein the fourth filtration unit (40) comprises a receptacle (410) and a filter membrane (420), wherein the permeate obtained in step h enters the receptacle (410) of the fourth filtration unit (40); j. passing fluid from the receptacle (410) of the fourth filtration unit through the filter membrane (420) of the fourth filtration unit to obtain a retentate and a permeate; and i. creating a second counter-current flow by re-circulating a fraction of the feed fluid from the receptacle (310) of the third filtration unit (30) to the receptacle of the second filtration unit; and k. recovering the molecule of interest from the fourth filtration unit; and optionally l. re-circulating the permeate obtained in the fourth filtration unit to the first filtration unit as a wash liquid.

[0241] Embodiment 103. The method according to embodiment 101 or embodiment 102, wherein each receptacle is a feed tank or a mixing unit and / or wherein each counter-current flow is created by passing a portion of the fluid from a subsequent feed tank to a preceding feed tank.

[0242] Embodiment 104. The method according to any one of embodiments 101 to 103, wherein the molecule of interest is a recombinantly produced polypeptide or polypeptide precursor.

[0243] Embodiment 105. The method according to any one of embodiments 101 to 104, wherein the molecule of interest is a GLP-1 peptide, such as an extended GLP-1 peptide, such as polypeptide 1.

[0244] Embodiment 106. The method according to any one of embodiments 101 to 105, wherein the molecule of interest is an amylin analogue or precursor thereof, such as polypeptide 2.

[0245] Embodiment 107. The method according to any one of embodiments 101 to 106, wherein each counter-current flow individually is in the range of

[0246] 0.01 0.99.

[0247] Embodiment 108. The method according to any one of embodiments 101 to 107, wherein each counter-current flow individually is in the range of

[0248] 0.05 < — < 0.95, such as 0.1 < — < 0.90.

[0249] 1+y 1+y

[0250] Embodiment 109. The method according to any one of embodiments 101 to 108, wherein the impurity is / are host cell proteins (HCP).

[0251] Embodiment 110. The method according to any one of embodiments 101 to 109, wherein the molecule of interest has been recombinantly produced, optionally wherein the molecule of interest has been produced in yeast, such as in s. cerevisiae. Embodiment 111. The method according to any one of embodiments 101 to 110, wherein at least one filter membrane has a separation factor in the range of 2-100, optionally wherein one or more filter membrane(s) has / have a MWCO selected from the list consisting of about 30 kDa, about 40 kDa and about 50 kDa.

[0252] Embodiment 112. The method according to any one of embodiments 101 to 111, wherein the separation factor of the first filter membrane differs from the separation factor of the fourth filter membrane.

[0253] Embodiment 113. The method according to any one of embodiments 101 to 112, wherein the fourth filter membrane has a MWCO of about 5 kDa.

[0254] Embodiment 114. The method according to any one of embodiments 101 to 113, wherein the feed fluid passes through a filter membrane to reduce NTU before entering the first filtration unit.

[0255] Embodiment 115. The method according to any one of embodiments 101 to 114, wherein each receptacle individually is kept at a temperature in the range of -5 °C to 15 °C, such as 0 °C to 10 °C or at a range of 30°C to 85 °C, such as 35 °C to 55 °C or 35 °C to 45 °C.

[0256] Embodiment 116. The method according to any one of the embodiments 101 to 115, wherein the method results in 0.5 to 10 log reduction in HCP, optionally in immunogenic HCP.

[0257] Embodiment 117. The method according to any one of the embodiments 101 to 115, wherein the method results in 1 to 10 log reduction in HCP, optionally in immunogenic HCP.

[0258] Embodiment 118. The method according to any one of the embodiments 101 to 115, wherein the method results in 2 to 10 log reduction in HCP, optionally in immunogenic HCP.

[0259] Embodiment 119. The method according to any one of the embodiments 101 to 115, wherein the method results in 3 to 10 log reduction in HCP, optionally in immunogenic HCP.

[0260] Embodiment 120. The method according to any one of the embodiments 101 to 115, wherein the method results in 4 to 10 log reduction in HCP, optionally in immunogenic HCP.

[0261] Embodiment 121. The method according to any one of the embodiments 101 to 115, wherein the method results in 5 to 10 log reduction in HCP, optionally in immunogenic HCP.

[0262] Embodiment 122. The method according to any one of the embodiments 101 to 115, wherein the method results in 6 to 10 log reduction in HCP, optionally in immunogenic HCP. Embodiment 122. The method according to any one of the embodiments 101 to 115, wherein the method results in 2-5 log, such as 3-5 log reduction in HCP, optionally in immunogenic HCP.

[0263] Embodiment 123. The method according to any one of the embodiments 101 to 115, wherein the last filter membrane may be replaced by a chromatography column.

[0264] Embodiment 124. The method according to any one of the embodiments 101 to 115, wherein a chromatography column is added after the last filter.

[0265] Embodiment 126. The method according to embodiment 125 or embodiment 126, wherein the chromatography column is selected from the list consisting of a reversed phase column, an ion exchanger, a hydrophobic interaction chromatography column, an affinity column, and a size-exclusion column.

[0266] Embodiment 126. The method according to embodiment 125 or embodiment 126, wherein the chromatography column is an anion or a cation exchanger.

[0267] Embodiment 127. The method according to any one of embodiments 1 to 34, wherein the first filter membrane has a molecular weight cut-off selected from the range of 1-500 kDa, such as 1-200 kDa, such as 1-150 kDa.

[0268] Embodiment 128. The method according to any one of embodiments 1 to 45, wherein the third filter membrane has a molecular weight cut-off selected from the range of 1-500 kDa, such as 1-200 kDa, such as 1-150 kDa.

[0269] Embodiment 129. Use of a counter-current filtration system as defined in any one of embodiments 71 to 95 for the removal of an impurity in a protein preparation, wherein the protein comprises a molecule of interest and the impurity.

[0270] Embodiment 130. The use according to embodiment 129, wherein the impurity is HCP.

[0271] Embodiment 131. The use according to embodiment 129 or embodiment 130, wherein the molecule of interest is a GLP-1 analogue or an amylin analogue.

[0272] Embodiment 132. The use according to embodiment 129 or embodiment 130, wherein the molecule of interest is semaglutide or a precursor thereof.

[0273] Embodiment 133. The use according to embodiment 129 or embodiment 130, wherein the molecule of interest is cagrilintide or a precursor thereof. Examples

[0274] Abbreviations: ACN acetonitrile

[0275] TFA trifluoroacetic acid

[0276] Table 1. Symbol list for equations 1 to 20.

[0277] 1where i-1 indicates a step upstream (preceding) from “i” and i+1 indicates a step downstream (subsequent) from “i”. For instance, referring to Fig. 2, filtration unit (100) could be referred to as filtration unit “i-1”, filtration unit (200) could be referred to as filtration unit “i” and filtration unit (300) could be referred to as filtration unit “i+1 ”.

[0278] 2“j” is used to refer to any component. For instance, the molecule of interest can be referred to as “j”, but also the impurity can be referred to as “j”. However, in specific instances “j” will be replaced by “A” to refer to the molecule of interest and “X” will be used to refer to the impurity. This is only used to help readability of the equations.

[0279] Materials and Methods

[0280] Polypeptide 1 is an extended semaglutide precursor (extended [Arg34]GLP- 1(9-37) and has a molecular weight of 4712 g / mol.

[0281] Polypeptide 2 is an extended cagrilintide precursor (extended [Glu14, Arg17,Pro37]- pramlintide.) and has a molecular weight of 4555 g / mol.

[0282] General Methods of Detection and Characterisation

[0283] LIPLC method 1 and 2 can be used to determine the area of the specific components: HCP, polypeptide 1 , and polypeptide 2.

[0284] The area of the specific components {e.g., HCP + extended GLP-1) is used to calculate the ratio between the specific components.

[0285] LIPLC method 1

[0286] • Waters LIPLC system with UV detector with a Waters CSH130 C18, 50x2.1 mm (1,7pm, 130A) column with a column temperature of 50 °C

[0287] • flow rate of 0.5 mL / min

[0288] • injector volume of 10 pL.

[0289] • Mobile phase A: 0.05% TFA in water

[0290] • Mobile phase B consists of 0.045% v / v TFA in 80% in ACN. Elute samples with a gradient of 20% B to 70% B at 22.5 min.

[0291] • tR(polypeptide 1): ~ 11 min (see Fig. 1)

[0292] • tR(polypeptide 2): ~ 5.5 min (see Fig. 7)

[0293] • tR(HCP): ~ 12-18 min

[0294] UPLC method 2:

[0295] Corresponds to UPLC method 1 but using a flow rate of 0.7 mL / min and eluting the samples with a gradient of 20% B to 70% B at 7.66 min.

[0296] • tR(polypeptide 1): ~ 4.5 min (see Fig. 1)

[0297] • tR(HCP): ~ 5-7 min HCP is normally analysed by ELISA, and this is used as the validated analytical method, however in the UPLC-method the HCP normally elutes as a number of peaks / broad peak after the main peak. This area is often used as a fast indication of the reduction in HCP in the filtrations.

[0298] HCP quantification using ELISA

[0299] Sandwich ELISA was used to quantify the ratio between the molecule of interest (e.g., polypeptide 1, polypeptide 2) and the impurity (e.g., HCP).

[0300] To quantitatively assess the level of immunogenic HCP, a sandwich-type enzyme-linked immunosorbent assays (ELISA) was employed. The assay was developed as a platform ELISA assay for recombinant peptides and in particular GLP-1 recombinant peptides according to guidelines, e.g. Ph. Eur 2.6.34.

[0301] The pool of HCP antigens for producing antibodies for the ELISA assay were derived from a mock run of the active substance manufacturing process up to a step capable of generating a broad spectrum of HCP in sufficient quantities. Antibodies generated in rabbits were tested for coverage of the HCP present in samples from the polypeptide 1 manufacturing process. The pool of HCP antigens was also used as HCP reference standard in the final assay.

[0302] In brief, the assay was performed by coating 96 well microtiter plates with the selected pool of anti HCP antibodies. The coated plates were incubated with a dilution series of the reference standard, internal control samples, and samples to be measured. After washing 3 times to remove unbound proteins, the bound HCP were detected using a combination of biotinylated pool of antibodies and horseradish conjugated streptavidin followed by TMB (3,3’,5,5’-tetramethylbenzidine) to develop a colour reaction. The amount of HCP in the samples was back calculated for the intensity of the colour reaction of the calibration curve generated by the dilution series of the reference standard.

[0303] General Methods of Selecting a Filter

[0304] 1) Selecting a filter membrane to separate the molecule of interest from the impurity(ies)

[0305] At the outset, the separation when using a filter membrane is based on the size of the molecules to be separated and the respective pore size of the filter. Put differently, the pore size of the filter determines the transmission of the molecules to be separated through the membrane. A filter having a pore size which is smaller than molecules to be separated (where the molecules to be separated are the molecule(s) of interested and the impurity(ies)) will retain both molecules to be separated and hence no separation occurs. The transmission will be close to 0. Conversely, a filter membrane having a pore size larger than the molecules to be separated will allow them to pass the membrane and give a transmission for both product and impurities close to 1. However, no satisfactory separation will occur.

[0306] For the separation to work satisfactorily there should be a difference in the transmission of the molecule of interest and the transmission of the impurity. The ratio between the concentrations of the respective components on the retentate and permeate side is given in the separation factor, a. where cR,xis the concentration of the impurity in the retentate;

[0307] CR, is the concentration of the molecule of interest in the retentate; cP,xis the concentration of the impurity in the permeate; and

[0308] Cp is the concentration of the product in the permeate.

[0309] The concentration ratio is chosen such that the value of the separation factor is greater than 1.

[0310] A selected filter membrane should preferably have high separation factor between the product and the impurity and a low transmission of HCP.

[0311] The transmission of the product is given by

[0312] T>CP,A L r,A ~ R,A

[0313] This should preferably also be high since this eases the transport of product through the series of filter membrane, and avoids build-up of high concentrations.

[0314] The transmission, Trj is defined as the ratio of the concentration of the specific component, j, e.g. the molecule of interest (“A“) or the impurity (“X”), in the permeate and the concentration of the specific component, j, in the retentate. The transmission indicates how much of a specific component in the feed stream can pass through the membrane and can be expressed as percentage. The higher the transmission, the more of the component can pass through the membrane. For instance, a “medium pore filter” would be selected such that the molecule of interest could easily pass through, but the impurity would be mostly retained.

[0315] A membrane with a high flux is also desirable since this reduces the overall size of the equipment.

[0316] Proof of concept

[0317] The purpose of the experiment was to investigate if repeated filtrations of a mixture of product and HCP lead to further reductions compared to just a single filtration.

[0318] At the outset, the ratio between HCP (component X) and polypeptide 1 (component A) in the feed stream was estimated via LIPLC (using analytical method 01). The HCP content was estimated from the area eluting in the interval around 14-18 min. polypeptide 1 eluted at around 11 min (see Fig.1).

[0319] A filtration system was set up comprising a first filtration unit with a 30 kDa Hydrosart filter membrane and a second filtration unit with a 5 kDa Hydrosart filter membrane. The second filtration unit was connected such that it received the permeate from the first filtration unit. The process was run in batch mode at ambient temperature and without added countercurrent flow.

[0320] Feed fluid: about 2 litres of fermentation broth comprising polypeptide 1 where the yeast cells have been removed and the pH has been adjusted to pH 8.5.

[0321] Method:

[0322] The feed fluid was added to the feed tank of the first filtration unit after which the pumps were started. Initially the permeate from the 5 kDa filter was collected in the 5 kDa permeate tank to concentrate the product in the 30 kDa feed tank, and later the permeate was recycled back to the 30 kDa feed tank to wash out the product from the first tank (“step-wise process”).

[0323] The content of the feed tank of the second filtration unit was collected and analyzed after the first filtration (“feed filtered once”). The result can be seen in Figure 1 : “filtered once”. Subsequently, the feed tank of the first filtration unit was emptied and cleaned whereafter the “feed filtered once” was added to the first filtration unit and the filtration process was run again. The result of this second filtration step can be seen in Figure 1 : “filtered twice”.

[0324] Referring to Fig. 1, the ratio is given as a ratio between HCP area and polypeptide 1 area, which is calculated as the area of the peaks in the HCP-range divided by the area of the main peak.

[0325] As can be seen from Figure 1, the ratio was measured to be 0.87. After the first filtration the ratio was 0.18. The impurity / product-ratio was hence reduced by approximately a factor 5. After the second filtration, the ratio was reduced from 0.18 to 0.03 corresponding to a factor 6 reduction.

[0326] Conclusion:

[0327] Multiple filtrations can lead to a further reduction in the ratio between HCP and product.

[0328] General equations for a counter-current filtration

[0329] MASS BALANCE

[0330] Adding multiple filters in series the mass balances can be set up under simplified assumptions to study the expected behaviour of this separation process.

[0331] At the outset, the following assumptions are made for derivation of the equations:

[0332] • mass is conserved for each element or compound on either molar or weight basis;

[0333] • all volumes in filters and filtration units are constant; and

[0334] • fluids are incompressible and have the same density.

[0335] Mathematically the mass balance for a system without a chemical reaction can be defined by equation (8):

[0336] Total Input = Total Output + Accumulation (10), where the accumulation term accounts for the change in the amount of a substance in a system of over time due to the addition or removal of that substance (e.g., component j). The accumulation term therefore represents the next input (or output) of the substance in the system. The purpose of including the accumulation term in the mass balance equation is to ensure that the equation accurately reflects the changes in the amount of the substance in the system and considers any inflows or outflows that may occur.

[0337] The system is any process or portion of a process chosen for analysis. A system is said to be “open” if material flows across the system boundary during the interval of the time being studies; “closed” if there are no flows in or out. Accumulation is usually the rate of change of holdup of material within the system. If material is increasing, accumulation is positive; if it is decreasing, it is negative. If the system does not change with time, it is said to be at steady state, and the net accumulation will be zero.

[0338] For the processes described herein, a mass balance equation is applied to each control volumes, the control volume for the filter and the control volume for the filtration unit. It follows that two mass balances are required:

[0339] 1) Mass balance around the filter;

[0340] 2) Mass balance around the filtration unit.

[0341] Each of these mass balances can be established for the total mass and the individual component(s) (e.g., impurity (“X”), product (“A”)).

[0342] The relationship between the mass flow and the volume flow depends on the density of the substance being transported. If the density of the substance is constant, then the mass flow rate and the volume flow rate are directly proportional and can be related by equation (11): m = pQ (11), where m is the mass flow rate; p is the density of the substance; and

[0343] Q is the volume flow rate.

[0344] 1) Mass balance around the filter

[0345] 1. 1 Total mass balance around the filter

[0346] Based on the assumption that the fluids are incompressible and that the density is constant, the mass balance around the filter can be described by equation (12):

[0347] QF = QR + Qp (12), where QF defines the flow of the feed entering the filter, QRdefines the flow of the retentate, and QPdefines the flow of the permeate.

[0348] With reference to equation (10) and taking into account equation (11), “QF” corresponds to “total input” and the sum of “QR” and “QP” corresponds to “total output”. As can be seen in Figure 8, QF is the only volumetric flow entering the filter, but both volumetric flows, QR and QP, respectively, are flowing out of the filter, and hence their sum corresponds to the “total output”.

[0349] It follows from equation (12) that the flow entering the filter (QF) must be equal to the sum of the flow leaving the filter (QR + QP).

[0350] 1.2 Mass balance around the filter for the individual components

[0351] As explained above, the mass balance can also be defined for each individual component of the feed using equation (13):

[0352] QF • cF,j = QR • cR+ Qp • cPJ(13), where

[0353] CFJ defines the concentration of component j in the feed entering the filter,

[0354] CRJ defines the concentration of component j in the retentate, and Cp defines the concentration of component j in the permeate.

[0355] As can be derived from equation (13), the flow comprising component j (CFJ) entering the filter corresponds to the sum flows comprising component j (CRJ + CPJ) leaving the filter.

[0356] The concentration of component j in the retentate and the permeate depends on the transmission, Trj, of the filter. Hence, the relation between the concentration of component j in the respective outflow feeds, QPJ and QRJ can be calculated based on equation (14). Put differently, the transmission, Trj of each component j defines how much of each component j is able to pass through the filter.

[0357] Combining equations (10) to (14) results in equation (15):

[0358] The fraction in equation (15) can then be replaced by “P resulting in equation (16) showing the relationship between the feed concentration and the retentate concentration.

[0359] CF,J = Pj • cR,j (16)

[0360] Normally the retentate flow is significantly higher, such as for instance ten times higher, than the permeate flow and j is close to one.

[0361] BATCH PROCESS

[0362] Batch processes operate to a batch cycle and are therefore non-steady state. Materials are added to the system in one operation, then the process is carried out, and then the batch cycle can be repeated.

[0363] 1) Mass balance for the filtration unit

[0364] In the following equations the index j for the components is omitted for readability.

[0365] Figure 8 shows a batch process comprising three filtration units in series, filtration unit “i”, filtration unit “i-1” preceding filtration unit “i”, and filtration unit “i+1” following filtration unit “i”. The exemplary filtration unit “i” is indicated by the dotted line.

[0366] 1.1 Total mass balance for the filtration unit

[0367] For calculating the total mass balance for a filtration unit / , it is assumed that the total volume within the filtration unit / is kept constant and no accumulation of total mass occurs. It therefore follows that the sum of the flows entering the filtration unit / must be equal to the sum of the flows leaving the filtration unit / . Hence, the mass balance for a filtration unit / can be calculated using equation (17):

[0368] Qp,i-i + QB,i+i= QB.I + QP.I (17), where QP -1 defines the permeate flow from the previous filtration unit (i-1) entering the filtration unit i,

[0369] QB +I defines the counter-current flow from the following filtration unit (i+1) entering the filtration unit i,

[0370] QBJ defines the counter-current flow / leaving the filtration unit / , and

[0371] Qp,i defines permeate / leaving the filtration unit / (as shown in Figure 8).

[0372] It is noted that the counter-current and permeate flows are related as shown in equation (18):

[0373] Qp,o=Qp,t ~ QB,I+I=QP,N-I=QP,N (^8) where

[0374] QP,O corresponds to the flow going into the first filter, which in Figure 2 is identical to the permeate flow from the last filter hence QP,O=QP,N.

[0375] Isolating QP and defining Y QBJ / QP.O the permeate from unit i can be defined as shown in equation (19):

[0376] Qp,i=QP,0 + QB,I+1=Qp,o( + Ti+1) (19)

[0377] 1.2 Mass balance for the filtration unit for the individual components

[0378] It is assumed that for each component the transmission is the same for all filters of the same type. It is assumed that product and impurities are fully soluble and do not interact with each other.

[0379] The total mass / volume is constant in each filtration unit. The masses of the components however are not constant. The four flows (QP, M; QB +I, QB , QPJ) as shown in eq. (17) each have a concentration with the same index as shown eq. (13). Considering a time interval from t to t+At, each of these terms will have the form Q c-At. The accumulation term will be the final volume time the final concentration minus the initial volume times the initial concentration. The corresponding component balance for filtration unit / is given by equation

[0380] (20), where:

[0381] At represents the change in time between two time points. Dividing equation (20) by At and lim results in an equation to calculate the instantaneous M->0 rate of change of the mass within the control volume, namely equation (21):

[0382] Since the volume, Vi, was assumed constant, the equation (21) for results in equation (22):

[0383] Combining equation (22), equation (16), and equation (14) leads to equation (23) and looking at Figure 8 it can be seen that the counter-current flow is taken from the feed tank giving CB = CF. These expressions for concentrations can be inserted to express all in retentate concentrations:

[0384] Inserting the expression for the permeate flow and the counter-current flow results in equation (24) after dividing by QP,O and

[0385] The parameters on the right side are now expressed as dimensionless variables. This can also be done on the left side by introducing the dimensionless variable, 0, as shown in equation (25) with the derivative as shown in equation

[0386] Insertion in equation (24) gives equation (27) which after multiplication by Vi / Vj gives equation (28):

[0387] Equation (28) is the equation for one filtration unit, / , with counter-current flow.

[0388] Analyzing a number of filtration units connected in series gives a set of coupled differential equations which, in matrix form, result in Equation (22):

[0389] Here all transmission for the medium pore size have been set without index and the low cutoff has been set for i=4 corresponding to the Fig. 2.

[0390] Having both product and impurities initially in the first tank the initial concentration vector will be c=[ci nit, 0, 0, 0] where Ci ,init is the initial concentration in the first tank.

[0391] Exemplary solutions to this matrix using two different values of y are set out below, which demonstrate the effect of the counter current flow, and how the counter-current flow reduces the transfer of the impurities to the last feed tank.

[0392] A) Modelling Example A with v > 0 (Figs. 4A-D)

[0393] The modelling example is based on the setup shown in Fig.10 using equation (29). The parameters selected for the modelling are shown in Table 2.

[0394] The calculations are made for 0 up to 20, corresponding to a permeate volume into the first filter 20 times the volume in “Feed 1”, and y (y2,Ys) being set to 0.25. Theta was set to 20 to assure in the modelling calculation that the total amount of molecule of interest reaches the last filtration unit and to account for the added volume. It will be understood that by adding or removing filtration units (and hence filters) the separation can be influenced. A quick modelling using equation (29) will guide on how many filtration units are necessary to obtain the desired separation. Table 2. Selected parameters.

[0395] The results are shown in Figs. 4A-D (“Example 02B”), where Fig. 4A and Fig. 4B show the product concentration in normal and semi-log plot, respectively (corresponding to a transmission of 50% of the product) and where Fig. 4C and Fig. 4D show the impurity concentration (HCP) in normal and semi-log plot, respectively (corresponding to a transmission of 5% of the impurity).

[0396] Referring to Fig. 4C it can be seen that the respective slopes are close to zero which reflects the hampered transfer of impurities from the first tank to the last. It additionally shows that the concentrations will reach an almost "steady state” I “pseudo steady state” after the initial change in impurity concentration. The large distance between the lines suggests a significant reduction of the impurities in each filtration unit during the process. Put differently, each filtration unit results in a significant reduction of impurities.

[0397] B) Modelling Example B with y = 0 (Figs. 3A-D)

[0398] To investigate the effect of the added counter-current flow in the filtration process, the modelling example described above under A) was repeated, but setting y=0.

[0399] The results are shown in Figs. 3A-D, where Fig. 3A and Fig. 3B show the product concentration in normal and semi-log plot, respectively (corresponding to a transmission of 50% of the product) and where Fig. 3C and Fig. 3D show the impurity concentration in normal and semi-log plot, respectively (corresponding to a transmission of 5% of the impurity). As can be seen in Figs. 3A-D, the product / impurity is washed out of the first filter and transferred to the second filter and is then washed out of the second filter and transferred to the third filter before the product is collected in filtration unit 4. Looking at the y-axis of the semi-log plots, it can be seen that both, the concentration of the product and the concentration of the impurity, decreases exponentially, and the intermediate filters have the same slope as the first filter and a short distance between the lines. As can be seen from Figs. 3A-D adding filters in series without an added counter-current flow does not lead to a good separation between product / molecule of interest and impurity. Conversely, impurity and product / molecule of interest end up in the last filtration tank.

[0400] It can be seen in Fig. 3C that without counter-current flow, the impurity concentration in “Filter 1” at 0=20 is below 40% of the initial concentration, which means that more than 60% of the impurities have been washed out of Filter 1. In fact, looking at Filter 4, the amount of impurity is just below 60%. Comparing the product and the impurity plots, Figs. 3A and 3C, respectively, two left plots, at the range 71 <0<1O it can be seen that the product concentration in filtration unit 4 only increases slightly. Conversely, the concentration of the impurities increases with the increasing wash volume 0. This leads to an increased ratio between impurities and product almost following the curve for the impurities in “Feed 4”. It follows that with increased wash volume the impurity will start to accumulate in the last filtration unit together with the product. Hence, there will not be any efficient separation between product and impurity.

[0401] Conclusion

[0402] Comparing the semi-log plots (impurity) for both modelled processes, i.e., with added counter-current and without, (see Fig. 4D and Fig. 3D, respectively) it can be seen how the curves for y=0 have the same slope and the distances between each curve (filter 1, filter 2, filter 3, filter 4) is small (Fid. 3D). The addition of a counter-current flow corresponding to Y2=Y3=0.25 leads to a significantly lower slope (close to 0) and a significantly larger distance between the curves (Fig. 4D). The smaller slope reflects the hampered transfer of impurities from the first tank to the last, meaning that the concentrations will reach an almost "steady state” I “pseudo steady state”. The large distance between the curves shows a significant reduction of the impurities over each filtration unit in the process. Each filtration unit gives a reduction in impurity level. Modelling also shows that additional filtration units with medium pores reduces the impurity level in the last tank event further. Hence, using multiple filtration units and adding a counter-current will results in both high yield and high purity.

[0403] Example 1 : Removal of host cells protein from a feed stream comprising an extended peptide precursor

[0404] The purpose of the experiment was experimentally to verify the results from modelling namely to investigate if a molecule of interest (e.g., a peptide) and an impurity (e.g., HCP) could be separated effectively using the suggested setup with counter-current flow.

[0405] Example 1.1 - Removal of host cell proteins from a feed stream comprising a GLP-1 precursor

[0406] Molecule of interest: polypeptide 1

[0407] Impurity: HCP

[0408] Setup

[0409] The parameters used for the filtration are given in Table 3.

[0410] Table 3. Selected parameters. All filters were Hydrosart. TWB is water bath temp, connected to the feed tanks.

[0411] Equipment:

[0412] • jacketed Duran® Laboratory bottles

[0413] • Longer micro-gear pumps were used to created circulation over the filter membranes

[0414] • a Watson Marlow 323 was used to create a counter-current flow. polypeptide 1 was expressed in yeast. The yeast cells were then removed before centrifugation and the pH was adjusted to pH 8.5. Approximately 3.75 L of centrifuged and pH adjusted starting material were used (and fed into filtration unit 1).

[0415] The feed tank of the first filtration unit had a volume of 2L, which meant that the starting material was added “step-wise”, i.e., as filtration progressed more volume could be added to the feed tank in filtration unit 1.

[0416] The feed tanks in filtration units 2 and 3 had an initial volume 0 but were filled to the values given in the table during the initial part of the experiment.

[0417] Tank 4 was also initially empty. The 5kDa permeate was collected initially to reduce the volume. Once the feed volume was reduced, the permeate obtainable in the fourth filtration unit could be re-circulated to the first filtration unit as wash liquid.

[0418] The volumes were kept almost constant in the feed tanks of filtration units 2-4 after the initial filling process.

[0419] Only the volume in the feed tank of the first filtration unit was reduced, as can be seen in the Fig. 5. For practical reasons the volume was hence not kept constant during the experiment, as assumed when deriving the equations for the batch-process.

[0420] At the end of the experiment the product was collected from the last feed tank.

[0421] Results

[0422] Permeate and retentate samples were taken from all four filter filtration units at different points in time. All samples were analyzed by LIPLC.

[0423] The Fig. 5 shows overlay chromatograms of samples taken from the retentate and permeate portion of the feed fluid from filtration units 1 to 4, respectively, at the same time points.

[0424] Polypeptide 1 elutes in a sharp peak around 4 to 4.5 min and the HCP elute in the range of 5 to 7 min in the plots (LIPLC method 2). The transmission of polypeptide 2 and HCP can be estimated by the areas below the curves for the two samples.

[0425] Each row represents a time point, and each column represents a filtration unit. The time and filtration unit are given in the title of each plot. Looking at column 2, it can be seen that the concentration of polypeptide 1 decreases in the retentate during the run (i.e., when comparing an earlier time point to a later time point) while the HCP peak stays almost at a constant level - after an initial increase - as predicted in the modelling example.

[0426] Turning now to the last row, it can be seen that the HCP is mainly concentrated in filter one and some of the HCP can be seen in filter two. Filter four has a very low level of impurities.

[0427] Conclusion

[0428] The hypothesis, based on the modelling experiment, that it should be possible to wash out a molecule of interest from a first filtration unit to a last filtration unit while keeping the impurity in the first filtration unit(s) has been confirmed experimentally.

[0429] Example 1.2 - Removal of host cell proteins from a feed stream comprising an amylin analog precursor (polypeptide 2).

[0430] The setup and equipment used was the same as for Example 1.1 except for some parameters as shown in Table 4.

[0431] Table 4. Selected parameters. All filters were Hydrosart. TWB is water bath temp, connected to the feed tanks. polypeptide 2 was expressed recombinantly in yeast, centrifuged to remove the yeast cells, and filtered through a 100 / 5 kDa Hydrosart filter membrane to lower the NTU. The pH was adjusted to approx. 8.5.

[0432] After an initial filling of the feed tanks of the respective filtration units 2 to 4 with the permeate from filtration unit 1, the volume in the feed tank of filtration unit 1 was kept constant around 700 mL during the rest of the experiment. Several permeate samples from unit 10 were collected at different time points, and the UPLC-areas of the molecule of interest and HCP was calculated and plotted as a function of time in Fig. 6. The figure shows the exponential wash-out of the molecule of interest and an almost constant level of HCP, hence confirming the results from the modelling.

[0433] Fig. 7 shows an overlay of chromatograms of samples taken at:

[0434] • the start of the process ( / .e., from feed stream initially fed into filtration unit 1) - “start" (sample 1);

[0435] • the HCP-rich fraction (at the feed tank 1 - end of experiment) of filtration unit 1 at the end of the process - “HCP” (sample 2), and

[0436] • the product-enriched fraction of filtration unit 4 at the end of the process - “product” (sample 3).

[0437] Looking at Fig. 7, it can be seen that sample 1 contained both the product eluting at round 5 min and HCP eluting at around 12-17 min. Looking at the chromatogram of sample 2, it can be seen that the peak eluting in the 12-17 min range is similar in height to the one in sample 1. In addition, only trace amount of product is found in the sample. Hence, it can be assumed that the process also is a high yielding process. Similarly, the chromatogram of sample 3 almost exclusively shows that product peak which elutes at around 5 minutes and thus indicating that HCP have successfully been removed.

[0438] The difference between the impurities / product areas ratio is 10,000 where the ratio between the impurities, X, and the product, A, is given by

[0439] AreaxRatio = -

[0440] AreaAindicating a good separation.

[0441] The process ran to almost all product was washed out of feed tank 1. At the end of the run the permeate line was closed, and the feed pumps continued to run.

[0442] After harvest of the material from feed tank 4, water was added to wash out / displace the liquid in the filter and the dead volumes. The mass balance is shown in Table 5 . It is seen that almost all product ends up in the feed 4 tank, and the wash of the filter 4, which is also harvested as product. Less than one percent of the product is at the end in feed 1 to 3.

[0443] Table 5 Mass balance from experiment.

[0444] 1HPLC samples were taken at the start and at the end point of the process form the respective feed tanks (see Fig. 10).

[0445] Conclusion:

[0446] The data confirms the exponential wash-out of the molecule of interest and an almost constant level of the HCP, as predicted by the mode.

[0447] The suggested separation method having multiple filtration unit in series and an added counter-current flow efficiently separates a molecule of interest (i.e. , polypeptide 2) and an impurity (i.e., HCP) and in good yield.

[0448] Example 2: Determination of immunogenic HCP levels using ELISA

[0449] Polypeptide 1 was prepared as described in Example 1.1.

[0450] Setup

[0451] Polypeptide 1 was expressed in yeast, the yeast cells were removed by centrifugation and the centrifuged material was pre-filtered using a large pore filter membrane (e.g., a 100 kDa filter membrane) to reduce NTU.

[0452] The filter setup consisted of four filtration units with Hydrosart filter membrane, the pore sizes are given in the Table 6. The volume in the feed tank of filtration unit 1 was reduced during the experiment and was hence not kept constant. At the end of the experiment the product in the last filtration unit was concentrated by collecting the permeate in the permeate tank (e.g., as shown in Fig. 10).

[0453] Table 6. Selected parameters. All filters were Hydrosart. TWB is water bath temp, connected to the feed tanks.

[0454] Results

[0455] Table 7 shows the results from the ELISA and LIPLC measurement of the starting material (feed fluid) as well as for the samples taken at the end of the experiment at each feed tank.

[0456] The reductions over the two filters with counter-current flow is seen to be around two orders of magnitude.

[0457] As can be seen from Table 7 the concentration of the product was around 0.1 g / L for the intermediate tanks (e.g., feed tanks of filtration units 2 and 3 and around 9.3 g / L for the feed tank of the last filtration unit indicating a high yield.

[0458] Table 7. Analytical results from ELISA and UPLC and ratio between these concentrations for the starting material, and samples taken at the end of the experiment from each of the feed tanks.

[0459] Sample1HCP [ng / mL] c [g / L] HCP / c [ng / mg]

[0460] Start 458,513 2.2 208,415

[0461] Feed 1 2,340,250 0.1 23,402,500

[0462] Feed 2 20,699 0.1 206,990

[0463] Feed 3 324 0.1 3,240

[0464] Feed 4 176 9.3 19

[0465] 1The samples “Feed / ” (where i=1 ,2,3, and 4) were taken at the end of the process from each of the feed tanks of the respective filtration units. The sampel “start” is taken from the feed fluid at the beginning of the process.

[0466] Conclusion Table 7 shows that the HCP / product-ratio can be reduced four orders of magnitude by counter-current filtration. More specifically, the two steps with the added counter-current flow reduces the ratio with about two orders of magnitude. While certain features of the invention have been illustrated and described herein, many modifications, substitutions, changes, and equivalents will now occur to those of ordinary skill in the art. It is, therefore, to be understood that the appended claims are intended to cover all such modifications and changes as fall within the true spirit of the invention.

Claims

CLAIMS1. A method of removing or reducing an impurity in a protein preparation comprising a molecule of interest recombinantly produced in a host cell, the method comprising the steps of: a. introducing the protein preparation as a fluid into a first filtration unit (10), wherein the first filtration unit comprises a receptacle (110) and a filter membrane (120); b. passing the fluid through the filter membrane (120) of the first filtration unit (10) to obtain a retentate and a permeate; c. re-circulating the retentate obtainable in step b to the receptacle (110) of the first filtration unit (10), and passing the permeate obtainable in step b to a second filtration unit (20), wherein the second filtration unit comprises a receptacle (210), and a filter membrane (220), and wherein the permeate obtainable in step b moves into the receptacle (210) of the second filtration unit; d. passing fluid from the receptacle (210) of the second filtration unit (20) through the filter membrane (220) of the second filtration unit (20) to obtain a retentate and a permeate; e. re-circulating the retentate obtainable in step d to the receptacle (210) of the second filtration unit (20), and f. collecting the molecule of interest from the second filtration unit or passing the permeate obtainable in step d to a third filtration unit (30); and g. creating a first counter-current flow by re-circulating a fraction of the fluid from the receptacle (210) of the second filtration unit (20) to the receptacle (110) of the first filtration unit (10); and wherein the molecule of interest is a recombinantly produced polypeptide, polypeptide precursor or extended polypeptide.

2. The method according to claim 1, wherein step f. consists of passing the permeate obtainable in step d to the third filtration unit (30), and wherein the third filtration unit comprises a receptacle (310) and a filter membrane (320), and wherein the permeate obtainable in step d enters the receptacle (310) of the third filtration unit; and wherein the method further comprises h. passing fluid from the receptacle (310) of the third filtration unit through the filter membrane (320) of the third filtration unit to obtain a retentate and a permeate;i. re-circulating the retentate obtainable in step h to the receptacle (310) of the third filtration unit, and passing the permeate obtainable in step h to a fourth filtration unit (40), wherein the fourth filtration unit (40) comprises a receptacle (410) and a filter membrane (420), wherein the permeate obtainable in step h enters the receptacle (410) of the fourth filtration unit (40); j. passing fluid from the receptacle (410) of the fourth filtration unit through the filter membrane (420) of the fourth filtration unit to obtain a retentate and a permeate; and i. creating a second counter-current flow by re-circulating a fraction of the fluid from the receptacle (310) of the third filtration unit (30) to the receptacle of the second filtration unit; and k. collecting the molecule of interest from the fourth filtration unit or passing the permeate obtainable in step j to a subsequent filtration unit; and optionally l. re-circulating the permeate obtainable in the fourth filtration unit to the first filtration unit as a wash liquid.

3. The method according to claim 1 or claim 2, wherein each receptacle individually is a feed tank or a mixing unit.

4. The method according to any one of claims 1 to 3, wherein the molecule of interest is a GLP-1 peptide, such as an extended GLP-1 peptide, such as polypeptide 1.

5. The method according to any one of claims 1 to 4, wherein the molecule of interest is an amylin analogue or precursor thereof, such as polypeptide 2.

6. The method according to any one of claims 1 to 5, wherein each counter-current flow individually is in the range of7. The method according to any one of claims 1 to 6, wherein each counter-current flow individually is in the range of0.05 < — 1+y < 0.95, such as 0.1 < — 1+y < 0.90.

8. The method according to any one of claims 1 to 7, wherein the impurity is / are host cell proteins (HCP).

9. The method according to any one of claims 1 to 8, wherein the molecule of interest has been produced in yeast, such as in s. cerevisiae.

10. The method according to any one of claims 1 to 9, wherein at least one filter membrane has a separation factor in the range of 2-100.

11. The method according to any one of claims 1 to 10, wherein one or more filter membrane(s) has / have a MWCO selected from the list consisting of about 30 kDa, about 40 kDa and about 50 kDa.

12. The method according to any one of claims 2 to 11, wherein the separation factor of the first filter membrane differs from the separation factor of the fourth filter membrane.

13. The method according to any one of claims 2 to 12, wherein the fourth filter membrane has a MWCO of about 5 kDa.

14. The method according to any one of claims 1 to 13, wherein the feed fluid passes through a filter membrane to reduce NTU before entering the first filtration unit.

15. The method according to any one of claims 1 to 14, wherein each receptacle individually is kept at a temperature in the range of -5 °C to 10 °C, such as 0 °C to 10 °C or at a range of 30°C to 85 °C, such as 35 °C to 55 °C or 35 °C to 45 °C.

Images