Rapid purification of monoclonal antibodies from upstream cell culture materials during the process.

Abstract

This invention generally relates to methods for enriching antibodies of interest. Specifically, this invention relates to enriching therapeutic antibodies from upstream cell culture processes using protein A chromatography in the form of a centrifugal column for product quality property profiling analysis.

Description

[0001] Cross-references to related applications

[0002] This application claims the benefit of U.S. Provisional Patent Application No. 63 / 544,767, filed October 18, 2023, and U.S. Provisional Patent Application No. 63 / 708,147, filed October 16, 2024, each of which is incorporated herein by reference in its entirety. Background Technology

[0003] The biophysical properties of therapeutic peptides and proteins can affect their safety, efficacy, and shelf life. For example, the presence of different size variants or charge variants can alter protein solubility, binding, and stability. Variations in attributes that are important to product quality can be termed critical product quality attributes.

[0004] Therapeutic peptides or proteins (such as antibodies) can acquire various variants and become heterogeneous due to a variety of post-translational modifications (PTMs), protein degradation, enzymatic modifications, and chemical modifications. These changes in biophysical properties can occur at almost any time during and after peptide and protein production. Because these changes in biophysical properties can affect the safety, efficacy, and shelf life of therapeutic peptides and proteins, it is important to identify different variants of specific therapeutic peptides or proteins as early as possible.

[0005] To characterize the product quality attributes of therapeutic peptides or proteins, sufficient amounts of the peptides or proteins must first be enriched in the sample. For therapeutic antibodies, protein A chromatography is typically used for enrichment or purification. However, conventional protein A chromatography methods require large amounts of material and can be both time-consuming and resource-intensive. Conventional large-scale methods are incompatible with rapid or high-throughput analysis of materials throughout the preparation and development process, especially in upstream cell culture.

[0006] Therefore, it should be understood that there is a need for systems and methods to rapidly and in a high-throughput manner enrich, purify, and / or characterize therapeutic peptides or proteins at any stage of cell culture, preparation, or development. Summary of the Invention

[0007] This disclosure provides a method for enriching proteins of interest from cell culture samples.

[0008] In some exemplary embodiments, the method may include: (a) contacting a cell culture sample comprising the protein of interest with a centrifuge column comprising an affinity resin to produce an immobilized sample, wherein the affinity resin specifically binds to the protein of interest; (b) subjecting the immobilized sample to at least one washing step; and (c) subjecting the immobilized sample from (b) to at least one elution step to produce enriched protein of interest.

[0009] On the one hand, the proteins of interest are selected from the group consisting of: therapeutic proteins, receptors, antigen-binding proteins, antibodies, monoclonal antibodies, multispecific antibodies, bispecific antibodies, antibody-derived proteins, fusion proteins, receptor fusion proteins, trap proteins, fragments thereof, variants thereof, and combinations thereof.

[0010] On the one hand, the protein of interest is a monoclonal antibody.

[0011] On the one hand, the protein of interest is dupilumab.

[0012] On the one hand, the protein of interest is aflibercept.

[0013] On the one hand, the cell culture sample is a mammalian cell culture or an insect cell culture. On the other hand, the cell culture sample is derived from CHO cell culture, CHO-K1 cell culture, BHK cell culture, HEK 293 cell culture, Sf9 insect cell culture, or a variant thereof.

[0014] On the one hand, the cell culture sample is a clarified cell culture sample.

[0015] On the one hand, the cell culture samples were collected from the cell culture on one of the following days: day 1 to day 20, day 3 to day 15, day 3 to day 12, day 3 to day 10, day 5 to day 10, day 5 to day 12, or day 5 to day 13.

[0016] On the one hand, the cell culture sample was collected from the cell culture on a day selected from the group consisting of: day 1, day 2, day 3, day 4, day 5, day 6, day 7, day 8, day 9, day 10, day 11, day 12, day 13, day 14, day 15, day 16, day 17, day 18, day 19, or day 20.

[0017] On one hand, the method further comprises repeating steps (a)-(c) at least once. In one specific aspect, the method is repeated using at least a first cell culture sample and a second cell culture sample collected from the same cell culture. In a more specific aspect, the first cell culture sample is collected on a first day, and the second cell culture sample is collected on a second day. In another specific aspect, the first cell culture sample and the second cell culture sample are collected at a time between approximately 3 hours, approximately 6 hours, approximately 12 hours, approximately 18 hours, approximately 24 hours, approximately 1 day, approximately 2 days, approximately 3 days, approximately 4 days, approximately 5 days, approximately 6 days, approximately 7 days, approximately 8 days, approximately 9 days, or approximately 10 days after the sample collection.

[0018] In one aspect, the method further comprises performing steps (a)-(c) in parallel on at least two cell culture samples. In one specific aspect, the at least two cell culture samples are derived from two different cell cultures.

[0019] On one hand, the contact step includes combining the cell culture sample and a binding buffer. In one specific aspect, the binding buffer comprises Tris-buffered saline, sodium phosphate, HEPES, or Tris. In a more specific aspect, the binding buffer further comprises sodium chloride or calcium chloride. In one specific aspect, the binding buffer comprises sodium phosphate. In a more specific aspect, the binding buffer further comprises sodium chloride. In another specific aspect, the pH of the binding buffer is from about 6 to about 8.

[0020] On one hand, the contact step includes adding a combined volume of binding buffer and the cell culture sample to the column, the combined volume being 250 µL to 1000 µL, 300 µL to 900 µL, 400 µL to 800 µL, 500 µL to 700 µL, 550 µL to 650 µL, 590 µL to 610 µL, 599 µL to 601 µL, or about 600 µL.

[0021] On one hand, the contact step includes adding a certain volume of the cell culture sample to the column, the volume being 50 µL to 100 µL, 60 µL to 90 µL, 70 µL to 80 µL, about 70 µL, about 71 µL, about 72 µL, about 73 µL, about 74 µL, about 75 µL, about 76 µL, about 77 µL, about 78 µL, about 79 µL, or about 80 µL.

[0022] On one hand, the contacting step includes adding a certain amount of protein to the column, the amount being 100.5 µg to 804 µg, 250 µg to 1 g, 350 µg to 900 µg, 450 µg to 804 µg, 500 µg to 700 µg, 550 µg to 650 µg, 575 µg to 625 µg, 590 µg to 610 µg, about 595 µg, about 596 µg, about 597 µg, about 598 µg, about 599 µg, about 600 µg, about 601 µg, about 602 µg, about 603 µg, about 604 µg, or about 605 µg.

[0023] On the one hand, the affinity resin is protein A resin, protein G resin, or a combination thereof.

[0024] On one hand, the at least one washing step comprises adding a washing buffer to the column and centrifuging the column to produce a washed flow solution. In one specific aspect, the washing buffer comprises Tris-buffered saline, sodium phosphate, sodium acetate, HEPES, or Tris. In a more specific aspect, the washing buffer further comprises sodium chloride or calcium chloride. In another specific aspect, the washing buffer comprises sodium phosphate or sodium acetate. In a more specific aspect, the washing buffer further comprises sodium chloride. In yet another specific aspect, the pH of the washing buffer is from about 6 to about 8. In yet another specific aspect, the volume of the washing buffer is about 600 µL. In yet another specific aspect, the centrifugation is performed at about 100 relative centrifugal force (RCF). In still another specific aspect, the centrifugation is performed for about 1 minute.

[0025] On one hand, the number of washing steps is one, two, or three. On the other hand, the number of washing steps is two. In yet another aspect, the washing buffer in the first washing step comprises sodium phosphate and sodium chloride, and the washing buffer in the second washing step comprises sodium acetate.

[0026] In one aspect, the at least one elution step comprises adding an elution buffer to the column and centrifuging the column to produce an eluent. In one specific aspect, the elution buffer comprises acetic acid or glycine. In a more specific aspect, the concentration of the acetic acid is about 0.24% or about 40 mM. In another specific aspect, the concentration of the acetic acid is about 0.125% or about 20 mM. In yet another specific aspect, the concentration of the glycine is about 0.1 M. In yet another specific aspect, the volume of the elution buffer is about 400 µL. In yet another specific aspect, the pH of the elution buffer is 1 to 4, 2 to 4, 2.5 to 3.5, 2.8 to 3.2, about 1, about 1.5, about 2, about 2.5, about 3, or about 3.5. In still another specific aspect, the centrifugation is performed at about 100 RCF. In yet another specific aspect, the centrifugation is performed for about 1 minute.

[0027] On the one hand, the number of elution steps is one, two, or three.

[0028] On one hand, the at least one elution step includes adding a neutralization buffer to the column. In one specific aspect, the neutralization buffer comprises a Tris base. In a more specific aspect, the concentration of the Tris base is 1 M to 2 M, about 1 M, about 1.5 M, or about 2 M. In another specific aspect, the volume of the neutralization buffer is 5 µL to 50 µL, about 5 µL, about 10 µL, about 20 µL, about 30 µL, about 40 µL, or about 50 µL.

[0029] On the one hand, the enriched protein of interest yields a yield of more than 50%, more than 60%, more than 70%, more than 80%, more than 90%, more than 95%, more than 99%, approximately 60%, approximately 70%, approximately 75%, approximately 80%, approximately 85%, approximately 90%, approximately 91%, approximately 92%, approximately 93%, approximately 94%, approximately 95%, approximately 96%, approximately 97%, approximately 98%, approximately 99%, or approximately 100%.

[0030] On the one hand, the amount of protein in the enriched protein of interest is greater than 10 µg, greater than 20 µg, greater than 50 µg, greater than 100 µg, greater than 200 µg, greater than 300 µg, greater than 400 µg, greater than 500 µg, greater than 600 µg, greater than 700 µg, greater than 800 µg, greater than 900 µg, greater than 1000 µg, about 10 µg, about 20 µg, about 50 µg, about 100 µg, about 200 µg, about 300 µg, about 400 µg, about 500 µg, about 600 µg, about 700 µg, about 800 µg, about 900 µg, or about 1000 µg.

[0031] On the one hand, the concentration of the enriched protein of interest is greater than 0.01 µg / µL, greater than 0.05 µg / µL, greater than 0.1 µg / µL, greater than 0.2 µg / µL, greater than 0.5 µg / µL, greater than 1 µg / µL, greater than 2 µg / µL, about 0.05 µg / µL, about 0.1 µg / µL, about 0.2 µg / µL, about 0.5 µg / µL, about 1 µg / µL, about 1.5 µg / µL, about 2 µg / µL, or about 2.5 µg / µL.

[0032] On the one hand, the duration of the method is less than 24 hours, less than 12 hours, less than 6 hours, less than 3 hours, less than 2 hours, less than 1 hour, less than 30 minutes, about 3 hours, about 2 hours, about 1.5 hours, about 1 hour, about 50 minutes, about 45 minutes, about 40 minutes, about 30 minutes, or about 20 minutes.

[0033] On the other hand, the method further includes characterizing at least one product quality attribute of the enriched protein of interest. In yet another aspect, the method further includes subjecting the enriched protein of interest to chromatography, mass spectrometry, spectroscopy, capillary electrophoresis, gel electrophoresis, and / or ligand binding assays.

[0034] On one hand, the method further comprises characterizing at least one size variant of the enriched protein of interest. On the other hand, the method further comprises characterizing at least one high molecular weight species of the enriched protein of interest. In one specific aspect, the characterization comprises subjecting the enriched protein of interest to size exclusion chromatography (SEC) analysis.

[0035] On one hand, the method further includes characterizing at least one fragment of the enriched protein of interest. In one specific aspect, the characterization includes subjecting the enriched protein of interest to capillary electrophoresis and sodium dodecyl sulfate (CE-SDS) analysis.

[0036] On one hand, the method further includes characterizing at least one charge variant of the enriched protein of interest. In one specific aspect, the characterization includes subjecting the enriched protein of interest to imaging capillary isoelectric focusing electrophoresis (iCIEF).

[0037] On one hand, the method further includes at least one glycan characterizing the enriched protein of interest. In a specific aspect, the characterization includes subjecting the enriched protein of interest to hydrophilic interaction chromatography (HILIC) analysis.

[0038] In one specific aspect, the method further includes using the at least one product quality attribute to determine whether the cell culture should continue or be terminated. In another specific aspect, the method further includes using the at least one product quality attribute to determine whether the cell culture should be modified.

[0039] These and other aspects of the invention will be better appreciated and understood when considered in conjunction with the following description and accompanying drawings. While the following description illustrates various embodiments and numerous specific details therein, it is given by way of illustration rather than limitation. Many substitutions, modifications, additions, or rearrangements can be made within the scope of the invention. Attached Figure Description

[0040] Figure 1 The general process for producing recombinant proteins (such as monoclonal antibodies (mAbs)) according to an exemplary aspect is shown.

[0041] Figure 2 The steps of a protein A chromatography method according to an exemplary aspect are shown.

[0042] Figure 3 An aggregation pathway of monoclonal antibodies according to an exemplary aspect is demonstrated.

[0043] Figure 4 An example of immunoglobulin breakage is demonstrated.

[0044] Figure 5 The upstream stage of an antibody preparation process according to an exemplary aspect is shown, and the sample for the method and system of the present invention originates from said upstream stage.

[0045] Figure 6A The workflow of the protein A chromatography method and system of the present invention according to an exemplary aspect is shown.

[0046] Figure 6B The workflow of one step of the protein A chromatography method and system of the present invention according to an exemplary aspect is shown.

[0047] Figure 7A The diagram illustrates a protein obtained based on the amount of incoming protein loaded per protein A column according to an exemplary aspect.

[0048] Figure 7B The protein yield based on the loading of each protein A column is shown according to an exemplary aspect.

[0049] Figure 8 The N-glycan structure of enriched antibodies is shown, analyzed using hydrophilic interaction chromatography (HILIC) according to an exemplary aspect.

[0050] Figure 9A The protein yield of dupilumab based on theoretical column loading is shown according to an exemplary aspect, such as by UV-Vis spectroscopy.

[0051] Figure 9B The protein yield of dupilumab based on theoretical column loading, as measured by titration, is shown according to an exemplary aspect.

[0052] Figure 9C A comparison of protein yields of dupilumab based on theoretical column loading, measured according to an exemplary aspect such as by UV-Vis spectroscopy or titration, is shown.

[0053] Figure 10 The product quality profile of dupilumab purified by the method and system of the present invention, according to an exemplary aspect, is shown compared to conventional protein A methods.

[0054] Figure 11 The results of aggregation tests of dupilumab purified from different small-scale bioreactors using the methods and systems of the present invention, according to an exemplary aspect, are shown compared to conventional small-scale purification methods.

[0055] Figure 12 The charge variant distribution of dupilumab purified from different small-scale bioreactors using the methods and systems of the present invention, according to an exemplary aspect, is shown compared to conventional small-scale purification methods.

[0056] Figure 13 The results of a cleavage test of dupilumab purified from different small-scale bioreactors using the method and system of the present invention, according to an exemplary aspect, are shown compared to conventional small-scale purification methods.

[0057] Figure 14 This paper presents an overview of the product quality of aflibercept purified using the method and system of the present invention, according to an exemplary aspect, compared to conventional small-scale purification methods using manufacturer-provided buffers.

[0058] Figure 15A The average total step yield percentage relative to total column load (µg) is shown according to an exemplary aspect.

[0059] Figure 15B The average step yield percentage of eluent 1 according to an exemplary aspect is shown relative to the total column loading (µg).

[0060] Figure 16 The results of aggregation tests of aflibercept purified by the method and system of the present invention at various total column loading levels are shown, according to an exemplary aspect, compared with conventional small-scale purification methods.

[0061] Figure 17 The charge variant distribution of aflibercept purified by the method and system of the present invention is shown at various total column loading levels, compared to conventional small-scale purification methods, according to an exemplary aspect.

[0062] Figure 18 The results of an aggregation test of aflibercept purified by the method and system of the present invention, compared with conventional small-scale purification methods, are shown according to an exemplary aspect, with cell culture samples collected at different time points.

[0063] Figure 19 The diagram illustrates the charge variant distribution of aflibercept purified by the method and system of the present invention, compared to conventional small-scale purification methods, with respect to an exemplary aspect, and cell culture samples collected at different time points. Detailed Implementation

[0064] Analyzing key product quality attributes of biotherapeutic agents is crucial for ensuring the safety, identity, potency, purity, and quality of pharmaceutical products delivered to patients. For the production of therapeutic antibodies, exemplary product quality attributes include size variations (including aggregation and fragmentation), charge variations, and glycosylation variations. The assessment of these product quality attributes requires adequately enriched samples, typically prepared using protein A chromatography. However, conventional methods utilize large-scale protein A chromatography, requiring the preparation of large quantities of cell culture material, processing only one sample at a time, and are time- and resource-intensive. Large-scale protein A methods may, for example, require a week of processing time and approximately 10,000 L of cell culture material, while laboratory-scale protein A methods may, for example, require several days of processing time and approximately 250 L of cell culture material.

[0065] Characterizing the protein quality properties of biotherapeutic agents early in the production process during upstream cell culture is advantageous. Further advantageous is the ability to rapidly (in minutes or hours rather than days) enrich therapeutic antibodies from small amounts of material in parallel with multiple samples. To meet these needs, novel systems and methods for the rapid, small-scale enrichment and purification of biotherapeutic agents from upstream cell culture samples have been developed, which will be described in detail below.

[0066] One advantage of the disclosed invention is its ability to enrich biological therapeutic agents (e.g., dupilumab) on a small scale, such as tens or hundreds of µL of cell culture medium and hundreds of µg of protein from any given day of cell culture. Another advantage is the ability to enrich biological therapeutic agents on short timescales, such as in about 30 minutes or hours, rather than over days in laboratory-scale or large-scale systems. Many samples can be enriched simultaneously because a single centrifuge can process dozens of centrifuge tubes at a time, thereby allowing for parallel processing of replicas or comparisons.

[0067] The rapid, small-scale capability of the disclosed invention allows for the enrichment of biotherapeutic agents throughout the upstream cell culture process, thereby obtaining the time course of product quality attributes throughout the process. Early and / or periodic monitoring of product quality attributes can inform cell culture parameters and whether a product batch should continue or be discontinued, saving significant time and resources compared to waiting for results from large-scale enrichment and downstream quality analysis. Periodic monitoring can also be useful, for example, to experiment with different cell culture conditions and feeding strategies to determine how product quality attributes are affected at different times during cell culture.

[0068] Glossary

[0069] Unless otherwise stated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Although methods and materials similar to or equivalent to any of the methods and materials described herein may be used in practice or testing, specific methods and materials are described hereafter.

[0070] The term “a / a kind” should be understood to mean “at least one / a kind”; and the terms “about” and “approximately” should be understood to allow for variations in standards, as would be understood by one of ordinary skill in the art; and where the scope is provided, endpoints are included. As used herein, the terms “include,” “includes,” and “including” are intended to be non-limiting and are understood to mean “comprise,” “comprises,” and “comprising,” respectively.

[0071] As used herein, the term "protein" or "protein of interest" can include any amino acid polymer having covalently linked amide bonds. A protein comprises one or more amino acid polymer chains, commonly referred to in the art as a "peptide." A "peptide" is a polymer consisting of amino acid residues, associated naturally occurring structural variants, and their synthetic, non-natural analogs linked by peptide bonds. As used herein, the term peptide includes proteins, their variants, fragments, and peptides, whether synthetic, naturally occurring, or derived from larger peptides, for example, through digestion or truncation. "Synthetic peptide or polypeptide" refers to a non-naturally occurring peptide or polypeptide. Synthetic peptides or polypeptides can be synthesized, for example, using an automated peptide synthesizer. Various solid-phase peptide synthesis methods are known to those skilled in the art. Proteins can contain one or more polypeptides to form a single functional biomolecule. In another exemplary aspect, proteins can include antibody fragments, nanobodies, recombinant antibody chimeras, cytokines, chemokines, peptide hormones, etc.

[0072] The protein or peptide of interest may include any of the following: biotherapeutic proteins, recombinant proteins for research or therapy, trap proteins and other chimeric receptor Fc fusion proteins, chimeric proteins, antibodies, monoclonal antibodies, polyclonal antibodies, human antibodies, and bispecific antibodies. The protein may be produced using recombinant cell-based production systems such as insect baculovirus systems, yeast systems (e.g., Pichia sp.), and mammalian systems (e.g., CHO cells and CHO derivatives, such as CHO-K1 cells). For a recent review discussing biotherapeutic proteins and their production, see Ghaderi et al., “Production platforms for biotherapeutic glycoproteins. Occurrence, impact, and challenges of non-human sialylation” (Darius Ghaderi et al., Production platforms for biotherapeutic glycoproteins. Occurrence, impact, and challenges of non-human sialylation), 28 Biotechnology and Genetic Engineering Reviews 147-176 (2012), the full teachings of which are incorporated herein by reference.

[0073] In some exemplary aspects, proteins include modifications, adducts, and other covalently linked moieties. These modifications, adducts, and moieties include, for example, avidin, streptavidin, biotin, glycans (e.g., N-acetylgalactosamine, galactose, neuraminic acid, N-acetylglucosamine, fucose, mannose, and other monosaccharides), PEG, polyhistidine, FLAGtag, maltose-binding protein (MBP), chitin-binding protein (CBP), glutathione S-transferase (GST) myc epitopes, fluorescent labels, and other dyes. Proteins can be classified based on composition and solubility, and therefore can include simple proteins, such as globular and fibrous proteins; conjugated proteins, such as nucleoproteins, glycoproteins, mucins, chromoproteins, phosphoproteins, metalloproteins, and lipoproteins; and derived proteins, such as primary and secondary derived proteins.

[0074] As used herein, the term "recombinant protein" refers to a protein resulting from the transcription and translation of a gene carried on a recombinant expression vector that has been introduced into a suitable host cell. In some respects, a recombinant protein may be an antibody, such as a chimeric antibody, a humanized antibody, or a fully human antibody. In some respects, a recombinant protein may be an isotype antibody selected from the group consisting of IgG, IgM, IgA1, IgA2, IgD, or IgE. In some respects, the antibody molecule is a full-length antibody (e.g., IgG1), or alternatively, the antibody may be a fragment (e.g., an Fc fragment or a Fab fragment).

[0075] As used herein, the term "antibody" includes an immunoglobulin molecule comprising four polypeptide chains, two heavy (H) chains and two light (L) chains linked by disulfide bonds, and their polymers (e.g., IgM). Each heavy chain comprises a heavy chain variable region (abbreviated herein as HCVR or VH) and a heavy chain constant region. The heavy chain constant region contains three domains CH1, CH2, and CH3. Each light chain comprises a light chain variable region (abbreviated herein as LCVR or VL) and a light chain constant region. The light chain constant region contains one domain (CL1). The VH and VL regions can be further subdivided into hypervariable regions called complementarity-determining regions (CDRs), which are interspersed with more conserved regions called framework regions (FRs). Each VH and VL consists of three CDRs and four FRs arranged in the following order from the amino terminus to the carboxyl terminus: FR1, CDR1, FR2, CDR2, FR3, CDR3, and FR4. Common amino acid sequences can be defined based on the side-by-side analysis of two or more CDRs.

[0076] As used herein, the term "antibody" also includes the antigen-binding fragment of a complete antibody molecule. As used herein, the terms "antigen-binding portion" of an antibody, "antigen-binding fragment" of an antibody, etc., include any naturally occurring, enzymatically obtainable, synthetic, or genetically engineered polypeptide or glycoprotein that specifically binds to an antigen to form a complex. The antigen-binding fragment of an antibody can be derived from a complete antibody molecule, for example, using any suitable standard technique, such as proteolytic digestion or recombinant genetic engineering techniques involving the manipulation and expression of DNA encoding variable and optionally constant domains of the antibody. Such DNA is known and / or readily available from, for example, commercial sources, DNA libraries (including, for example, phage-antibody libraries), or can be synthesized. DNA can be sequenced and manipulated by chemical methods or by using molecular biology techniques, for example, to arrange one or more variable and / or constant domains into suitable conformations, or to introduce codons, generate cysteine ​​residues, modify, add or delete amino acids, etc.

[0077] As used herein, “antibody fragment” includes a portion of a complete antibody, such as the antigen-binding or variable region of an antibody. Examples of antibody fragments include, but are not limited to, Fab fragments, Fab' fragments, F(ab')2 fragments, Fc fragments, Fc / 2 fragments, scFv fragments, Fv fragments, dsFv biantibodies, dAb fragments, Fd' fragments, Fd fragments, and separated complementarity-determining regions (CDRs), as well as triantibodies, tetraantibodies, linear antibodies, single-chain antibody molecules, and multispecific antibodies formed from antibody fragments. An Fv fragment is a combination of variable regions of the immunoglobulin heavy and light chains, and a scFv protein is a recombinant single-chain polypeptide molecule in which the variable regions of the immunoglobulin light and heavy chains are linked by peptide linkers. In some aspects, an antibody fragment contains a sufficient amino acid sequence of a parent antibody that is a fragment of the parent antibody such that it binds to the same antigen as the parent antibody; in other aspects, the fragment binds to an antigen having an affinity comparable to that of the parent antibody and / or competes with the parent antibody for antigen binding.

[0078] Antibody fragments can be produced in any manner. For example, antibody fragments can be produced enzymatically or chemically by breaking down an intact antibody, and / or antibody fragments can be produced recombinantly from a gene encoding a portion of the antibody sequence. Alternatively or additionally, antibody fragments can be produced entirely or partially synthetically. Antibody fragments may optionally comprise single-chain antibody fragments. Alternatively or additionally, antibody fragments may comprise multiple chains, for example, linked together by disulfide bonds. Antibody fragments may optionally comprise multi-molecular complexes. Functional antibody fragments typically contain at least about 50 amino acids, and more typically contain at least about 200 amino acids.

[0079] The term "bispecific antibody" refers to antibodies capable of selectively binding to two or more epitopes. Bispecific antibodies typically comprise two distinct heavy chains, each specifically binding to a different epitope on two different molecules (e.g., antigens) or on the same molecule (e.g., the same antigen). If a bispecific antibody is capable of selectively binding to two different epitopes (a first epitope and a second epitope), the affinity of the first heavy chain for the first epitope will typically be at least one to two, three, or four orders of magnitude lower than the affinity of the first heavy chain for the second epitope, and vice versa. The epitopes recognized by the bispecific antibody can be located on the same or different targets (e.g., on the same or different proteins). Bispecific antibodies can be prepared, for example, by combining heavy chains that recognize different epitopes of the same antigen. For example, a nucleic acid sequence encoding a variable sequence of a heavy chain that recognizes different epitopes of the same antigen can be fused with a nucleic acid sequence encoding a constant region of a different heavy chain, and such sequences can be expressed in cells expressing immunoglobulin light chains.

[0080] A typical bispecific antibody has two heavy chains, each with three heavy chain CDRs, and an immunoglobulin light chain. Following the CDRs are a CH1 domain, a hinge, a CH2 domain, and a CH3 domain. The immunoglobulin light chain does not confer antigen-binding specificity but can associate with each heavy chain, or can associate with each heavy chain and bind to one or more epitopes that bind via the antigen-binding region of the heavy chain, or can associate with each heavy chain such that one or two of the heavy chains can bind to one or two epitopes. bsAbs can be divided into two main categories: those with an Fc region (IgG-like) and those lacking an Fc region, the latter typically smaller than Fc-containing IgG and IgG-like bispecific molecules. IgG-like bsAbs can have different formats, such as, but not limited to, trifunctional antibodies, kih IgG, crossMab, orthogonal Fab IgG, dual variable domain Ig (DVD-Ig), dual-acting or dual-functional Fab (DAF), IgG-single-chain Fv (IgG-scFv), or κλ bodies. Non-IgG-like formats include tandem scFv, biantibody formats, single-chain biantibodies, tandem biantibodies (TandAb), dual-affinity retargeting molecules (DART), DART-Fc, nanobodies, or antibodies generated via dock-and-lock (DNL) methods (Gaowei Fan, Zujian Wang, and Mingju Hao, Bispecific antibodies and their applications, 8 Journal of Hematology & Oncology 130; Dafne Müller and Roland E. Kontermann, Bispecific Antibodies, Handbook of Therapeutic Antibodies 265–310 (2014), the entire teachings of which are incorporated herein by reference). Methods for generating bsAbs are not limited to tetrahybridoma techniques based on somatic cell fusion of two different hybridoma cell lines, chemical conjugation involving chemical cross-linking agents, and gene methods utilizing recombinant DNA technology.

[0081] As used herein, a “multispecific antibody” refers to an antibody that has binding specificity to at least two different antigens. While such molecules will typically bind to only two antigens (i.e., bispecific antibodies, bsAb), antibodies with additional specificity (such as trispecific antibodies and KIH trispecific antibodies) can also be addressed using the systems and methods disclosed herein.

[0082] As used herein, the term "monoclonal antibody" is not limited to antibodies produced by hybridoma technology. Monoclonal antibodies can be derived from a single clone in any manner available or known in the art, including any eukaryotic clone, prokaryotic clone, or phage clone. Monoclonal antibodies usable in this disclosure can be prepared using a variety of techniques known in the art, including hybridoma technology, recombinant technology, and phage display technology, or combinations thereof.

[0083] In some respects, the proteins or peptides of interest are antibodies, human antibodies, humanized antibodies, chimeric antibodies, monoclonal antibodies, multispecific antibodies, bispecific antibodies, antibody fragments, antigen-binding antibody fragments, single-chain antibodies, biantibodies, tri- or tetra-antibodies, Fab fragments or F(ab')2 fragments, IgD antibodies, IgE antibodies, IgM antibodies, IgG antibodies, IgG1 antibodies, IgG2 antibodies, IgG3 antibodies, IgG4 antibodies, fusion proteins, receptor fusion proteins, antibody-derived proteins, or combinations thereof. On one hand, the antibody is an IgG1 antibody. On another hand, the antibody is an IgG2 antibody. On another hand, the antibody is an IgG4 antibody. On another hand, the antibody is a chimeric IgG2 / IgG4 antibody. On another hand, the antibody is a chimeric IgG2 / IgG1 antibody. On another hand, the antibody is a chimeric IgG2 / IgG1 / IgG4 antibody.

[0084] In some respects, the antibody is selected from the group consisting of: anti-programmed cell death 1 antibody (e.g., anti-PD1 antibody, as described in U.S. Patent Application Publication No. US2015 / 0203579A1), anti-programmed cell death ligand-1 antibody (e.g., anti-PD-L1 antibody, as described in U.S. Patent Application Publication No. US2015 / 0203580A1), anti-DII4 antibody, anti-angiogenic-2 antibody (e.g., anti-ANG2 antibody, as described in U.S. Patent No. 9,402,898), and anti-angiogenic-like 3 antibody (e.g., anti-AngPtl3 antibody, as described in U.S. Patent No. 9,000A1). Antiplatelet-derived growth factor receptor antibodies (e.g., anti-PDGFR antibodies, as described in U.S. Patent No. 9,265,827), anti-Erb3 antibodies, anti-prolactin receptor antibodies (e.g., anti-PRLR antibodies, as described in U.S. Patent No. 9,302,015), anti-complement 5 antibodies (e.g., anti-C5 antibodies, as described in U.S. Patent Application Publication No. US2015 / 0313194A1), anti-TNF antibodies, anti-epidermal growth factor receptor antibodies (e.g., anti-EGFR antibodies, as described in U.S. Patent No. 9,132,192, or anti-EGF antibodies), anti-platelet-derived growth factor receptor antibodies (e.g., anti-EGFR antibodies, as described in U.S. Patent No. 9,132,192, or anti-EGF antibodies), anti-platelet-derived growth factor receptor antibodies (e.g., anti-PDGFR antibodies, as described in U.S. Patent No. 9,265,827), anti-Erb3 antibodies, anti-prolactin receptor antibodies (e.g., anti-PRLR antibodies, as described in U.S. Patent No. 9,302,015), anti-complement 5 antibodies (e.g., anti-C5 antibodies, as described in U.S. Patent Application Publication No. US2015 / 0313194A1), anti-TNF antibodies, anti-epidermal growth factor receptor antibodies (e.g., anti-EGFR antibodies, as described in U.S. Patent No. 9,132,192, or anti-EGF antibodies), anti-Erb3 antibodies, anti-prolactin ... RvIII antibodies, as described in U.S. Patent Application Publication No. US2015 / 0259423A1, anti-proprotein convertase subtilisin Kexin-9 antibodies (e.g., anti-PCSK9 antibodies, as described in U.S. Patent No. 8,062,640 or U.S. Patent Application Publication No. US2014 / 0044730A1), anti-growth and differentiation factor-8 antibodies (e.g., anti-GDF8 antibodies, also known as anti-myosostatin antibodies, as described in U.S. Patent Nos. 8,871,209 or 9,260,515, also known as anti-myosostatin antibodies), anti-glucagon antibodies, etc. Anti-leukocyte antibodies (e.g., anti-GCGR antibodies, as described in U.S. Patent Application Publication No. US2015 / 0337045A1 or US2016 / 0075778A1), anti-VEGF antibodies, anti-IL1R antibodies, interleukin-4 receptor antibodies (e.g., anti-IL4R antibodies, as described in U.S. Patent Application Publication No. US2014 / 0271681A1 or U.S. Patent Nos. 8,735,095 or 8,945,559), and anti-interleukin-6 receptor antibodies (e.g., anti-IL6R antibodies, as described in U.S. Patent Nos. 7,582,298, 8,043,617 or 9,173).Anti-IL1 antibodies, anti-IL2 antibodies, anti-IL3 antibodies, anti-IL4 antibodies, anti-IL5 antibodies, anti-IL6 antibodies, anti-IL7 antibodies, anti-interleukin 33 (e.g., anti-IL33 antibodies, as described in U.S. Patent Application Publication No. US2014 / 0271658A1 or No. US2014 / 0271642A1), and anti-differentiation cluster 3 antibodies (e.g., anti-CD3 antibodies, as described in U.S. Patent Application Publication No. US2014 / 0088295A1 and No. US20150). Anti-differentiation cluster 20 antibodies (e.g., anti-CD20 antibodies, as described in U.S. Patent Application Publication No. US2014 / 0088295A1 and US20150266966A1, and U.S. Patent No. 7,879,984), anti-CD19 antibodies, anti-CD28 antibodies, anti-differentiation cluster 48 antibodies (e.g., anti-CD48 antibodies, as described in U.S. Patent No. 9,228,014), anti-Fel d1 antibodies (e.g., as described in U.S. Patent No. 9,079,948), anti-influenza virus antibodies, anti-respiratory syncytial virus antibodies (e.g., anti-RSV antibodies, as described in U.S. Patent Application Publication No. US2014 / 0271653A1), anti-Middle East Respiratory Syndrome virus antibodies (e.g., anti-MERS-CoV antibodies, as described in U.S. Patent Application Publication No. US2015 / 0337029A1), and anti-Ebola virus antibodies (e.g., as described in U.S. Patent Application Publication No. US2016 / 021). Anti-Zika virus antibodies, anti-Severe Acute Respiratory Syndrome (SARS) antibodies (e.g., anti-SARS-CoV antibodies), anti-COVID-19 antibodies (e.g., anti-SARS-CoV-2 antibodies), anti-lymphocyte activation gene 3 antibodies (e.g., anti-LAG3 antibodies or anti-CD223 antibodies), and anti-nerve growth factor antibodies (e.g., anti-NGF antibodies, as described in U.S. Patent Application Publication No. US2016 / 0017029 and U.S. Patent Nos. 8,309,088 and 9,353).(as described in patent application No. 176) and anti-activin A antibodies. In some aspects, the bispecific antibody is selected from the group consisting of: anti-CD3 × anti-CD20 bispecific antibodies (as described in U.S. Patent Application Publications Nos. US2014 / 0088295A1 and US20150266966A1), anti-CD3 × anti-mucin 16 bispecific antibodies (e.g., anti-CD3 × anti-Muc16 bispecific antibodies), anti-CD3 × BCMA bispecific antibodies, and anti-CD3 × anti-prostate-specific membrane antigen bispecific antibodies (e.g., anti-CD3 × anti-PSMA bispecific antibodies). In one aspect, the protein or polypeptide of interest comprises a combination of any of the foregoing items.

[0085] In some respects, the protein or polypeptide of interest is selected from the group consisting of alirocumab, sarilumab, fasinumab, nesvacumab, dupilumab, trevogrumab, evinacumab, and rinucumab, as well as their modifications, truncations, and variants. In other respects, the protein of interest comprises any combination of the foregoing.

[0086] Dupilumab (DUPIXENT®) is a monoclonal antibody developed in collaboration between Regeneron and Sanofi. It was approved by the U.S. Food and Drug Administration in March 2017 as the first antibody-based treatment for adult atopic dermatitis (Thibodeaux et al., 2019, *Human Vaccines & Immunother.*, 15:2129-2139; Rodrigues et al., 2019, *Italian Journal of Dermatology and Venereology*, 154). Atopic dermatitis (AD) is a chronic inflammatory skin condition affecting up to 20% of the global population, characterized by dryness, erythema, and lichenified papules and plaques. Dupilumab can be used in conjunction with topical corticosteroids or as the sole treatment (D'Ippolito and Pisano, 2018, Pharmacy and Therapeutics, 43(9):532).

[0087] In October 2018, DUPIXENT® was approved as an adjunct maintenance therapy for the treatment of moderate to severe asthma with an eosinophilic phenotype in patients aged 12 years and older, designed to suppress atopic symptoms and improve patient quality of life by reducing symptoms and morbidity (Thibodeaux et al.). DUPIXENT® is also approved for the treatment of AD in children 6 months and older, asthma with an eosinophilic phenotype or oral corticosteroid-dependent asthma in children 6 years and older, eosinophilic esophagitis (EoE) in patients 12 years and older, nodular prurigo (PN) in adults, and as adjunct therapy for chronic sinusitis with nasal polyps (CRSwNP) in adults (Patient Information - DUPIXENT® (dupilumab) subcutaneous injection, Regeneron Corporation, regeneron.com / downloads / dupixent_ppi.pdf, (accessed May 25, 2023); for DUPIXENT®... IXENT® (dupilumab) takes action, dupixent.com (accessed July 24, 2023). Additionally, dupilumab is currently undergoing numerous clinical trials, including new indications such as: Chronic obstructive pulmonary disease (COPD) (a pivotal study evaluating the efficacy, safety, and tolerability of dupilumab in patients with moderate to severe COPD and type 2 inflammation - ClinicalTrials.gov, classic.clinicaltrials.gov / ct2 / show / NCT04456673, July 2023). Accessed on July 24, 2023), bullous pemphigoid (a study evaluating the efficacy and safety of dupilumab in adult patients with bullous pemphigoid - ClinicalTrials.gov, classic.clinicaltrials.gov / ct2 / show / NCT04206553, (accessed on July 24, 2023)), chronic sinusitis without nasal polyps (CRSsNP) (dupilumab in CRSsNP - ClinicalTrials.gov, classic.clinicaltrials.gov) / ct2 / show / NCT04678856 (accessed July 24, 2023)), Keloids (Study evaluating the efficacy and safety of dupilumab for the treatment of keloids - ClinicalTrials.gov, classic.clinicaltrials.gov / ct2 / show / NCT05128383 (accessed July 24, 2023)), Alopecia Areata (Regeneron AA multicenter (dupilumab) - ClinicalTrials.gov, classic.clinicaltrials.gov.gov / ct2 / show / NCT05551793, (accessed July 24, 2023)), chronic hepatitis pruritus (dupilumab for the treatment of moderate to severe chronic hepatitis pruritus - ClinicalTrials.gov, classic.clinicaltrials.gov / ct2 / show / NCT04256759, (accessed July 24, 2023)), eosinophilic gastritis (dupilumab for eosinophilic gastritis - ClinicalTrials.gov, classic.clinicaltrials.gov / ct2 / show / NCT03678545, (accessed July 24, 2023)), milk allergy (dupilumab and milk OIT for the treatment of milk allergy - ClinicalTrials.gov, classic.clinicaltrials.gov (accessed July 24, 2023) and Peanut Allergy (Studies in Peanut Allergy-Evaluating the Efficacy and Safety of Dupilumab as an Adjuvant to AR101 (Peanut Oral Immunotherapy) - ClinicalTrials.gov, classic.clinicaltrials.gov / ct2 / show / NCT03682770, (accessed July 24, 2023)), and as Neoadjuvant Therapy in Men with High-Risk Prostate Cancer (The Role of Neoadjuvant Dupilumab in Men with Locally High-Risk Prostate Cancer - ClinicalTrials.gov, classic.clinicaltrials.gov / ct2 / show / NCT03886493, (accessed July 24, 2023)).

[0088] Dupilumab is a fully human IgG4 monoclonal antibody with a molecular weight of approximately 147 kDa, produced using Chinese hamster ovary (CHO) cell suspension cultures (D'Ippolito and Pisano; Patient Information). The antibody binds to the IL-4Rα subunit of both type 1 and type 2 IL-4 receptors, inhibiting the IL-4 and IL-13 signaling pathways. This reduces the release of cytokines and chemokines as inflammatory mediators, as well as the release of nitric oxide and IgE, and results in increased serum levels of IL-4 and IL-13. IL-4 and IL-13 play crucial roles in type 2 inflammation, a component of atopic diseases such as asthma, atopic dermatitis, or chronic sinusitis with nasal polyps. IL-4 induces the differentiation of primary CD4+ T cells into Th2 effector cells, and IL-13 is involved in goblet cell metaplasia, smooth muscle alterations, fibrosis, excessive mucus secretion, and increased airway hyperresponsiveness. In addition, both IL-4 and IL-13 promote eosinophil chemotaxis toward sites of inflammation and class switching of B cell immunoglobulins to IgE and IgG4 (in humans) or IgG1 (in mice) (Thibodeaux et al.; Le Floc'h et al., 2020, Allergy, 75:1188-1204).

[0089] Compared to placebo, dupilumab has been shown to reduce circulating concentrations of FeNO (exhaled nitric oxide fraction) and total IgE, allergen-specific IgE, eosinophil chemokine-3, periosteal protein, and chemokine CCL17 in asthmatic patients (patient information). In the case of atopic dermatitis, compared to placebo, dupilumab has been shown to reduce the expression of genes involved in epidermal proliferation (MKi67 and K16) and Th2 inflammatory responses (IL-4, IL-14, CCL17, CCL18, CCL26), thereby reducing the thickness of the affected skin (Thibodeaux et al.). When used to treat chronic sinusitis with nasal polyps, dupilumab was found to reduce polyp size and the concentrations of type 2 inflammatory biomarkers (eosinophil chemokine-3 and total IgE) in blood, nasal secretions and polyp tissue, compared to placebo, and to rapidly improve the sense of smell in patients with CRSwNP (Jonstam et al., 2019, Allergy, 74:743-752).

[0090] In some aspects, the protein or polypeptide of interest is a recombinant protein containing an Fc moiety and another domain (e.g., an Fc fusion protein). In some aspects, the Fc fusion protein is a receptor Fc fusion protein containing one or more extracellular domains of a receptor coupled to the Fc moiety. In some aspects, the Fc moiety comprises a hinge region, followed by the CH2 and CH3 domains of IgG. In some aspects, the receptor Fc fusion protein contains two or more distinct receptor chains that bind to a single ligand or multiple ligands. For example, the Fc fusion protein is a TRAP protein, such as an IL-1 trap (e.g., rilonacept, which contains an IL-1RAcP ligand-binding region fused to the extracellular region of II-1R1 fused to the Fc of hIgG1; see U.S. Patent No. 6,927,004, which is incorporated herein by reference in its entirety), or a VEGF trap (e.g., aflibercept or ziv-aflibercept), which contains an Ig domain 2 of VEGF receptor Flt1 fused to Ig domain 3 of VEGF receptor Flk1 fused to the Fc of hIgG1; see U.S. Patent Nos. 7,087,411 and 7,279,159). In other aspects, the Fc fusion protein is a ScFv-Fc fusion protein containing one or more antigen-binding domains (such as variable heavy chain fragments and variable light chain fragments) of an antibody conjugated to the Fc portion.

[0091] The following identifies and describes proteins prepared in cell culture that can be produced, used, or characterized according to the present invention. Cells containing the essential DNA encoding these proteins can be cultured to produce them according to the present invention.

[0092] For example, regarding antibody production, this invention is applicable to research and production uses for diagnosis and treatment based on all major antibody classes (i.e., IgG, IgA, IgM, IgD, and IgE). IgG is a preferred class and includes subclasses IgG1 (including IgG1λ and IgG1κ), IgG2, IgG3, and IgG4. Other antibody embodiments include human antibodies, humanized antibodies, chimeric antibodies, monoclonal antibodies, multispecific antibodies, bispecific antibodies, antigen-binding antibody fragments, single-chain antibodies, biantibodies, triantibodies, or tetraantibodies, Fab fragments or F(ab')2 fragments, IgD antibodies, IgE antibodies, IgM antibodies, IgG antibodies, IgG1 antibodies, IgG2 antibodies, IgG3 antibodies, or IgG4 antibodies. In one embodiment, the antibody is an IgG1 antibody. In one embodiment, the antibody is an IgG2 antibody. In one embodiment, the antibody is an IgG4 antibody. In one embodiment, the antibody is a chimeric IgG2 / IgG4 antibody. In one embodiment, the antibody is a chimeric IgG2 / IgG1 antibody. In one embodiment, the antibody is a chimeric IgG2 / IgG1 / IgG4 antibody. The aforementioned derivatives, components, domains, chains, and fragments are also included.

[0093] Other antibody embodiments include human antibodies, humanized antibodies, chimeric antibodies, monoclonal antibodies, multispecific antibodies, bispecific antibodies, trispecific antibodies, antigen-binding antibody fragments, single-chain antibodies, bispecific antibodies, trispecific or tetraspecific antibodies, Fab fragments or F(ab')2 fragments, IgD antibodies, IgE antibodies, IgM antibodies, IgG antibodies, IgG1 antibodies, IgG2 antibodies, IgG3 antibodies, or IgG4 antibodies. In one embodiment, the antibody is an IgG1 antibody. In one embodiment, the antibody is an IgG2 antibody. In one embodiment, the antibody is an IgG4 antibody. In another embodiment, the antibody is a chimeric IgG2 / IgG4 antibody. In another embodiment, the antibody is a chimeric IgG2 / IgG1 antibody. In another embodiment, the antibody is a chimeric IgG2 / IgG1 / IgG4 antibody.

[0094] In another embodiment, the antibody is selected from the group consisting of: anti-programmed cell death 1 antibody (e.g., anti-PD1 antibody, as described in U.S. Patent Application Publication No. US2015 / 0203579A1), anti-programmed cell death ligand-1 (e.g., anti-PD-L1 antibody, as described in U.S. Patent Application Publication No. US2015 / 0203580A1), anti-Dll4 antibody, anti-angiogenic-2 antibody (e.g., anti-ANG2 antibody, as described in U.S. Patent No. 9,402,898), and anti-angiogenic-like 3 antibody (e.g., anti-AngPtl3 antibody, as described in U.S. Patent No. 9,402,898). Antiplatelet-derived growth factor receptor antibodies (e.g., anti-PDGFR antibodies, as described in U.S. Patent No. 9,265,827), anti-Erb3 antibodies, anti-prolactin receptor antibodies (e.g., anti-PRLR antibodies, as described in U.S. Patent No. 9,302,015), anti-complement 5 antibodies (e.g., 25 anti-C5 antibodies, as described in U.S. Patent Application Publication No. US2015 / 0313194A1), anti-TNF antibodies, and anti-epidermal growth factor receptor antibodies (e.g., anti-EGFR antibodies, as described in U.S. Patent No. 9,132,192). Anti-EGFRvIII antibodies, such as those described in U.S. Patent Application Publication No. US2015 / 0259423A1, anti-proprotein convertase subtilisin Kexin-9 antibodies (e.g., anti-PCSK9 antibodies, as described in U.S. Patent No. 8,062,640 or U.S. Patent Application Publication No. US2014 / 0044730A1), anti-growth and differentiation factor-8 antibodies (e.g., anti-GDF8 antibodies, as described in U.S. Patent Nos. 8,871,209 or 9,260,515, also known as anti-myosostatin antibodies), anti-glucagon receptor antibodies, etc. Anti-GCGR antibodies (e.g., anti-GCGR antibodies, as described in U.S. Patent Application Publication No. US2015 / 0337045A1 or US2016 / 0075778A1), anti-VEGF antibodies, anti-IL1R antibodies, interleukin 4 receptor antibodies (e.g., anti-IL4R antibodies, as described in U.S. Patent Application Publication No. US2014 / 0271681A1 or U.S. Patent Nos. 8,735,095 or 8,945,559), and anti-interleukin 6 receptor antibodies (e.g., anti-IL6R antibodies, as described in U.S. Patent Nos. 7,582,298, 8,043,617 or 9,173).Anti-IL1 antibodies, anti-IL2 antibodies, anti-IL3 antibodies, anti-IL4 antibodies, anti-IL5 antibodies, anti-IL6 antibodies, anti-IL7 antibodies, anti-interleukin 33 (e.g., anti-IL33 antibodies, as described in U.S. Patent Application Publication No. US2014 / 0271658A1 or No. US2014 / 0271642A1), anti-respiratory syncytial virus antibodies (e.g., anti-RSV antibodies, as described in U.S. Patent Application Publication No. US2014 / 0271653A1), and anti-differentiation cluster 3 (e.g., anti-CD3 antibodies, as described in U.S. Patent Application Publication No. US2014 / 0271653A1). The publications include US2014 / 0088295A1 and US20150266966A1, and US Application No. 62 / 222,605, anti-differentiation cluster 20 (e.g., anti-CD20 antibody, as described in US Patent Application Publications US2014 / 0088295A1 and US20150266966A1, and US Patent No. 7,879,984), anti-CD19 antibody, anti-CD28 antibody, anti-differentiation cluster 48 (e.g., anti-CD48 antibody, as described in US Patent No. 9,228,014), and anti-Fel. Antibodies against d1 (e.g., as described in U.S. Patent No. 9,079,948), antibodies against Middle East Respiratory Syndrome Virus (e.g., anti-MERS antibodies, as described in U.S. Patent Application Publication No. US2015 / 0337029A1), antibodies against Ebola virus (e.g., as described in U.S. Patent Application Publication No. US2016 / 0215040), antibodies against Zika virus, antibodies against lymphocyte activation gene 3 (e.g., anti-LAG3 antibodies or anti-CD223 antibodies), antibodies against nerve growth factor (e.g., anti-NGF antibodies, as described in U.S. Patent Application Publication No. US2016 / 0017029 and U.S. Patent Nos. 8,309,088 and 9,353), and antibodies against nerve growth factor (e.g., anti-NGF antibodies, as described in U.S. Patent Application Publication No. US2016 / 0017029 and U.S. Patent Nos. 8,309,088 and 9,353).(as described in US Patent Application Publication No. 176) and anti-activin A antibody. In some embodiments, the bispecific antibody is selected from the group consisting of: anti-CD3 x anti-CD20 bispecific antibodies (as described in US Patent Application Publications Nos. US2014 / 0088295A1 and US20150266966A1), anti-CD3 x anti-mucin 16 bispecific antibodies (e.g., anti-CD3 x anti-Muc16 bispecific antibodies), and anti-CD3 x anti-prostate-specific membrane antigen bispecific antibodies (e.g., anti-CD3 x anti-PSMA bispecific antibodies). See also US Patent Publication No. US 2019 / 0285580 A1. This also includes Met x Met antibodies, anti-NPR1 agonist antibodies, LEPR agonist antibodies, BCMA x CD3 antibodies, MUC16 x CD28 antibodies, GITR antibodies, IL-2Rg antibodies, EGFR x CD28 antibodies, factor XI antibodies, antibodies against SARS-CoC-2 variants, Fel d 1 multi-antibody therapy, and Bet v 1 multi-antibody therapy. Derivatives, components, domains, chains, and fragments of the above are also included.

[0095] Cells that produce exemplary antibodies can be cultured, used, or characterized according to the present invention. Exemplary antibodies include alikumab, atoltivimab, maftivimab, odesivimab, odesivivmab-ebgn, casirivimab, imdevimab, cemiplimab, and cemiplimab-rwlc (a human IgG4 monoclonal antibody that binds to PD-1), dupilumab, etc. Euphratib (a human monoclonal antibody against the IgG4 subclass that binds to the IL-4Rα(α) subunit and thereby inhibits interleukin 4 (IL-4) and interleukin 13 (IL-13) signaling), ivexumab, ivexumab-dgnb, fasnurumab, fianlimab, garetosmab, itepekimab, nevasumab, odrononextamab, pozelimab, sarrelurumab, trogolurumab, and linusumab.

[0096] Other exemplary antibodies include ravulizumab-cwvz, abciximab, adalimumab, adalimumab-atto, ado-trastuzumab, alemtuzumab, atezolizumab, avelumab, basiliximab, belimumab, benalizumab, bevacizumab, bezlotoxumab, blinatumomab, brentuximab vedotin, brodalumab, canakinumab, and capromab. Pendetide, certolizumab pegol, cetuximab, denosumab, dinutuximab, durvalumab, eculizumab, elotuzumab, emicizumab-kxwh, emtansinealirocumab, evolocumab, golimumab, guselkumab, ibritumomabtiuxetan, idarucizumab, infliximab, infliximab-abda, infliximab-dyyb, ipilimumab, ixekizumab, mepolizumab, necitumumab, nivolumab, obiltoxaximab, obinutuzumab, ocrelizumab, ofatumumab, olaratumab, omalizumab Panitumumab, pembrolizumab, pertuzumab, ramucirumab, ranibizumab, raxibacumab, reslizumab, rinucumab, rituximab, secukinumab, siltuximab, tocilizumab, trastuzumab, ustekinumab, and vedolizumab.

[0097] In addition to next-generation products, this invention is also applicable to the production and / or characterization of biosimilars. The definition of a biosimilar varies by jurisdiction, but it shares common characteristics with a previously approved biological product (often referred to as a “reference product”) in that jurisdiction. According to the World Health Organization, a biosimilar is a biological therapeutic product that is similar in quality, safety, and efficacy to an approved reference biological therapeutic product, and is welcomed in many countries.

[0098] Recombinant protein production

[0099] As used herein, a “sample” can be obtained from any step of a biological process, such as cell culture medium (CCF), harvested cell culture medium (HCCF), any step in downstream processing, final concentration tank (FCP), drug substance (DS) or drug product (DP) containing the final formulated product. In some respects, samples can be obtained from cell cultures on days 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, and / or 15. In some respects, samples can be obtained from culture vessels (e.g., plates, flasks, bags, or bioreactors). In some respects, the volume of the culture vessel can be, for example, 250 mL, 500 mL, 1 L, 2 L, 5 L, 10 L, 50 L, 500 L, 1,000 L, 2,000 L, 3,000 L, 5,000 L, 10,000 L, 15,000 L, 20,000 L or 25,000 L.

[0100] In some specific aspects, the sample may be selected from any step of a downstream process, such as clarification, chromatographic generation, or filtration. In some specific exemplary aspects, the pharmaceutical product may be selected from pharmaceutical products prepared in clinical, transport, storage, or processing conditions.

[0101] In some exemplary aspects, the protein or peptide of interest may be derived from mammalian cells. Mammalian cells may be of human or non-human origin and may include primary epithelial cells (e.g., keratinocytes, cervical epithelial cells, bronchial epithelial cells, tracheal epithelial cells, renal epithelial cells, and retinal epithelial cells), established cell lines and strains thereof (e.g., HEK293 embryonic kidney cells, BHK cells, HeLa cervical epithelial cells and PER-C6 retinal cells, MDBK (NBL-1) cells, 911 cells, CRFK cells, MDCK cells, CHO cells, BeWo cells, Chang cells, Detroit 562 cells, HeLa 229 cells, HeLa S3 cells, Hep-2 cells, KB cells, LSI80 cells, LS174T cells, NCI-H-548 cells, RPMI2650 cells, SW-13 cells, T24 cells, WI-28 cells, etc.). VA13, 2RA cells, WISH cells, BS-CI cells, LLC-MK2 cells, clone M-3 cells, 1-10 cells, RAG cells, TCMK-1 cells, Yl cells, LLC-PKi cells, PK(15) cells, GHi cells, GH3 cells, L2 cells, LLC-RC 256 cells, MHiCi cells, XC cells, MDOK cells, VSW cells and TH-I, B1 cells, BSC-1 cells, RAf cells, RK cells, PK-15 cells or their derivatives), fibroblasts from any tissue or organ (including but not limited to the heart, liver, kidney, colon, intestine, esophagus, stomach, nervous tissue (brain, spinal cord), lung, vascular tissue (arteries, veins, capillaries), lymphoid tissue (lymph nodes, glands, tonsils, bone marrow and blood), spleen and fibroblasts and fibroblast-like cell lines) (e.g.,CHO cells, TRG-2 cells, IMR-33 cells, Don cells, GHK-21 cells, citrullinemia cells, Dempsey cells, Detroit 551 cells, Detroit 510 cells, Detroit 525 cells, Detroit 529 cells, Detroit 532 cells, Detroit 539 cells, Detroit 548 cells, Detroit 573 cells, HEL 299 cells, IMR-90 cells, MRC-5 cells, WI-38 cells, WI-26 cells, Midi cells, CHO cells, CV-1 cells, COS-1 cells, COS-3 cells, COS-7 cells, Vero cells, DBS-FrhL-2 cells, BALB / 3T3 cells, F9 cells, SV-T2 cells, M-MSV-BALB / 3T3 cells, K-BALB cells, BLO-11 cells, NOR-10 cells, C3H / IOTI / 2 cells, HSDMiC3 cells, KLN205 cells, McCoy cells, mouse L cells, strain 2071 (mouse L) cells, LM strain (mouse L) cells, L-MTK' (mouse L) cells, NCTC clones 2472 and 2555, SCC-PSA1 cells, Swiss / 3T3 cells, Indian muntjac cells (Indian muntjac cells) (Muntjac cells), SIRC cells, Cn cells, Jensen cells, Sp2 / 0, NS0, NS1 cells, or their derivatives).

[0102] The recombinant proteins of interest (e.g., antibodies) can be isolated from cell culture media, serum, plasma, ascites, or bacterial culture media, and require multiple purification steps to isolate any contaminants associated with the product, preparation process, or host cell (Rathore and Bhambure, 2014, Methods in Molecular Biology, 29-37).

[0103] Large-scale antibody drug production typically utilizes cell culture media generated from a master cell bank, multiple seed cells, and a production bioreactor, followed by cell removal steps, antibody purification (usually including affinity chromatography), virus inactivation (using detergents or low pH), purification steps, and virus filtration. The product is then concentrated, percolated, and formulated (Chahar et al., 2020, Biologicals, 63:1-13; Jin et al., 2019, Monoclonal Antibodies, 11:1479-1491). Figure 1 An exemplary workflow for mAb generation is shown in the figure.

[0104] The upstream process typically involves culturing genetically modified mammalian cells (e.g., CHO cells) to produce the antibody of interest. Frozen cells in a cell bank are thawed and cultured. These cells are then gradually mass-produced over approximately two weeks to generate seed cultures for the bioreactor tank.

[0105] In bioreactors (with volumes up to, for example, 20,000 liters (Vázquez-Rey and Lang, 2011, Biotechnol. Bioeng., 108:1494-1508)), optimal conditions are maintained for cell growth and protein production. This includes providing the necessary metabolic substrates and growth factors, as well as controlling temperature, pH, dissolved oxygen, and other gases, while protecting the environment from microbial contamination (Jozala et al., 2016, Braz. J. Microbiol., 47:51-63). These conditions influence the synthesis of the antibody of interest and its potential degradation throughout the upstream preparation process, and are one of the determinants of the quality profile of the resulting drug (Das et al., 2020, J. Pharm. Sci., 109:116-133).

[0106] When the production phase ends, microfiltration or centrifugation is typically used to separate the cells, and the harvested cell culture supernatant is used in downstream processes to recover biopharmaceutical products (Vázquez-Rey and Lang).

[0107] Following the upstream processing stage of recombinant protein production is downstream processing. The purpose of downstream processing is to isolate the protein from other components of the cell culture mixture, and to isolate the antibody of interest from other proteins without loss of their chemical integrity and biological activity. Typically, at least two or three different separation steps are employed in sequence (Rathore and Bhambure, 2014), and these separation steps can be based on solubility, hydrophobicity, density, charge and charge distribution, isoelectric point, ligand binding affinity, reversible association, metal binding, post-translational modification, size, or shape. The initial purification steps typically consist of higher-capacity steps at lower resolution, while subsequent steps are typically lower-capacity steps at higher resolution to illustrate the reduction in protein content throughout the purification process (Labrou, 2014, *Methods in Molecular Biology*, 3-10).

[0108] The most commonly used purification technique for monoclonal antibodies is affinity chromatography (Arora et al., 2014, *Methods in Molecular Biology*, 497-516), while protein A (ProA) affinity chromatography, due to its high reliability, stability, and reproducibility, is often specifically the first step in downstream processes (dos Santos et al., 2017, *Biotechnol. Adv.*, 35:41-50). This step is typically followed by virus inactivation (VI), as both VI and antibody elution from the protein A resin require low pH conditions. Purification steps after virus inactivation typically involve ion exchange chromatography (IEC) and membrane filtration (Vázquez-Rey and Lang). The product can then be formulated and packaged. The resulting antibody product should be free of contaminants such as host cells and their proteins, nucleic acids, viruses, pyrogens, leachates and cell culture medium components, and undesirable protein isotypes (Labrou) that may result from post-translational modifications.

[0109] While proteins produced for some purposes can be used in their original state, the biopharmaceutical industry and its stringent drug product quality regulations demand extremely high purity, thus emphasizing the importance of robust purification procedures to ensure the safety, identity, potency, purity, and quality of biotherapeutic agents delivered to patients, and significantly increasing costs. Downstream processing is estimated to account for 50–80% of the cost of protein preparation, with chromatography accounting for up to 60% of downstream costs (Labrou; Bracewell et al., 2015, BioPharm International, 28(3)).

[0110] Protein A chromatography is often considered a bottleneck because it not only uses particularly expensive resins, but also requires multiple cycles per batch and has capacity and diffusion limitations (dos Santos). Over the past few decades, as upstream generation processes have become increasingly efficient, downstream processes have become a limiting factor due to their inability to purify increasing amounts of material at the same rate. Therefore, maximizing yield and minimizing the number of separation steps is crucial when developing downstream purification procedures (Shi and Sun, 2020, *Chinese Journal of Chemical Engineering*, 30:194-203; Labrou).

[0111] Chromatographic techniques have been improved through enhancements in resin properties (such as optimizing pore structure, size, and volume to enhance dynamic binding capacity) and ligand chemistry (such as immobilization on beads), which has increased ligand accessibility (Rathore et al., 2018, Biotechnol. Lett., 40:895-905). However, the development of chromatographic purification steps faces many challenges, including insufficient understanding of protein-related processes, poor characterization of raw or feed materials, product instability, and low feed concentrations (Rathore and Bhambure).

[0112] Alternative chromatography methods are better suited for large-scale production in upstream processes at an industrial scale and have been explored for protein purification, including liquid-liquid extraction, membrane processing, precipitation and crystallization, and magnetic separation (dos Santos). However, despite limitations, chromatography (especially ProA chromatography) remains the most commonly used technique in downstream processing due to its high resolution for complex mixtures and high selectivity for biotherapeutic molecules (Rathore et al., 2018) (Labrou).

[0113] In some respects, samples may be prepared before or after enrichment, separation, and / or analytical steps. Preparation steps may include alkylation, reduction, denaturation, digestion, derivatization, and / or deglycosylation.

[0114] As used herein, the term "protein alkylating agent" refers to an agent used to alkylate certain free amino acid residues in a protein. Non-limiting examples of protein alkylating agents include iodoacetamide (IOA), chloroacetamide (CAA), acrylamide (AA), N-ethylmaleimide (NEM), methyl methanethiosulfonate (MMTS), and 4-vinylpyridine or combinations thereof.

[0115] As used herein, “protein denaturation” can refer to the process by which the three-dimensional shape of a molecule changes from its native state. Protein denaturation can be performed using protein denaturing agents. Non-limiting examples of protein denaturing agents include heat, high or low pH, reducing agents such as DTT (see below), or exposure to a dissociating agent. Several dissociating agents can be used as protein denaturing agents. Dissociating solutes increase the entropy of a system by interfering with intramolecular interactions mediated by non-covalent forces such as hydrogen bonds, van der Waals forces, and hydrophobic interactions. Non-limiting examples of dissociating agents include butanol, ethanol, guanidine hydrochloride, lithium perchlorate, lithium acetate, magnesium chloride, phenol, propanol, sodium dodecyl sulfate, thiourea, N-lauroyl sarcosine, urea, and their salts.

[0116] As used herein, the term "protein reducing agent" refers to a reagent used to reduce disulfide bonds in proteins. Non-limiting examples of protein reducing agents used for protein reduction include dithiothreitol (DTT), β-mercaptoethanol, Ellman's reagent, hydroxylamine hydrochloride, sodium cyanoborohydride, tris(2-carboxyethyl)phosphonic acid hydrochloride (TCEP-HCl), or combinations thereof. Conventional methods for protein analysis (i.e., reduced peptide mapping) involve protein reduction prior to LC-MS analysis. In contrast, non-reducing peptide mapping omits the sample preparation reduction step to preserve endogenous disulfide bonds. In some respects, non-reducing agents may be used, for example, to preserve endogenous disulfide bonds between the Fab arms of antibodies or antibody-derived proteins. In other respects, partially reducing agents may be used, for example, to reduce disulfide bonds between the Fab arms of antibodies or antibody-derived proteins without completely reducing the protein.

[0117] As used herein, the term "digestion" refers to the hydrolysis of one or more peptide bonds in a protein or polypeptide. Several methods exist for digesting proteins in a sample using appropriate hydrolytic agents, such as enzymatic digestion or non-enzymatic digestion.

[0118] As used herein, the term "digestive enzyme" refers to any of a large number of different agents capable of digesting proteins or polypeptides. Non-limiting examples of hydrolysants capable of enzymatic digestion include proteases from *Aspergillus Saitoi*, elastase, *Bacillus subtilis* protease, protease XIII, pepsin, trypsin, Tryp-N, chymotrypsin, *Aspergillus pepsin I*, LysN protease (Lys-N), LysC endonuclease (Lys-C), asp-N endonuclease (Asp-N), Arg-C endonuclease (Arg-C), Glu-C endonuclease (Glu-C), outer membrane protein T (OmpT), immunoglobulin-degrading enzyme (IdeS) from *Streptococcus pyogenes*, thermophilic proteases, papain, streptomycin, V8 protease, or bioactivated fragments or homologs thereof, or combinations thereof. For a recent review of the available techniques for protein digestion, see Switzar et al., “Protein Digestion: An Overview of the Available Techniques and Recent Developments” (Linda Switzar, Martin Giera and Wilfried MA Niessen, 12 Journal of Proteomics Research 1067–1077 (2013)).

[0119] In some exemplary aspects, IdeS or variants thereof are used to cleave antibodies below the hinge region, yielding Fc and Fab2 fragments. Digestion of the analyte may be advantageous because the size reduction can increase the sensitivity and specificity for characterization and detection of the analyte using LC-MS. When used for this purpose, digestion that isolates the Fc fragment and retains the Fab2 fragment for analysis may be preferred. This is because the Fab2 fragment contains the variable region of interest (such as the antibody's complementarity-determining region (CDR)), while the Fc fragment may be relatively homogeneous between antibodies and therefore provides less relevant information. Alternatively or additionally, digestion that isolates the Fab2 fragment and retains the Fc fragment for analysis may be preferred because the Fc fragment contains the N-glycosylation site of interest.

[0120] IdeS digestion is highly efficient, resulting in high analyte recoveries. The digestion and elution processes can be carried out under natural conditions, allowing for easy coupling with natural LC-MS systems. IdeS or its variants are commercially available and can be used, for example, as a fabricator. ® Or FabRICATOR Z ® Sale.

[0121] As used herein, the term "liquid chromatography" refers to a process in which a biological / chemical mixture carried by a liquid may be separated into components due to the differential distribution of components as it flows through (or into) a stationary liquid or solid phase. Non-limiting examples of liquid chromatography include reversed-phase (RP) liquid chromatography, ion-exchange (IEX) chromatography, size exclusion chromatography (SEC), affinity chromatography, hydrophobic interaction chromatography (HIC), hydrophilic interaction chromatography (HILIC), or mixed-mode chromatography (MMC). In some aspects, a sample may be subjected to any of the aforementioned chromatographic methods or combinations thereof. Analytes separated by chromatography will be characterized by different retention times, which reflect the rate at which the analyte migrates through the chromatographic column. Analytes can be compared using chromatograms that plot retention times on one axis and measured signals on another axis, wherein the measured signals may be generated by, for example, UV detection or fluorescence detection.

[0122] In some exemplary aspects, the methods and systems of the present invention may include subjecting a sample to affinity chromatography. Affinity chromatography is an analytical technique based on the specific and reversible interactions between proteins and their ligands (e.g., hormones and receptors, enzymes and substrates, or antibodies and their target antigens), thereby allowing the selective binding and separation of target proteins from complex mixtures. Affinity chromatography columns typically consist of ligands covalently immobilized on a solid support (such as agarose gel or agarose). As the sample passes through the column, the target protein binds to the affinity ligands, and unbound or weakly bound sample components are removed in a washing step. The target protein can then be eluted by altering factors such as polarity, ionic strength, or pH, or by adding denaturing agents or competitive protein analogs to the column (Arora et al., 2017, Methods, 116:84-94; Labrou; Arora et al., 2014; Urh et al., 2009, Methods in Enzymology, 417-438).

[0123] Affinity chromatography can involve subjecting biological samples to a column containing a suitable protein A resin. When used herein, the term "protein A" encompasses protein A recovered from its natural source, protein A produced synthetically (e.g., via peptide synthesis or recombinant technologies), and protein A retained with a C10-2000 ppm.H 2 / C H A variant of the protein-binding ability of region 3. Protein A can additionally bind to human IgG molecules containing the IgG F(ab')2 fragment from the human VH3 gene family (Roben et al., J Immunol., 1995, 154(12):6437-45). In some respects, by specifically interacting with the Fc portion of the molecule (if it has the aforementioned region), protein A resins can be used for affinity-based generation and separation of various antibody isotypes.

[0124] In protein A affinity chromatography, the ProA ligand is a single polypeptide chain with a molecular weight of approximately 46.8 kDa, atomized from Staphylococcus aureus or Escherichia coli. Protein A contains homologous domains E, D, A, B, and C, which can bind to the Fc region between the CH2 and CH3 (constant heavy chain 2 and 3) domains of the antibody's IgG class. It can also interact with the variable region between the complementarity-determining regions CDR2 and CDR3 in the heavy chain and interact with polyclonal antibodies (Shi and Sun; Arora et al., 2014; Urh et al.; Ramos-de-la-Peña et al., 2019, J. Sep. Sci., 42:1816-1827).

[0125] Several commercial sources of protein A resins are available. Suitable resins include, but are not limited to, MabSelect PrismA, MabSelect SuRe, MabSelect SuRe LX, MabSelect, MabSelect SuRe pcc, MabSelect Xtra, rProtein A Sepharose, ProSep HC, and ProSep Ultra from Cytiva, ProSep Ultra Plus from EMD Millipore, MabCapture from Thermo Fisher Scientific, and Amsphere™ A3 from JSR Life Sciences.

[0126] Another commonly used ligand (G protein) is derived from group C and group G streptococci (Ramos-de-la-Peña et al.) and has a molecular weight of approximately 21.6 kDa. Protein G is generally less preferred than protein A because protein A can bind to albumin, kininogen, and α2-macroglobulin, leading to reduced antibody purification efficiency. Additionally, G proteins have lower binding affinity and are less stable under acidic conditions (required for the elution step) (Arora et al., 2014). Other proteins used for antibody purification include protein B (also from *Streptococcus*) which binds to human IgA and protein L (from *Peptostreptococcus magnus*) which binds to the light chain of the Fab region, making them particularly useful for purifying antibodies lacking the Fc region (Chahar et al.; Urh et al.).

[0127] Preparing samples for antibody purification using protein A affinity chromatography typically involves clarifying cell cultures, which involves removing most insoluble components, such as cells, by centrifugation, filtration, or precipitation with ammonium sulfate, caprylic acid, or polyethylene glycol (PEG) (Arora et al., 2017; Arora et al., 2014). The target antibody selectively binds to the protein A ligand, while other sample components (including host cell proteins, nucleic acids, culture medium components, product isotypes, and fragments) pass through the resin (Ramos-de-la-Peña et al.). Optimal binding usually occurs at near-neutral pH (i.e., 8.2) (similarly, pH 7–7.5 for protein G chromatography and pH 7.5 for protein L chromatography). The resin is washed to remove any remaining unbound or weakly bound antibodies and impurities, such as nucleic acids or host cell proteins. The bound antibody is then eluted by lowering the pH (typically to 2.5–4) and thus weakening its interaction with the protein ligand. Then, in antibody preparation, a virus inactivation step is performed before the elution buffer is neutralized to improve protein stability and minimize or prevent denaturation, aggregation and loss of biological activity (Chahar et al.; Urh et al.; Zhang et al., 2019, Protein Expression and Purification, 158:65-73). Figure 2 An exemplary workflow for protein A chromatography using a centrifugal (rotary) column is shown in the figure.

[0128] Although protein A affinity chromatography is the method of choice for antibody purification, it still has significant limitations, including relatively low binding capacity, high cost of protein A resin (nearly 50% more expensive than resins used in conventional chromatography), and additional impurities due to ligand co-elution in a fragment known as protein A leachate (Bracewell et al.; Kateia et al., 2018, J. Chromatogr. A), 1579:60-72). Furthermore, affinity chromatography is not very effective in removing aggregates, which can bind more strongly than monomeric antibodies but can still be eluted within the pH range used in the elution step and can even form at low elution pH (Yu et al., 2016, J. Chromatogr. A, 1457:66-75; Amritkar et al., 2020, Biotechnol. Adv., 44:107632). Therefore, it is important to implement purification steps (e.g., ion exchange chromatography (IEC) (Ramos-de-la-Peña et al.) to remove aggregates and other impurities in the downstream process (Zhang et al., 2019).

[0129] Furthermore, in some manufacturing plants, smaller protein A columns are used for multiple cycles per batch, instead of using a large column as a single step. This results in longer processing times, lower throughput, and the concentrated reuse of columns leads to resin damage, ligand degradation and leaching, as well as a reduction in usable surface area (Bracewell et al.).

[0130] Alternative protein A ligands, including aptamers, short peptides, avidin, affinity compounds, or ankyrin repeats, have been investigated, but protein A affinity chromatography remains the gold standard for mAb purification due to its high recovery (>95%) and purity (95% to >99%) achieved by its high binding affinity to the target protein (Curling, 2017, Process Scale Purification of Antibodies, 23-54; Ramos-de-la-Peña et al.). Furthermore, affinity chromatography can be used to separate active and inactive forms of proteins, or to concentrate low-concentration samples (Urh et al.). Therefore, protein A affinity chromatography is used for both purifying antibodies during preparation and as an analytical tool (Rodriguez et al., 2020, Journal of Chromatography, 1157:122332).

[0131] In some respects, affinity columns can be centrifugal affinity columns (or "rotational columns"), where bound or unbound components can be removed from the column using centrifugation. In some respects, using centrifugal columns can have the advantage of being able to process small volumes of samples and can process multiple samples simultaneously compared to conventional flow-based columns.

[0132] Before sample loading, the affinity column can be equilibrated with a suitable buffer. The pH for protein A loading can be, for example, about 6 to about 8, about 6 to about 7, about 7 to about 8, or about 6. After loading the column, it can be washed once or several times with a suitable wash buffer. The column can then be eluted with a suitable elution buffer (e.g., glycine-HCl, acetic acid, or citric acid). The eluent can be monitored using techniques well known to those skilled in the art, such as a UV detector. The elution fraction of interest can be collected and then prepared for further processing.

[0133] Protein A wash buffers can be selected based on their ability to disrupt protein-protein interactions (e.g., disrupting the interaction between the protein of interest and an impurity (such as HCP) without disrupting the interaction between the protein of interest and the chromatographic material). Suitable wash buffers for HCP removal may contain, for example, salts (e.g., sodium salts (such as sodium phosphate or sodium chloride), potassium salts (such as potassium sorbate), magnesium salts, salts containing hydrochloride (such as guanidine hydrochloride), or Tris-containing salts), surfactants (e.g., polysorbate 20, polysorbate 80), polar materials (e.g., isopropanol, ethanol), or amino acids (e.g., arginine). Suitable wash buffers may have a pH of about 5 to about 9, about 5, about 5.5, about 6, about 6.5, about 7, about 7.5, about 8, about 8.5, or about 9.

[0134] Product quality attributes

[0135] Identifying critical product quality (cPQ) attributes of biological drugs is crucial for ensuring the efficacy and safety of the product. Product quality attributes can also be referred to as PQA, and critical quality attributes can be referred to as CQA. According to the ICH Q8(R2) Scientific Guidance for Drug Development, a critical quality attribute is "a physical, chemical, biological or microbiological characteristic or feature that should be within appropriate limits, ranges or distributions to ensure the desired quality of the product" (ICH Guidance Q8(R2) on Drug Development, European Medicines Agency, ema.europa.eu / en / documents / scientific-guideline / international-conference-harmonisation-technical-requirements-registration-pharmaceuticals-human-use_en-11.pdf (accessed 16 June 2023)). Understanding cPQ properties and analyzing their variability is essential for defining acceptance criteria and quality objectives product profiles (QTPPs) (Alt et al., 2016, Biologicals, 44:291-305), and understanding how various process parameters affect cPQ allows for the use of quality-by-design (QbD) approaches to product development (Reusch and Tejada, 2015, Glycobiology, 25:1325-1334).

[0136] cPQ attributes affect product purity, stability, strength, and drug release, as well as other aspects specific to the formulation type, such as the adhesive properties of patches, the sterility of parenteral products, or the aerodynamic properties of inhaled medications (ICH guidelines).

[0137] Evaluation of cPQ includes not only impurities related to the detection and quantification process, but also aggregates, fragments, and product variants, typically including charge variants (basic or acidic), size variants, oxidation-related variants, structural variants, and variants resulting from glycosylation of the Fc region of the antibody (Alt et al.). Examples of such monoclonal antibody variants are listed in Table 1 below.

[0138] Table 1. Examples of biopharmaceutical product variants.

[0139]

[0140] Risk assessments can be performed to analyze the impact of cPQ attributes and process parameters on safety, pharmacokinetics (PK), biological activity, and immunogenicity. This is typically done early in the development process based on initial experiments and prior knowledge, and the assessment is refined in later stages, incorporating more experimental data and mathematical models (ICH Guidelines; Alt et al.). Impact scores can be used to better define ambiguous terms, such as “high impact” or “low impact” using point scales (Alt et al.). Non-limiting examples of assays suitable for assessing product quality attributes include chromatography (including RPLC, IEX, AEX, CEX, SEC, HIC, HILIC, and MMC), mass spectrometry (including whole mass analysis, peptide mapping, and amino acid sequencing), spectroscopy (including UV / vis spectroscopy), capillary electrophoresis (including free-flow electrophoresis, isoelectric focusing, capillary isoelectric focusing, imaging capillary isoelectric focusing, and capillary region electrophoresis), gel electrophoresis (including SDS-PAGE and Western blotting), and ligand binding assays (including biolayer interferometry, enzyme-linked immunosorbent assay, and surface plasmon resonance).

[0141] Aggregation This is one of the most common cPQ-related problems in biopharmaceutical preparation because the drug can lead to a loss of protein activity and product yield, purity, and stability (Jing et al., 2012, Process Biochem., 47:69-75; Torkashvand and Vaziri, 2017, Iran Biomed. J., 21:131-141). Large aggregates are also suspected of causing adverse immune responses, such as allergic reactions, in patients. It is hypothesized that such large aggregates are mistaken by the immune system for bacterial cell walls, thus generating immunogenicity. However, due to the complexity of the phenomenon and contributing factors related to both the biopharmaceutical and the patient, the cause of the immune response cannot be solely attributed to the aggregates. Opponents of this hypothesis argue that the number of aggregates administered is too small to generate immunogenicity (Eon-Duval et al., 2012, Biotechnol. Prog., 28:608-622).

[0142] Based on their local characteristics, aggregates can be divided into different categories, as summarized in Table 2.

[0143] Table 2. Classification of protein aggregates.

[0144]

[0145] Aggregation can be caused by mechanical stress (shaking, stirring, pumping) or physicochemical stress, including changes in the pH and molar osmotic pressure of the cell culture medium, as well as changes in temperature, oxygen concentration, protein concentration, and exposure to air or metal surfaces (Jin et al.; Torkashvand and Vaziri; Eon-Duval et al.; Cromwell et al., 2006, *Journal of the Association of American Pharmaceutical Scientists (AAPS J.)*, 8:E572-E579). Therefore, aggregates can form at any step during preparation, formulation, or storage (Eon-Duval et al.). In cell culture media, intracellular aggregation can occur after protein expression and after secretion into the medium (Vázquez-Rey and Lang).

[0146] The mechanism of aggregation is not fully understood, but it is suspected that monomers with modified secondary or tertiary structures act as precursors or intermediates for aggregation, such as... Figure 3 As shown in (Jefferis, 2018, Generics & Biosimilars Initiat. J., 7:63-69). The altered structure exposes hydrophobic regions that were previously shielded by the unchanged protein, potentially leading to aggregate formation (Jin et al.).

[0147] Aggregates generated during the upstream process can be removed in the downstream purification step. While protein A affinity chromatography is not used for aggregate removal because aggregates may bind to the resin along with the monomeric antibody (Vázquez-Rey and Lang), other techniques such as size exclusion chromatography, cation exchange and anion exchange chromatography, or ultrafiltration (in the case of insoluble aggregates) can be used (Cromwell et al.). However, this may result in reduced yields; therefore, minimizing aggregation from the outset is more efficient (Jing et al.).

[0148] Aggregates can also form in downstream processes, particularly during the elution of antibodies captured by protein A resin and during the virus inactivation step, both of which require acidic conditions (Vázquez-Rey and Lang; Cromwell et al.). Furthermore, poor mixing during the VI step can lead to the formation of regions with pH levels lower than the expected VI conditions (Jin et al.).

[0149] The pH of the culture medium affects the charge distribution on the antibody surface, and thus influences intermolecular and intramolecular interactions. Strong electrostatic interactions have been found to cause protein unfolding, meaning that ionic strength is also an important factor. When high ionic strength shields the surface charge of an antibody, especially at low pH, the antibody may be unstable (Jin et al.).

[0150] Aggregation increases when the cell culture medium temperature is increased for a prolonged period (Cromwell et al.). Aggregate levels also show an increase with increasing protein concentration due to the greater likelihood of protein-protein interactions (Jin et al.).

[0151] Due to the biophysical similarity between aggregates and monomers, the removal of aggregates is more challenging than that of other impurities. Their removal typically relies primarily on purification steps, as affinity chromatography is ineffective (Yu et al.).

[0152] Providing optimal conditions for cell culture media, including feed strategy, conductivity, osmotic pressure, temperature, and pH, can help minimize aggregation levels. Osmotic agents that stabilize proteins (such as amino acids, sugars, and polyols) are often added to reduce aggregate levels (Torkashvand and Vaziri).

[0153] The presence of reducing and oxidizing agents containing divalent copper ions, cysteine, and cystine has been found to reduce protein aggregation (Jing et al.). Another study found that the addition of sodium chloride can reduce aggregation (Ju et al., 2009, J. Biotechnol., 143:145-150).

[0154] Other methods to reduce the buildup rate include using tank systems designed to minimize splashing and pumping, rapidly freezing the bulk solution and adding cryoprotectants, and eliminating top space in storage containers (Vázquez-Rey and Lang).

[0155] A variety of techniques can be used to measure aggregates, including sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), fluorescence spectroscopy, multi-angle laser scattering (MALLS), and size exclusion chromatography (SEC). SEC is the standard method in the biopharmaceutical industry because it can qualitatively and quantitatively measure aggregates and minimizes the impact on the environment and protein structure due to the use of a mild mobile phase (Fekete et al., 2014, J. Pharm. Biomed. Anal., 101:161-173).

[0156] Unlike other chromatographic techniques, SEC separates analytes through an entropy process based on the hydrodynamic radius of the analyte, rather than on the chemical properties of the analyte and its adsorption to the stationary phase. SEC columns contain porous particles with a controllable pore size distribution that can be permeated by smaller molecules. The largest molecules excluded from the pores (oligomers in this case) elute first, followed by smaller molecules (dimers, monomers, antibody fragments) (Brusotti et al., 2017, Chromatographia, 81:3-23).

[0157] Analytes eluted from an SEC column can be separated into different fractions based on elution time. For example, analytes eluted earlier than the functional form of the protein of interest (e.g., monomeric form) can be broadly classified as high molecular weight (HMW) species. HMW fractions can be further subdivided into, for example, very high molecular weight (vHMW) fractions and dimer fractions (representing the elution time of the dimer form of the protein of interest). Analytes eluted later than the functional form of the protein of interest can be broadly classified as low molecular weight (LMW) species, and can be further subdivided into LMW fractions and late tail fractions. Variants of the protein of interest with higher or lower molecular weights than the dominant species or the expected product can be referred to as "size variants."

[0158] The chromatographic material may include a size-limiting material, wherein the size-limiting material is a resin or membrane. The matrix used for size-limiting is preferably an inert gel medium, which may be a complex of cross-linked polysaccharides, such as cross-linked agarose and / or dextran in the form of spherical beads. The degree of cross-linking determines the pore size present in the swollen gel beads. Molecules larger than a certain size will not enter the gel beads and therefore move through the chromatographic bed the fastest. Smaller molecules (such as detergents, proteins, DNA, etc.) that enter the gel beads to varying degrees according to their size and shape are hindered as they pass through the bed. Therefore, molecules generally elute in order of decreasing molecular size.

[0159] Porous chromatographic resins suitable for size exclusion chromatography can be made from dextrose, agarose, polyacrylamide, or silica with different physical properties. Polymer combinations can also be used. The most commonly used is the polymer combination "SEPHADEX," available from Amersham Biosciences. Other size exclusion supports from different structural materials are also suitable, such as Toyopearl 55F (polymethacrylate, from Tosoh Bioscience, Montgomery Pa.) and Bio-Gel P-30 Fine (BioRad Laboratories, Hercules, California).

[0160] Fragmentation It is one of the key product quality attributes for assessing degradation in biopharmaceutical products, commonly seen in antibodies produced in CHO cells (Hu et al., 2021, *Protein Expression and Purification*, 186:105907). Cleavage describes the breakdown of proteins into smaller fragments (LMW, low molecular weight species) due to chemical disruption of covalent peptide bonds or enzymatic cleavage. Chemical disruption has been observed under high temperature and alkaline or acidic conditions (e.g., copper-induced cleavage of the hinge region of human IgG1 under high temperature and alkaline pH conditions, inhibited by EDTA (Vlasak and Ionescu, 2011, *Monoclonal Antibodies*, 3:253-263). Enzymatic cleavage is caused by proteolytic enzymes (e.g., protease or cathepsin D) (Eon-Duval et al.). Examples of immunoglobulin cleavage sites are found in… Figure 4 As shown in (O'Connor et al., 2017, Journal of Chromatography A, 1499:65-77).

[0161] In biopharmaceutical preparation, proteolytic enzymes are produced by host cells and released into the cell culture medium. The presence of these proteolytic enzymes leads to impurities and reduced stability, and thus reduces the half-life of the antibody of interest (Eon-Duval et al.). Fragmentation patterns can serve as a fingerprint for assessing product stability and preparation consistency, for example, when biopharmaceuticals are produced at multiple sites (Torkashvand and Vaziri; Vlasak and Ionescu).

[0162] The major cleavage sites of monoclonal antibodies are located around the domain interfaces and within the CH1 and CH2 (constant weight 1 and 2) domains. The former may lead to the loss of biological activity and other functions of one or more domains, while the latter may affect the structural integrity of the antibody, although the fragment may be preserved by disulfide bonds. Cleavage within variable domains may also affect the biological activity of the protein (Eon-Duval et al.).

[0163] Cleavages occurring in the complementarity-determining region (CDR) can affect the binding of monoclonal antibodies to their targets. In the case of hinge region cleavage, potency may be affected by reduced or lost function, depending on the presence of Fc and Fab regions in the fragment. Cleavages in the constant region can affect the circulating half-life and Fc-mediated effector function. Furthermore, cleavages may affect the aggregation rate (Vlasak and Ionescu).

[0164] Cleavage rate depends on the cell culture medium composition. The presence of cysteine ​​and EDTA, as well as trace elements such as zinc, manganese, and cobalt, has been found to decrease the cleavage rate, while the presence of copper has been reported to increase it (Torkashvand and Vaziri). Iron atoms have been found to catalyze peptide bond cleavage in the presence of histidine buffers containing IgG1 molecules with λ light chains (Vlasak and Ionescu).

[0165] Peptide bonds are inherently stable under physiological conditions, but they can undergo non-enzymatic hydrolysis under high temperatures or extreme pH conditions. Most cleavage occurs at cysteine, aspartic acid, glycine, asparagine, serine, and threonine. Except for glycine, all these residues have specific mechanisms that promote peptide bond cleavage, but other types of hydrolysis can also occur, such as those mediated by water or induced by free radicals (e.g., due to prolonged exposure to H₂O₂ (Vlasak and Ionescu)).

[0166] It has been observed that fragmentation increases even after short-term storage when the purification eluent is not neutralized with the acidic pH required for elution of protein A ligand. The difference in fragmentation rate is even more significant when the sample incubation temperature is increased from room temperature to 40°C (Hu et al.).

[0167] Cleavage can be detected and quantified based on the altered properties of the protein undergoing this process, including based on molecular size (using size exclusion chromatography (SEC), capillary electrophoresis using sodium dodecyl sulfate (CE-SDS), or sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE)), and based on the chemical properties of the protein's amino acid side chains (using a range of chromatographic techniques such as reversed-phase HPLC, cation exchange HPLC, or hydrophobic interaction HPLC). Additionally, mass spectrometry or, less commonly, N-terminal sequencing can be used to identify cleavage sites (Vlasak and Ionescu).

[0168] Size exclusion chromatography (SEC) is generally the preferred method for aggregate detection, but it can also be used to detect hinge region breakage. However, data processing can be difficult due to the resolution difference between the Fc-Fab peak and the monomer peak. Better resolution can be obtained when using SDS-PAGE and CE-SDS, and the latter can employ a fluorescence detector, thereby improving sensitivity (Vlasak and Ionescu).

[0169] When monoclonal antibodies undergo chemical degradation (including fragmentation, deamidation, isomerization, and oxidation) and post-translational modifications (such as glycosylation), elimination of the C-terminal lysine residue of the heavy chain, or N-terminal glutamine cyclization to pyroglutamic acid, they produce... Charge Variants (Torkashvand and Vaziri).

[0170] In the chemical degradation pathways responsible for acidic or basic species, the two most common covalent modifications observed in proteins and peptides are deamination and oxidation. Methionine, cysteine, histidine, tryptophan, and tyrosine are some of the most easily oxidized amino acids: Met and Cys are due to their sulfur atoms, while His, Trp, and Tyr are due to their aromatic rings.

[0171] Charge variants can also be caused by suboptimal cell culture conditions during the preparation process. The resulting heterogeneity leads to changes in the isoelectric point (pI), in vivo properties, stability, efficacy, and quality of antibodies (Wu et al., 2022, TrAC Trends in Analytical Chemistry, 150:116567), because charge heterogeneity can affect binding to other proteins and targets, and thus also affect the tissue distribution and pharmacokinetics of biotherapeutic agents (Torkashvand and Vaziri; Zhang et al., 2023, Journal of Pharmaceutical and Biomedical Analysis, 224:115178), especially for species with a pI difference ≥1.76.

[0172] Therefore, the analysis of charge variants as cPQ properties is crucial in the preparation of biopharmaceutical products. In fact, according to ICH guidelines, charge heterogeneity needs to be monitored during the production, storage, and transportation of biopharmaceuticals (Wu et al.).

[0173] The charged variants with relatively high isoelectric points are called basic variants, while those with lower pI are called acidic variants (Torkashvand and Vaziri). Basic variants have been reported to exhibit increased blood clearance and tissue retention, while acidic variants are associated with decreased systemic clearance and tissue retention (Chung et al., 2018, Biotechnology and Bioengineering, 115:1646-1665).

[0174] The main causes of acid variant formation include deamidation of asparagine residues in both constant and variable regions (especially complement-determining regions) (Torkashvand and Vaziri), as well as post-translational modification of sialic acid (Chung et al.).

[0175] Acidic charge variants are considered to have a greater negative impact on the pharmacokinetics and efficacy of biopharmaceuticals than basic variants, and therefore are usually preferentially reduced during preparation (Chung et al.). For example, it has been observed that antigen binding is reduced by 14-fold due to the deamidation of asparagine and glutamine (Huang et al., 2005, Analytical Chemistry, 77:1432-1439). It has been reported that oxidation of cysteine, methionine, tryptophan, histidine, and tyrosine reduces their binding to protein A (Bertolotti-Ciarlet, 2009, Molecular Immunology, 46:1878-1882; Gaza-Bulseco et al., 2008, Journal of Chromatography B, 870:55-62; Pan et al., 2009, Protein Science, 18:424-433), shortens the half-life (Gaza-Bulseco et al.), and leads to loss of activity (Hensel et al., 2011, PLoS ONE, 6:e17708).

[0176] The isomerization of aspartic acid (Asp-92) in human IgG2 has been found to produce a basic species that leads to complete antibody inactivation (Rehder et al., 2008, Biochemistry, 47:2518-2530). However, several processes that have been reported to cause the formation of basic species, including C-terminal modifications of lysine or arginine (Alt et al.), and N-terminal modifications of glutamine and glutamate (Manning et al., 2010, Pharm. Res., 27:544-575), have been reported to have little effect on pharmacokinetics, binding, or potency.

[0177] In some exemplary aspects, a sample including the protein of interest may contain more than one type of variant of the protein of interest. Such variants may include acidic and basic species. Acidic species are typically variants eluted earlier than the main peak from cation exchange chromatography (CEX) or later than the main peak from anion exchange chromatography (AEX), while basic species are variants eluted later than the main peak from CEX or earlier than the main peak from AEX. In one exemplary aspect, in normal polarity, basic species may migrate earlier than the main peak from isoelectric focusing (IEF), and acidic species may migrate later than the main peak from IEF. In one exemplary aspect, in reverse polarity, basic species may migrate later than the main peak from IEF, and acidic species may migrate earlier than the main peak from IEF.

[0178] As used herein, the terms “acidic species,” “AS,” “acidic region,” and “AR” refer to variants of proteins characterized by a total acidic charge.

[0179] In some respects, a sample may contain more than one type of acidic species variant. For example, but not limited to, total acidic species can be classified based on the chromatographic retention time of the peaks that appear or by using UV peaks or other absorbance peaks generated by IEF.

[0180] As used herein, the terms “oxidative species,” “OS,” or “oxidative variant” refer to variants of proteins formed through oxidation. Such oxidative species can also be detected by various methods, such as ion exchange, for example, WCX-10 HPLC (weak cation exchange chromatography), or IEF. Oxidative variants can be caused by oxidation occurring at histidine, cysteine, methionine, tryptophan, phenylalanine, and / or tyrosine residues.

[0181] As used herein, the terms "basic species," "basic region," and "BR" refer to a variant of a protein, such as an antibody or its antigen-binding moiety, characterized by a total basic charge relative to the dominant charge variant species present within the protein. For example, in recombinant protein formulations, such basic species can be detected by various methods, such as ion exchange, e.g., WCX-10 HPLC (weak cation exchange chromatography), or IEF. Exemplary variants may include, but are not limited to, lysine variants, isomerization of aspartic acid, succinimide formation at asparagine, methionine oxidation, amidation, incomplete disulfide bond formation, mutation from serine to arginine, glycosylation, cleavage, and aggregation. Typically, basic species elute later than the main peak during CEX analysis or earlier than the main peak during AEX analysis. (Chromatographic analysis of the acidic and basic species of recombinant monoclonal antibodies. Monoclonal Antibody, 1 Sep 2012; 4(5): 578–585. doi:10.4161 / mabs.21328, the entire teachings of which are incorporated herein by reference).

[0182] In some respects, a sample may contain more than one type of basic species variant. For example, but not limited to, total basic species can be segmented based on the chromatographic retention time of the peaks that appear or based on UV peaks or other absorbance peaks generated using IEF. Another instance where total basic species can be segmented can be based on the type of variant—e.g., structural variants or fragmentation variants.

[0183] In biopharmaceutical preparation, due to limited downstream purification capabilities, cell line development and upstream processes primarily govern the control of charge variant distribution. Cell line development includes protein engineering for controlling post-translational modifications, while upstream process development involves optimizing both process conditions (such as cell culture media) and post-culture treatments. For the reduction of acidic charge variants, pH and temperature are the main control factors, while the formation of basic charge variants is significantly influenced by other parameters, including dissolved oxygen and inoculum density in the production reactor (Chung et al.).

[0184] Lowering the temperature in a bioreactor can slow the rate of protein degradation, likely by reducing the amount of proteolytic enzymes released from dead cells (Yoon et al., 2003, Biotechnology and Bioengineering, 82:289-298). Indeed, lowering the temperature has been observed to reduce the amount of acidic charge variants, although said lowering the temperature increases the amount of basic charge variants formed (Torkashvand and Vaziri).

[0185] Many chromatographic methods have been used to characterize charge variants, including anion-exchange chromatography and cation-exchange chromatography (AEX and CEX), hydrophobic interaction chromatography, liquid chromatography-mass spectrometry (LC-MS), affinity chromatography, and reversed-phase high-performance liquid chromatography (RP-HPLC) (Chung et al.). However, most conventional chromatographic analytical techniques (except ion-exchange chromatography), even when coupled with mass spectrometry, cannot accurately distinguish charge variants with very similar isoelectric points (Zhang et al., 2023). Therefore, other methods have been developed for this purpose, including free-flow electrophoresis (FFE), isoelectric focusing gel electrophoresis (IEF), and more recently, imaging capillary isoelectric focusing electrophoresis (iCIEF) (Torkashvand and Vaziri).

[0186] As used herein, “isoelectric focusing” or “IEF,” also simply referred to as electrofocusing, is a technique for separating charged molecules (typically proteins or peptides) based on their isoelectric point (pI), such as the pH at which the molecule is uncharged. IEF works because molecules in an electric field that are at a pH gradient migrate toward their pI. Various techniques exist for performing IEF, all of which are encompassed under the term IEF as used herein. For example, in capillary isoelectric focusing (cIEF), the sample travels through a capillary based on an applied electric field. A UV detector can be used at a point along the capillary to detect the time it takes for an analyte (such as a protein) to travel through that point. Since the travel time through the capillary is directly related to the charge (pI) of the analyte, the change in the UV signal from a point in the capillary over time can be represented as a UV trace, which represents the change in charge (pI) of the sample component. In one exemplary embodiment, the UV trace generated by cIEF represents a charge variant of the protein of interest, where each UV peak represents a significant charge variant.

[0187] Variations of cIEF, such as imaging cIEF (icIEF), can also be used. Examples of suitable devices for icIEF analysis include CEInfinite (Advanced Electrophoresis Solutions), iCE3 (ProteinSimple), and Maurice (ProteinSimple). Following the focusing of charge variants in the sample, icIEF analysis may further include a mobilization step, where pressure is used to mobilize the focused sample through a detection window, and the focused sample may also be mobilized out of the separation capillary. Fractions can be collected from the mobilized sample. The collected fractions may correspond to the charge variant of interest. A fraction may contain more than one charge variant. Charge variants may be collected in more than one fraction.

[0188] iCIEF is routinely used in the biopharmaceutical industry due to its high throughput, high resolution, and reliability. UV detectors are used for quantification and obtaining the UV spectrum of products because of their high sensitivity. Furthermore, mass spectrometry methods such as electrospray ionization-mass spectrometry (ESI-MS) can be applied to identify charge variants by determining their molecular weight, although this technique is limited by interfering species commonly used in many IEF methods, such as methylcellulose or amphoteric electrolytes (Zhang et al., 2023; Schlecht et al., 2022, *Electrophoresis*, 44(5-6):540-548).

[0189] iCIEF can be used to analyze a range of species, including monoclonal antibodies, antibody-drug conjugates, glycoproteins, and vaccines. Recent developments in this technology include the use of fluorescence or chemiluminescence to improve detection sensitivity (Wu et al.). However, iCIEF involves complex operations and often requires repeated trials for optimization (Zhang et al., 2023).

[0190] Glycosylation This is a post-translational protein modification regulated by glycosyltransferases, characterized by the linkage of oligosaccharides to the polypeptide backbone of proteins. This occurs in the endoplasmic reticulum and Golgi apparatus of the cell (Ivarsson et al., 2014, Journal of Biotechnology, 188:88-96; Mao et al., 2023, Progress in Biotechnology, e3365).

[0191] Glycans can be O-linked or, more commonly, N-linked, depending on the linking site with the polypeptide backbone of the protein (the former is serine- or threonine-linked, and the latter is asparagine-linked (del Val et al., 2010, Progress in Biotechnology, 26:1505-1527)). Each IgG molecule is N-glycosylated at the asparagine (Asn-297) residue in the CH2 domain of the Fc region, and 15-20% of IgG is N-glycosylated in the light or heavy chain of the variable domain in the Fab region. Overall, glycosylation in the Fab region is more extensive due to the steric hindrance of glycosyltransferases in the Fc region caused by the proximity of heavy chains (Eon-Duval et al.).

[0192] Monoclonal antibodies produced by CHO cells exhibit an Fc region glycosylation pattern comparable to that of antibodies present in human blood (glycoproteins lacking α-galactose residues (Mao et al.)), but CHO cells can also produce low glycosyl and high mannose glycoforms, which are not present in most IgGs (Eon-Duval et al.).

[0193] Antibodies exhibit high glycogeneity: microheterogeneity—differences in glycoform distribution due to variations in Golgi in vivo processing; and macroheterogeneity—the absence or presence of glycans at glycosylation sites due to inefficient glycoprotein degradation or glycan transfer to sites (Reusch and Tejada; Mao et al.; del Val et al.).

[0194] The glycan profiles of therapeutic antibodies depend on process parameters such as cell culture conditions (glucose concentration (Villacrés et al., 2015, *Biotechnol. J.*, 10:1051-1066), redox potential (Dionne et al., 2017, *Biotechnol. J.*, 246:71-80), pH, shear stress, CO2, and ammonia content), the amount of expressed enzymes involved in glycosylation, and oligosaccharide substrate accessibility (Torkashvand and Vaziri; Ivarsson et al.). The high glycan heterogeneity of biopharmaceuticals and the significant impact of process parameter variations on antibody glycan profiles make monitoring PQ properties throughout the preparation process crucial for delivering consistent quality drug products (Ivarsson et al.).

[0195] Assessing the effects of glycosylation on the efficacy and safety of biopharmaceuticals is a complex process. Glycosylation patterns can affect the pharmacokinetics, immunogenicity, and bioactivity of biotherapeutic agents, including increasing protein solubility, reducing aggregation rates, preventing enzymatic degradation, and thus increasing cycle life in some cases (Kateja et al., 2018, Journal of Chromatography A, 1579:60-72), and affecting clearance rates and effector functions (Ivarsson et al.).

[0196] Non-glycosylated immunoglobulins exhibit loss of function, conformational differences, increased aggregation rates, and decreased thermal stability. On the other hand, antibodies with high galactosylation have been shown to lead to increased CDC activity; high-mannose variants may exhibit shortened serum half-life, while low levels of core fucose may lead to increased ADCC (Reusch and Tejada).

[0197] Therefore, glycan profiles are one of the key product quality attributes. Considering the potential impact of glycan profiles on the efficacy and safety of biotherapeutic agents, it is necessary to monitor glycan profiles in the preparation of therapeutic antibodies, reduce batch-to-batch variability, and avoid potential economic losses when batches with undesirable protein glycosylation patterns need to be discarded.

[0198] Assessing macroscopic heterogeneity includes identifying glycosylation sites, which can be accomplished through collision-induced dissociation experiments and mass spectrometry (del Val et al.) or capillary electrophoresis (Mao et al.). Characterizing microscopic heterogeneity (glycan profiling) involves detecting and quantifying oligosaccharides present at glycosylation sites (del Val et al.). Techniques used for glycan profiling include capillary electrophoresis (Mao et al.), mass spectrometry (Torkashvand and Vaziri), anion exchange chromatography, fluorophore-assisted carbohydrate electrophoresis (FACE) (del Val et al.), and hydrophilic interaction liquid chromatography with a fluorescence detector (HILIC) (Villacrés et al.).

[0199] HILIC uses a polar stationary phase and a relatively polar mobile phase (e.g., acetonitrile), and separates analytes based on the difference in hydrogen bonding ability between the analytes and the stationary phase. Glycan analysis using HILIC is performed using both conventional packed columns employed in HPLC and UPLC instruments and elution plates (del Val et al.; Lauber et al., 2013, Waters Application Note, 720004717EN).

[0200] Exemplary embodiments

[0201] This disclosure provides methods and systems for enriching a protein of interest from a sample, such as a cell culture sample. In some exemplary embodiments, the method may include: (a) contacting a cell culture sample comprising the protein of interest with a centrifugal (rotational) column comprising an affinity resin to produce an immobilized sample, wherein the affinity resin specifically binds to the protein of interest; (b) subjecting the immobilized sample to at least one washing step; and (c) subjecting the immobilized sample from (b) to at least one elution step to produce enriched protein of interest.

[0202] On one hand, the protein of interest is selected from the group consisting of: therapeutic proteins, receptors, antigen-binding proteins, antibodies, monoclonal antibodies, multispecific antibodies, bispecific antibodies, antibody-derived proteins, fusion proteins, receptor fusion proteins, trap proteins, fragments thereof, variants thereof, and combinations thereof. On the other hand, the protein of interest is a monoclonal antibody. In another aspect, the protein of interest is dupilumab (DUPIXENT®).

[0203] On the one hand, the protein of interest contains an Fc domain.

[0204] On one hand, the protein of interest is selected from the group consisting of: Fc fusion proteins, receptor Fc fusion proteins, and ScFv-Fc fusion proteins. On the other hand, the protein of interest is an anti-vascular endothelial growth factor (VEGF) antibody or an anti-VEGF receptor Fc fusion protein. In a particular aspect, the protein of interest is aflibercept.

[0205] On the one hand, the protein of interest is a recombinant protein.

[0206] On the one hand, the cell culture samples are derived from cell cultures selected from eukaryotic cell cultures, mammalian cell cultures, or insect cell cultures. On the other hand, the cell culture samples are selected from the group consisting of: CHO, CHO-K1, CHO DUX B-11, Veggie-CHO, GS-CHO, S-CHO, CHO lec, COS, Vero, CV-1, HEK293, MCDK, HaK, BHK2, HeLa, HepG2, WI38, MRC 5, Colo25, HB 8065, HL-60, Jurkat, Daudi, A431, U937, 3T3, L cells, C127, SP2 / 0, NS-0, MMT, variants thereof, and combinations thereof.

[0207] On the one hand, the cell culture samples are derived from CHO cell culture, CHO-K1 cell culture, BHK cell culture, HEK293 cell culture, Sf9 cell culture or variants thereof.

[0208] On the one hand, the cell culture sample is a clarified cell culture sample.

[0209] On one hand, the method further includes clarifying the cell culture sample prior to step (a). In one specific aspect, the clarification includes centrifugation, filtration, and / or precipitation of insoluble components.

[0210] On the one hand, the cell culture sample is an upstream cell culture sample.

[0211] On one hand, the cell culture samples are collected from the cell culture on one of the following days: day 1 to day 20, day 3 to day 15, day 3 to day 12, day 3 to day 10, day 5 to day 10, day 5 to day 12, or day 5 to day 13. On the other hand, the cell culture samples are collected from the cell culture on one of the following days: day 1, day 2, day 3, day 4, day 5, day 6, day 7, day 8, day 9, day 10, day 11, day 12, day 13, day 14, day 15, day 16, day 17, day 18, day 19, or day 20.

[0212] On one hand, the method further comprises repeating steps (a)-(c) (contact, washing, and elution) at least once. In one specific aspect, the method is repeated using at least a first cell culture sample and a second cell culture sample collected from the same cell culture. In a more specific aspect, the first cell culture sample is collected on a first day, and the second cell culture sample is collected on a second day. In another specific aspect, the first and second cell culture samples are collected at a time between approximately 3 hours, approximately 6 hours, approximately 12 hours, approximately 18 hours, approximately 24 hours, approximately 1 day, approximately 2 days, approximately 3 days, approximately 4 days, approximately 5 days, approximately 6 days, approximately 7 days, approximately 8 days, approximately 9 days, or approximately 10 days after the sample collection.

[0213] In one aspect, the method further comprises performing steps (a)-(c) (contact, washing, and elution) in parallel on at least two cell culture samples. In a specific aspect, the at least two cell culture samples are derived from two different cell cultures.

[0214] On one hand, the contact step includes combining the cell culture sample and a binding buffer. In one specific aspect, the binding buffer comprises Tris-buffered saline or sodium phosphate. In another specific aspect, the binding buffer further comprises sodium chloride. In yet another specific aspect, the concentration of sodium phosphate is 10 mM to 30 mM, 15 mM to 25 mM, about 10 mM, about 15 mM, about 20 mM, about 25 mM, or about 30 mM. In yet another specific aspect, the concentration of sodium chloride is 100 mM to 500 mM, about 100 mM, about 200 mM, about 300 mM, about 400 mM, or about 500 mM. In yet another specific aspect, the pH of the binding buffer is 7.1 to 7.3.

[0215] On one hand, the binding buffer comprises Tris-buffered saline, sodium phosphate, HEPES, or Tris. On another hand, the binding buffer further comprises sodium chloride or calcium chloride. On yet another specific hand, the pH of the binding buffer is between 6.0 and 8.0.

[0216] On one hand, the contact step includes adding a combined volume of binding buffer and the cell culture sample to the column, the combined volume being 250 µL to 1000 µL, 300 µL to 900 µL, 400 µL to 800 µL, 500 µL to 700 µL, 550 µL to 650 µL, 590 µL to 610 µL, 599 µL to 601 µL, or about 600 µL.

[0217] On one hand, the contact step includes adding a certain volume of the cell culture sample to the column, the volume being 50 µL to 100 µL, 60 µL to 90 µL, 70 µL to 80 µL, about 70 µL, about 71 µL, about 72 µL, about 73 µL, about 74 µL, about 75 µL, about 76 µL, about 77 µL, about 78 µL, about 79 µL, or about 80 µL.

[0218] On one hand, the contacting step includes adding a certain amount of protein to the column, the amount being 100.5 µg to 804 µg, 250 µg to 1 g, 350 µg to 900 µg, 450 µg to 804 µg, 500 µg to 700 µg, 550 µg to 650 µg, 575 µg to 625 µg, 590 µg to 610 µg, about 595 µg, about 596 µg, about 597 µg, about 598 µg, about 599 µg, about 600 µg, about 601 µg, about 602 µg, about 603 µg, about 604 µg, or about 605 µg.

[0219] On the one hand, the affinity resin is protein A resin, protein G resin, or a combination thereof.

[0220] On one hand, the at least one washing step comprises adding a washing buffer to the column and centrifuging the column to produce a washed flow solution. In one specific aspect, the washing buffer comprises Tris-buffered saline, sodium acetate, or sodium phosphate. In a more specific aspect, the washing buffer further comprises sodium chloride. In another specific aspect, the concentration of sodium phosphate or sodium acetate is 10 mM to 30 mM, 15 mM to 25 mM, about 10 mM, about 15 mM, about 20 mM, about 25 mM, or about 30 mM. In another specific aspect, the concentration of sodium chloride is 100 mM to 500 mM, about 100 mM, about 200 mM, about 300 mM, about 400 mM, or about 500 mM. In another specific aspect, the pH of the washing buffer is 7.1 to 7.3.

[0221] In one aspect, the washing buffer comprises Tris-buffered saline, sodium phosphate or sodium acetate, HEPES or Tris. In another aspect, the washing buffer further comprises sodium chloride or calcium chloride. In yet another aspect, the pH of the washing buffer is from 6.0 to 8.0.

[0222] In another specific aspect, the volume of the washing buffer is approximately 600 µL. In yet another specific aspect, the centrifugation is performed at approximately 100 relative centrifugal force (RCF). In still another specific aspect, the centrifugation is performed for approximately 30 seconds, approximately 1 minute, approximately 90 seconds, or approximately 2 minutes.

[0223] On the one hand, the number of washing steps is one, two, or three. On the other hand, the binding buffer and the washing buffer are the same.

[0224] On one hand, the at least one elution step comprises adding an elution buffer to the column and centrifuging the column to produce an eluent. In one specific aspect, the elution buffer comprises acetic acid or glycine. In a more specific aspect, the concentration of the acetic acid is 0.1% to 0.3%, 0.12% to 0.27%, 0.12% to 0.24%, about 0.12%, about 0.24%, 15 mM to 50 mM, 16 mM to 45 mM, 20 mM to 40 mM, about 15 mM, about 16 mM, about 20 mM, about 25 mM, about 30 mM, about 35 mM, about 40 mM, about 45 mM, or about 50 mM. In another specific aspect, the concentration of the glycine is about 0.1 M. In yet another specific aspect, the volume of the elution buffer is 300 µL to 500 µL, 350 µL to 450 µL, about 300 µL, about 350 µL, about 400 µL, about 450 µL, or about 500 µL. In still another specific aspect, the centrifugation is performed at about 100 RCF. In yet another specific aspect, the centrifugation is performed for about 30 seconds, about 1 minute, about 90 seconds, or about 2 minutes.

[0225] On the one hand, the pH of the elution buffer is below 4, 1 to 4, 2 to 4, 2.5 to 3.5, 2.8 to 3.2, about 1, about 1.5, about 2, about 2.5, about 3, or about 3.5.

[0226] On the one hand, the number of elution steps is one, two, or three.

[0227] On one hand, the at least one elution step includes adding a neutralization buffer to the column. In one specific aspect, the neutralization buffer comprises a Tris base. In a more specific aspect, the concentration of the Tris base is 1 M to 2 M, about 1 M, about 1.5 M, or about 2 M. In another specific aspect, the volume of the neutralization buffer is 5 µL to 50 µL, about 5 µL, about 10 µL, about 20 µL, or about 30 µL.

[0228] On the one hand, the enriched protein of interest yields a yield of more than 50%, more than 60%, more than 70%, more than 80%, more than 90%, more than 95%, more than 99%, approximately 60%, approximately 70%, approximately 75%, approximately 80%, approximately 85%, approximately 90%, approximately 91%, approximately 92%, approximately 93%, approximately 94%, approximately 95%, approximately 96%, approximately 97%, approximately 98%, approximately 99%, or approximately 100%.

[0229] On the one hand, the amount of protein in the enriched protein of interest is greater than 10 µg, greater than 20 µg, greater than 50 µg, greater than 100 µg, greater than 200 µg, greater than 300 µg, greater than 400 µg, greater than 500 µg, greater than 600 µg, greater than 700 µg, greater than 800 µg, greater than 900 µg, greater than 1000 µg, greater than 1100 µg, greater than 1200 µg, about 10 µg, about 20 µg, about 50 µg, about 100 µg, about 200 µg, about 300 µg, about 400 µg, about 500 µg, about 600 µg, about 700 µg, about 800 µg, about 900 µg, about 1000 µg, about 1100 µg, or about 1200 µg.

[0230] On the one hand, the concentration of the enriched protein of interest is greater than 0.01 µg / µL, greater than 0.05 µg / µL, greater than 0.1 µg / µL, greater than 0.2 µg / µL, greater than 0.5 µg / µL, greater than 1 µg / µL, greater than 2 µg / µL, about 0.05 µg / µL, about 0.1 µg / µL, about 0.2 µg / µL, about 0.5 µg / µL, about 1 µg / µL, about 1.5 µg / µL, about 2 µg / µL, or about 2.5 µg / µL.

[0231] On the one hand, the column can be reused to perform the method at least twice.

[0232] On the one hand, the duration of the method is less than 24 hours, less than 12 hours, less than 6 hours, less than 3 hours, less than 2 hours, less than 1 hour, less than 30 minutes, about 3 hours, about 2 hours, about 1.5 hours, about 1 hour, about 50 minutes, about 45 minutes, about 40 minutes, about 30 minutes, or about 20 minutes.

[0233] On one hand, the method further includes characterizing the enriched protein of interest. On the other hand, the method further includes characterizing at least one product quality attribute of the enriched protein of interest. In one specific aspect, the at least one product quality attribute is a critical product quality attribute. In yet another aspect, the method further includes subjecting the enriched protein of interest to chromatography, mass spectrometry, spectroscopy, capillary electrophoresis, gel electrophoresis, and / or ligand binding assays.

[0234] On one hand, the method further comprises characterizing at least one size variant of the enriched protein of interest. On the other hand, the method further comprises characterizing at least one high molecular weight species of the enriched protein of interest. In one specific aspect, the characterization comprises subjecting the enriched protein of interest to size exclusion chromatography (SEC) analysis.

[0235] On one hand, the method further includes characterizing at least one fragment of the enriched protein of interest. In one specific aspect, the characterization includes subjecting the enriched protein of interest to capillary electrophoresis and sodium dodecyl sulfate (CE-SDS) analysis.

[0236] On one hand, the method further includes characterizing at least one charge variant of the enriched protein of interest. In one specific aspect, the characterization includes subjecting the enriched protein of interest to imaging capillary isoelectric focusing electrophoresis (iCIEF).

[0237] On one hand, the method further includes at least one glycan characterizing the enriched protein of interest. In a specific aspect, the characterization includes subjecting the enriched protein of interest to hydrophilic interaction chromatography (HILIC) analysis.

[0238] In one specific aspect, the method further includes using the at least one product quality attribute to determine whether the cell culture should continue or be terminated. In another specific aspect, the method further includes using the at least one product quality attribute to determine whether the cell culture should be modified. In yet another specific aspect, modification of the cell culture may include modification of temperature, pH, dissolved oxygen or other gases or cell culture medium.

[0239] This disclosure also provides a method for monitoring at least one product quality attribute of a recombinant protein of interest. In some exemplary embodiments, the method may include (a) obtaining a first cell culture sample comprising the recombinant protein of interest from a first time point; (b) contacting the first cell culture sample with a centrifuge column comprising an affinity resin to produce an immobilized sample, wherein the affinity resin specifically binds to the protein of interest; (c) subjecting the immobilized sample to at least one washing step; (d) subjecting the immobilized sample from (c) to at least one elution step to produce a first enriched sample; (e) repeating steps (b)-(d) from at least one additional time point with at least one additional cell culture sample comprising the recombinant protein of interest to produce at least one additional enriched sample, wherein the first cell culture sample and the at least one additional cell culture sample are obtained from the same cell culture; and (f) characterizing at least one product quality attribute from the first enriched sample and the at least one additional enriched sample to monitor at least one product quality attribute of the recombinant protein of interest.

[0240] On one hand, the recombinant protein of interest is selected from the group consisting of: therapeutic proteins, receptors, antigen-binding proteins, antibodies, monoclonal antibodies, multispecific antibodies, bispecific antibodies, antibody-derived proteins, fusion proteins, receptor fusion proteins, trap proteins, fragments thereof, variants thereof, and combinations thereof. On the other hand, the recombinant protein of interest is a monoclonal antibody. In another aspect, the recombinant protein of interest is dupilumab (DUPIXENT®).

[0241] On the one hand, the recombinant protein of interest contains an Fc domain.

[0242] On one hand, the recombinant protein of interest is selected from the group consisting of: Fc fusion proteins, receptor Fc fusion proteins, and ScFv-Fc fusion proteins. On the other hand, the recombinant protein of interest is an anti-vascular endothelial growth factor (VEGF) antibody or an anti-VEGF receptor Fc fusion protein. In a particular aspect, the recombinant protein of interest is aflibercept.

[0243] On the one hand, the cell culture is a eukaryotic cell culture, a mammalian cell culture, or an insect cell culture. On the other hand, the cell culture is selected from the group consisting of: CHO, CHO-K1, CHO DUX B-11, Veggie-CHO, GS-CHO, S-CHO, CHO lec, COS, Vero, CV-1, HEK293, MCDK, HaK, BHK2, HeLa, HepG2, WI38, MRC 5, Colo25, HB 8065, HL-60, Jurkat, Daudi, A431, U937, 3T3, L cells, C127, SP2 / 0, NS-0, MMT, variants thereof, and combinations thereof.

[0244] On the one hand, the cell culture is a CHO cell culture, a CHO-K1 cell culture, a BHK cell culture, a HEK293 cell culture, an Sf9 insect cell culture, or a variant thereof.

[0245] On the one hand, the first cell culture sample and the at least one other cell culture sample are clarified cell culture samples.

[0246] On one hand, the method further includes clarifying the first cell culture sample and the at least one additional cell culture sample prior to step (a). In one specific aspect, the clarification includes centrifugation, filtration, and / or precipitation of insoluble components.

[0247] On the one hand, the first cell culture sample and the at least one other cell culture sample are upstream cell culture samples.

[0248] On one hand, the first cell culture sample and / or the at least one additional cell culture sample are collected from the cell culture on one of the following days: day 1 to day 20, day 3 to day 15, day 3 to day 12, day 3 to day 10, day 5 to day 10, day 5 to day 12, or day 5 to day 13. On the other hand, the first cell culture sample and / or the at least one additional cell culture sample are collected from the cell culture on one of the following days: day 1, day 2, day 3, day 4, day 5, day 6, day 7, day 8, day 9, day 10, day 11, day 12, day 13, day 14, day 15, day 16, day 17, day 18, day 19, or day 20.

[0249] On the one hand, the first cell culture sample and / or the at least one other cell culture sample were collected at some time between about 3 hours, about 6 hours, about 12 hours, about 18 hours, about 24 hours, about 1 day, about 2 days, about 3 days, about 4 days, about 5 days, about 6 days, about 7 days, about 8 days, about 9 days, or about 10 days.

[0250] On the one hand, steps (b)-(d) of the method are performed in parallel on the first cell culture sample and the at least one other cell culture sample.

[0251] On one hand, the contact step includes combining the first cell culture sample and / or the at least one additional cell culture sample with a binding buffer. In one specific aspect, the binding buffer comprises Tris-buffered saline or sodium phosphate. In another specific aspect, the sodium phosphate concentration is 10 mM to 30 mM, 15 mM to 25 mM, about 10 mM, about 15 mM, about 20 mM, about 25 mM, or about 30 mM. In yet another specific aspect, the pH of the binding buffer is 7.1 to 7.3.

[0252] On one hand, the contact step includes adding a combined volume of binding buffer and cell culture sample to the column, the combined volume being 250 µL to 1000 µL, 300 µL to 900 µL, 400 µL to 800 µL, 500 µL to 700 µL, 550 µL to 650 µL, 590 µL to 610 µL, 599 µL to 601 µL, or about 600 µL.

[0253] On one hand, the contact step includes adding a volume of the first cell culture sample or the at least one other cell culture sample to the column, the volume being 50 µL to 100 µL, 60 µL to 90 µL, 70 µL to 80 µL, about 70 µL, about 71 µL, about 72 µL, about 73 µL, about 74 µL, about 75 µL, about 76 µL, about 77 µL, about 78 µL, about 79 µL, or about 80 µL.

[0254] On one hand, the contacting step includes adding a certain amount of protein to the column, the amount being 100.5 µg to 804 µg, 250 µg to 1 g, 350 µg to 900 µg, 450 µg to 804 µg, 500 µg to 700 µg, 550 µg to 650 µg, 575 µg to 625 µg, 590 µg to 610 µg, about 595 µg, about 596 µg, about 597 µg, about 598 µg, about 599 µg, about 600 µg, about 601 µg, about 602 µg, about 603 µg, about 604 µg, or about 605 µg.

[0255] On the one hand, the affinity resin is protein A resin, protein G resin, or a combination thereof.

[0256] On one hand, the at least one washing step comprises adding a washing buffer to the column and centrifuging the column to produce a washed flow solution. In one specific aspect, the washing buffer comprises Tris-buffered saline, sodium acetate, or sodium phosphate or sodium acetate. In another specific aspect, the concentration of sodium phosphate or sodium acetate is 10 mM to 30 mM, 15 mM to 25 mM, about 10 mM, about 15 mM, about 20 mM, about 25 mM, or about 30 mM. In another specific aspect, the pH of the washing buffer is 7.1 to 7.3. In yet another specific aspect, the volume of the washing buffer is about 600 µL. In still another specific aspect, the centrifugation is performed at about 100 relative centrifugal force (RCF). In yet another specific aspect, the centrifugation is performed for about 30 seconds, about 1 minute, about 90 seconds, or about 2 minutes.

[0257] On the one hand, the number of washing steps is one, two, or three. On the other hand, the binding buffer and the washing buffer are the same.

[0258] On one hand, the at least one elution step comprises adding an elution buffer to the column and centrifuging the column to produce an eluent. In one specific aspect, the elution buffer comprises acetic acid or glycine. In a more specific aspect, the concentration of the acetic acid is 0.1% to 0.3%, 0.12% to 0.27%, 0.12% to 0.24%, about 0.12%, about 0.24%, 15 mM to 50 mM, 16 mM to 45 mM, 20 mM to 40 mM, about 15 mM, about 16 mM, about 20 mM, about 25 mM, about 30 mM, about 35 mM, about 40 mM, about 45 mM, or about 50 mM. In another specific aspect, the concentration of the glycine is about 0.1 M. In yet another specific aspect, the volume of the elution buffer is 300 µL to 500 µL, 350 µL to 450 µL, about 300 µL, about 350 µL, about 400 µL, about 450 µL, or about 500 µL. In still another specific aspect, the centrifugation is performed at about 100 RCF. In yet another specific aspect, the centrifugation is performed for about 30 seconds, about 1 minute, about 90 seconds, or about 2 minutes.

[0259] On the one hand, the pH of the elution buffer is below 4, 1 to 4, 2 to 4, 2.5 to 3.5, 2.8 to 3.2, about 1, about 1.5, about 2, about 2.5, about 3, or about 3.5.

[0260] On the one hand, the number of elution steps is one, two, or three.

[0261] On one hand, the at least one elution step includes adding a neutralization buffer to the column. In one specific aspect, the neutralization buffer comprises a Tris base. In a more specific aspect, the concentration of the Tris base is 1 M to 2 M, about 1 M, about 1.5 M, or about 2 M. In another specific aspect, the volume of the neutralization buffer is 5 µL to 50 µL, about 5 µL, about 10 µL, about 20 µL, or about 30 µL.

[0262] On the one hand, the enriched protein of interest yields a yield of more than 50%, more than 60%, more than 70%, more than 80%, more than 90%, more than 95%, more than 99%, approximately 60%, approximately 70%, approximately 75%, approximately 80%, approximately 85%, approximately 90%, approximately 91%, approximately 92%, approximately 93%, approximately 94%, approximately 95%, approximately 96%, approximately 97%, approximately 98%, approximately 99%, or approximately 100%.

[0263] On the one hand, the amount of protein in the enriched protein of interest is greater than 10 µg, greater than 20 µg, greater than 50 µg, greater than 100 µg, greater than 200 µg, greater than 300 µg, greater than 400 µg, greater than 500 µg, greater than 600 µg, greater than 700 µg, greater than 800 µg, greater than 900 µg, greater than 1000 µg, greater than 1100 µg, greater than 1200 µg, about 10 µg, about 20 µg, about 50 µg, about 100 µg, about 200 µg, about 300 µg, about 400 µg, about 500 µg, about 600 µg, about 700 µg, about 800 µg, about 900 µg, about 1000 µg, about 1100 µg, or about 1200 µg.

[0264] On the one hand, the concentration of the enriched protein of interest is greater than 0.01 µg / µL, greater than 0.05 µg / µL, greater than 0.1 µg / µL, greater than 0.2 µg / µL, greater than 0.5 µg / µL, greater than 1 µg / µL, greater than 2 µg / µL, about 0.05 µg / µL, about 0.1 µg / µL, about 0.2 µg / µL, about 0.5 µg / µL, about 1 µg / µL, about 1.5 µg / µL, about 2 µg / µL, or about 2.5 µg / µL.

[0265] On the one hand, the column can be reused to perform the method at least twice.

[0266] On the one hand, the duration of the method is less than 24 hours, less than 12 hours, less than 6 hours, less than 3 hours, less than 2 hours, less than 1 hour, less than 30 minutes, about 3 hours, about 2 hours, about 1.5 hours, about 1 hour, about 50 minutes, about 45 minutes, about 40 minutes, about 30 minutes, or about 20 minutes.

[0267] On one hand, the characterization includes subjecting the first enriched sample and the at least one additional enriched sample to chromatography, mass spectrometry, spectroscopy, capillary electrophoresis, gel electrophoresis and / or ligand binding determination.

[0268] On one hand, the at least one product quality attribute includes at least one size variant, high molecular weight species, or low molecular weight species of the recombinant protein of interest. In a specific aspect, the characterization includes subjecting the first enriched sample and the at least one additional enriched sample to size exclusion chromatography (SEC) analysis.

[0269] On one hand, the at least one product quality attribute includes at least one fragment of the recombinant protein of interest. On another hand, the characterization includes subjecting the first enriched sample and the at least one additional enriched sample to capillary electrophoresis and sodium dodecyl sulfate (CE-SDS) analysis.

[0270] On one hand, the at least one product quality attribute includes at least one charged variant of the recombinant protein of interest. On another hand, the characterization includes subjecting the first enriched sample and the at least one additional enriched sample to imaging capillary isoelectric focusing electrophoresis (iCIEF).

[0271] On one hand, the at least one product quality attribute includes at least one glycosylated variant of the recombinant protein of interest. On another hand, the characterization includes subjecting the first enriched sample and the at least one additional enriched sample to hydrophilic interaction chromatography (HILIC) analysis.

[0272] In one specific aspect, the method further includes using the at least one product quality attribute to determine whether the cell culture should continue or be terminated. In another specific aspect, the method further includes using the at least one product quality attribute to determine whether the cell culture should be modified. In yet another specific aspect, modification of the cell culture may include modification of temperature, pH, dissolved oxygen or other gases or cell culture medium.

[0273] This disclosure further provides a method for profiling product quality attributes of an antibody of interest from an upstream cell culture sample. In some exemplary embodiments, the method may include (a) contacting an upstream cell culture sample including the antibody of interest with a centrifuge column comprising a protein A resin to produce an immobilized sample; (b) subjecting the immobilized sample to at least one washing step; (c) subjecting the immobilized sample from (b) to at least one elution step to produce an enriched antibody of interest; and (d) characterizing at least two product quality attributes of the enriched antibody of interest to analyze the product quality attributes of the antibody of interest, wherein the at least two product quality attributes include aggregation, charge change, glycosylation, and / or fragmentation.

[0274] This disclosure further provides a method for evaluating the product quality attributes of antibodies of interest from upstream cell culture samples. In some exemplary embodiments, the method may include (a) contacting a clarified upstream cell culture sample, comprising an antibody of interest and a binding buffer comprising about 20 mM sodium phosphate (pH 7.10-7.30), with a centrifuge column comprising protein A resin to produce an immobilized sample, wherein the clarified upstream cell culture sample has a protein content of about 600 µg, and the combined volume of the clarified upstream cell culture sample and the binding buffer is about 600 µL; (b) subjecting the immobilized sample to two washing steps, wherein each washing step includes adding about 600 µL of about 20 mM sodium phosphate (pH 7.10-7.30) to the column and centrifuging the column at about 100 RCF for about 1 minute; and (c) subjecting the immobilized sample from (b) to two elution steps to produce an eluent, wherein the elution steps include adding about 400 µL of about 40 mM acetic acid (pH 2.80-3.20) to the column, agitating the column, and adding about 30 (d) Add approximately 2 µL of Tris base to the column and centrifuge the column at approximately 100 RCF for approximately 1 minute; and characterize at least two product quality attributes of the enriched antibody of interest to analyze the product quality attributes of the antibody of interest, wherein the at least two product quality attributes include aggregation, charge change, glycosylation, and / or fragmentation. In one aspect, the protein of interest is dupilumab.

[0275] In some exemplary embodiments, the method may include (a) contacting a clarified upstream cell culture sample, comprising the antibody of interest and a binding buffer comprising about 10 mM sodium phosphate and 500 mM NaCl, with a centrifuge column comprising protein A resin to produce an immobilized sample, wherein the clarified upstream cell culture sample has a protein content of about 600 µg and the combined volume of the clarified upstream cell culture sample and the binding buffer is about 600 µL; (b) subjecting the immobilized sample to two washing steps, wherein a first washing step comprises adding about 600 µL of about 10 mM sodium phosphate and 500 mM NaCl to the column, and a second washing step comprises adding about 600 µL of about 20 mM sodium acetate and centrifuging the column at about 100 RCF for about 1 minute; and (c) subjecting the immobilized sample from (b) to two elution steps to produce an eluent, wherein the elution steps comprise adding about 400 µL of about 20 mM acetate to the column, agitating the column, and adding about 5 µL of about 2 M... Tris base was added to the column, and the column was centrifuged at approximately 100 RCF for approximately 1 minute; and (d) characterizing at least two product quality attributes of the enriched antibody of interest to analyze the product quality attributes of the antibody of interest, wherein the at least two product quality attributes include aggregation, charge change, glycosylation, and / or fragmentation. In one aspect, the protein of interest is aflibercept.

[0276] It should be understood that the present invention is not limited to any of the aforementioned proteins, proteins of interest, polypeptides of interest, recombinant proteins, antibodies, antibody fragments, cells, cell types, cell lines, cell culture media, protein alkylating agents, protein denaturing agents, protein reducing agents, digestive enzymes, samples, chromatographic methods, or product quality attributes, and any protein, protein of interest, polypeptide of interest, recombinant protein, antibody, antibody fragment, cell, cell type, cell line, cell culture media, protein alkylating agents, protein denaturing agents, protein reducing agents, digestive enzymes, samples, chromatographic methods, or product quality attributes can be selected by any suitable method.

[0277] The invention will be more fully understood by referring to the following examples. However, the following examples should not be construed as limiting the scope of the invention.

[0278] Example

[0279] Cell culture medium, such as Figure 5The cell culture media shown, containing genetically modified Chinese hamster ovary (CHO) cells, the antibody of interest, other proteins, and culture medium components, originated from different stages of the upstream preparation process in the production bioreactor and from the in-house production facility at cell harvest. Dupilumab was produced in Chinese hamster ovary (CHO) cells and cultured and harvested using chemically defined media (CDM) in laboratory-scale or small-scale bioreactors. Aflibercept was produced in CHO cells and cultured and harvested using soybean hydrolysate in a small-scale bioreactor. Harvested samples were stored at 2–8°C or -80°C until use. Routine preparation samples were obtained from the bioreactor every other day or on day 10.4 before harvesting the entire cell culture batch. Samples were clarified by centrifugation to remove CHO cells, and the supernatant was collected, aliquoted, and stored at -80°C until use.

[0280] Antibody purification method development. Antibody purification was performed using protein A affinity chromatography with a commercially available rotating column containing protein A (ProA) resin. Initially, the purification method relied on the column manufacturer's user guide and buffer kits from the same manufacturer, requiring only minor adjustments. Subsequent runs were then performed using internally produced optimized buffers (see Tables 1 and 2) and the manufacturer's protocols, with minor variations and additional steps (see Tables 1 and 2). The obtained run-through and eluent, as well as the filtered cell culture samples, were analyzed using high-performance liquid chromatography (HPLC) with a photodiode array (PDA) detector or a UV detector. The concentration of the first eluent for each sample loading was analyzed using Solo-VPE via UV-Vis spectroscopy. Each titer assay run included internal standards specific to the antibody of interest for calibration. The workflow of the total protein A chromatography method is described in [link to workflow description]. Figure 6A The general workflow of each step in the protein A chromatography method is shown in [the diagram]. Figure 6B It is displayed in the middle.

[0281] The protein purification protocol and initial adjustments provided by the ProA column manufacturer are described below. Six replicates were used in this experiment, and the obtained samples (flux after antibody binding, washing, and elution steps, and cell culture samples) were analyzed using HPLC.

[0282] Buffer Preparation. The manufacturer's buffers were prepared according to the instructions provided with the dedicated buffer kit. The binding / washing buffer was prepared by diluting 5 mL of Tris-buffered saline (TBS buffer, containing 0.5 M Tris and 1.5 M NaCl) ten times with 45 mL of water. According to the manual, the recommended elution buffer (2.5% acetic acid) does not require dilution. A recommended neutralization buffer (1 M Tris-HCl) is not included in the buffer kit. It is prepared by diluting 2 M Tris base with water at a 1:1 ratio.

[0283] Internal buffers used in large-scale downstream preparation processes were tested and compared with manufacturer-supplied buffers (see Tables 1 and 2). A summary of the original and improved protein purification methods for dupilumab is shown in Table 1. A summary of the original and improved purification methods for aflibercept is shown in Table 2.

[0284] Table 1. Summary of buffer solutions, sample volumes, and centrifuge settings used throughout the dupilumab purification optimization process.

[0285]

[0286]

[0287] Table 2. Summary of buffer solutions, sample volumes, and centrifuge settings used in the optimized aflibercept purification process.

[0288]

[0289] Storage solution removal. First, resuspend the precipitate in the ProA column by inverting it. Then, open the bottom cap and place the column in a 2 mL centrifuge tube. The column manual recommends centrifuging at 70-100 RCF (relative centrifugal force) for 30 seconds per step; however, the time was adjusted to 1 minute, as this is the shortest possible centrifugation time on the available instrument. The lower limit of the RCF range (70 RCF) was chosen due to the relatively long centrifugation time. This resulted in incomplete transfer of solution from the column to the centrifuge tube. Centrifuge the column again at 100 RCF for 1 minute, and from then on, use these settings for each centrifugation step. Discard the runoff and place the column in a new centrifuge tube.

[0290] Equilibrate. Add 600 µL of binding / washing buffer to each column and resuspend the culture medium by inverting the column. Centrifuge the column, discard the runoff, and place the column in a new tube.

[0291] Antibody binding. The manufacturer's manual specifies that a maximum of 600 µL of antibody sample can be added to the column. The initial protocol was optimized to avoid column overload by reducing the volume of cell culture sample added to the column from 400 µL to 76 µL using 524 µL of binding / wash buffer, achieving a column loading of approximately 600 µg. Subsequently, the protocol was modified by determining the sample volume based on the theoretical volume loading calculated from linear and time-process studies, and adding binding buffer to the sample to achieve a total volume of 600 µL (Table 3). The column was stirred approximately every 20 seconds while incubating for 4 minutes, centrifuged, and transferred to new tubes. The runoff and eluent were stored for testing.

[0292] Table 3: Examples of theoretical column loadings, and the volumes of antibody and binding buffer determined based on theoretical column loadings.

[0293]

[0294] Washing 1 and 2. Wash the column twice by adding binding / washing buffer, followed by centrifugation, and transfer the column to a new centrifuge tube. To purify aflibercept, perform a second wash by adding 600 µL of 20 mM sodium acetate (see Table 2). Reserve the flow-through for testing or discard it.

[0295] Elutions 1 and 2. The elution steps were initially performed according to the manufacturer's instructions. 400 µL of elution buffer was added to the column and agitated to resuspend the culture medium. 30 µL of neutralization buffer was added to each centrifuge tube. The column was centrifuged, and the eluent was reserved for testing. Although the manufacturer's manual does not specify a second addition of elution buffer, it describes performing two elution steps. Therefore, the elution steps were performed twice. Further optimization of the dupilumab and aflibercept elution steps will be described further below.

[0296] Column Cleaning. The manufacturer does not provide instructions for column maintenance. Therefore, a column cleaning procedure was developed in which 600 µL of water and 30 µL of neutralization buffer are added to each column to neutralize any remaining elution buffer. The column contents are then agitated and centrifuged, and the flow is discarded. Then, 600 µL of a stock solution consisting of 20% ethanol is added to each column. The columns are then capped tightly and the contents are resuspended by inversion before storage at 4°C.

[0297] Titration assay. Cell culture samples and samples purified using the protocol described above were tested in six replicates using a titer assay, with absorbance measured at 280 nm to determine protein yield. Optionally, samples were filtered under vacuum using a Samplicity® filtration system prior to analysis using the titer assay.

[0298] Protein loss investigation. To optimize yield, the previously used column was studied to determine if any protein remained inside. Storage solution removal, equilibration, and two elution steps were performed, and the flow-through was tested with titer determination.

[0299] Elution and Neutralization Optimization. To purify dupilumab, a series of litmus paper tests were performed to find the optimal combination of elution and neutralization buffers. 30 µL, 60 µL, 100 µL, and 200 µL of 2 M Tris base were added to four tubes containing 400 µL of 2.5% acetic acid. Similarly, the pH of other buffers, 0.1 M glycine, and mixtures of different volumes of 1 M and 2 M Tris bases present in the column manufacturer's buffer kit were tested using litmus paper to find the optimal combination. The dupilumab purification protocol was then repeated in duplicate using either 400 µL of 0.1 M glycine or 400 µL of 0.24% acetic acid (40 mM acetic acid) as the elution buffer and 5 µL of 2 M Tris base as the neutralization buffer, with the first and / or second elution buffers reserved for titer testing. 5 µL of 2 M Tris base was used in the column cleaning step.

[0300] The aflibercept purification protocol was optimized to include 400 µL of 0.12% acetic acid (20 mM acetic acid) as the elution buffer and 30 µL of 2 M Tris base as the neutralization buffer. The first elution buffer was used for testing. 30 µL of 2 M Tris base was used in the column cleaning step.

[0301] Yield optimization. Experiments were repeated using the optimized combination of elution and neutralization buffers. The purified samples to be tested were collected into pre-weighed centrifuge tubes and weighed individually. The sample density was assumed to be 1 µg / µL, and the accuracy of calculations was improved by using actual sample weights instead of theoretical volumes.

[0302] Repeatability check. Based on the improved protocol, repeat the experiment using different centrifuges. Use precise eluent weights again to improve calculation accuracy.

[0303] Total protein yield was measured using UV-Vis spectroscopy. Protein loss outside of the binding and washing steps was investigated using UV-Vis spectroscopy to ensure consistency with established protocols. The runoff samples obtained during the repeatability check run described above were analyzed using a commercially available UV-Vis spectrophotometer. Concentrations were measured at 280 nm using pre-programmed extinction coefficient characteristics of the proteins. Binding-elution samples were excluded from the calculations due to their light brown color, which could lead to inaccurate measurements.

[0304] Linearity study of dupilumab column loading. The range of column loadings that could produce consistent and reliable step yields was evaluated by purifying different volumes of cell culture samples of known concentrations of dupilumab and determining the amount of purified protein by UV-Vis spectroscopy.

[0305] The theoretical column loading value was the target value selected for the first run of the linear study (Table 4). The volume of cell culture to be added was calculated based on an estimated cell culture protein concentration of 8.0 µg / µL. Each elution was weighed in a pre-weighed tube to obtain an accurate volume (assuming a density of 1 µg / µL), and the protein concentration was measured using a UV-Vis spectrophotometer.

[0306] Table 4. Column loading and buffer volume values ​​used in the first run of the linear study.

[0307]

[0308]

[0309] The second linearization experiment was repeated, this time using a different blank column (Table 5). Except for the antibody binding step, the antibody purification procedure was performed using a fresh column, with 600 µL of binding / wash buffer added instead of the cell culture sample diluted with buffer in the antibody binding step.

[0310] Table 5. Column load values ​​used in the second run of the linear study.

[0311]

[0312] A comparison of a small-scale purification method for dupilumab with a pre-established large-scale purification method. Dupilumab cell culture samples were purified in three replicates using a modified protocol (76 µL sample / column), with eluents 1 and 2 from each column combined. Aggregation of the eluent pool and dupilumab samples from the same cell culture batch purified using a pre-established large-scale Protein A method was tested using size exclusion chromatography (SEC), and charge variant distribution was analyzed using imaging capillary isoelectric focusing electrophoresis (iCIEF) according to standard operating procedures. The purification protocol was then repeated on a separate day to replicate the aggregation test.

[0313] Aggregation optimization of dupilumab purification method. The dupilumab purification protocol was further optimized to reduce aggregation by improving the elution buffer, centrifuge settings, and neutralization steps. All purified samples were then analyzed using size exclusion chromatography, as detailed below.

[0314] To further optimize the elution buffer, aggregate levels in samples eluted with 2.5% acetic acid were compared with those eluted with 0.1 M glycine to examine differences between the two elution buffers from the manufacturer's buffer kit. For each column, the eluates obtained from the two elution steps were combined.

[0315] Different columns were used to perform the same purification protocol, and the binding / washing and elution buffers were the same as those used in established large-scale methods. Both purification protocols omitted a neutralization step, which was determined to be unnecessary for aggregation assays because the sample does not need to be neutral for size exclusion chromatography.

[0316] To further optimize the centrifugation process, a purification protocol was performed using buffers from a buffer kit, including 0.1 M glycine as the elution buffer. Centrifugation steps were performed at 70 RCF, with the instrument manually stopped 30 seconds after each step. Alternatively, a purification protocol using internal buffers was performed using three columns, with the centrifuge settings reduced to 70 RCF and 30 seconds. The neutralization step was omitted in both methods.

[0317] To further optimize the neutralization buffer, the purification protocol was performed in triplicate using the internal buffer, and the centrifuge was set to 100 RCF for 1 minute. 10 μL or 30 μL of 2 M Tris base neutralization buffer (volume determined by litmus test) was added to each eluent, and eluent 1 and eluent 2 were combined, or eluent 1 was collected for each copy.

[0318] The above experiments were repeated on another day using different cell culture samples and different batches of binding / washing buffer, elution buffer, and neutralization buffer. Samples purified from the same cell cultures using established large-scale methods were also analyzed by size exclusion chromatography for comparison. Remaining samples were aliquoted and frozen at -80°C for further testing.

[0319] Charge variant analysis. One aliquot of each previously frozen sample was thawed, and its concentration was determined using UV-Vis spectroscopy. The spin-column purified sample and the large-scale purified sample were aliquoted into two vials, and half of each sample was dialyzed. The two large-scale purified samples were diluted with water to 1.0 mg / mL, and the spin-column purified sample was concentrated to 1.0 mg / mL using a commercially available centrifugal filter. All four samples were then analyzed using an iCIEF instrument.

[0320] Fracture Analysis. One aliquot of each previously frozen sample was thawed and diluted with water to 0.3 mg / mL. Protein fragmentation was analyzed using capillary electrophoresis with sodium dodecyl sulfate (CE-SDS) on a microchip electrophoresis (MCE) system, following valid standard operating procedures. Samples were treated with sodium dodecyl sulfate (SDS) under both non-reducing and reducing conditions. Reducing conditions were achieved by adding a commercially available reducing agent.

[0321] N-glycan profiling analysis. One aliquot of each previously frozen sample was thawed. The mass-purified sample was diluted with water to 2 mg / mL, and the column-purified sample was concentrated to 1.8 mg / mL using a centrifugal filter. The N-glycan profiles (G0F, G1F(1-6), G1F(1-3), and G2F) were then analyzed using a commercially available N-glycan kit according to valid standard operating procedures and hydrophilic interaction chromatography (HILIC).

[0322] Yield study of the optimized dupilumab purification method. Freshly thawed dupilumab cell culture harvest sample was purified in triplicate using the optimized method and placed in pre-weighed tubes. 81 µL of sample was added to each column to achieve a column loading of 600 µg, with a known cell culture concentration of 7.395 mg / mL at harvest. Additionally, a blank sample was prepared by performing the purification protocol with one variation: 600 µL of binding / wash buffer was added during the antibody binding step without adding the cell culture sample.

[0323] For any of the four columns, the eluents obtained after elutions 1 and 2 were not pooled. The precise weights of the eluents were recorded, and the concentration of each eluent was measured using UV-Vis spectroscopy. For the corresponding elution step, the concentration of the blank eluent was subtracted from the purified antibody sample. These values ​​were then used to calculate the theoretical amount of protein transferred, and the eluent concentrations and precise weights were used to calculate the obtained protein mass and yield.

[0324] The titers of the purified antibody and cell culture samples were determined. The obtained protein mass and yield were calculated in the same manner as described above.

[0325] Linearity study of column loading for optimized purification methods of dupilumab and aflibercept. The optimal column loading range for the optimized purification methods was investigated using both chromatographic and spectroscopic techniques. For the dupilumab purification method, selected column loading values ​​were 200 µg, 400 µg, 600 µg, 800 µg, and 1000 µg (Table 6). Additionally, a blank eluent with a column loading of 0 µg was prepared for background absorbance spectroscopic measurements. Based on historical data, assuming a cell culture concentration of 6.8 mg / mL, the volume of cell culture required to achieve the selected column loading was calculated.

[0326] Table 6. Column loading and buffer volume values ​​used in the linearity study of column loading for the optimized purification method of dupilumab.

[0327]

[0328] For the purification of aflibercept, column loading values ​​of 100.5 µg, 201 µg, 402 µg, 603 µg, and 804 µg were evaluated (Table 7). The high column loading value (804 µg) was selected based on the maximum loading typically achieved when harvesting material using the maximum permissible volume (600 µL) at typical titers. Lower column loading values ​​were selected to cover the low and medium column loading range.

[0329] Table 7. Column loading and buffer volume values ​​used in the linearity study of column loading for the optimized purification method of aflibercept.

[0330]

[0331] Except for a blank column, optimized purification protocols were performed in triplicate for different column loadings. The eluents obtained in elution steps 1 and / or 2 were collected in separate, pre-weighed tubes, and their precise weights were recorded. The concentration of each eluent was measured using UV-Vis spectroscopy. The cell culture samples and each eluent, except for the two blank samples, were subsequently analyzed by titer determination.

[0332] Comparative study of spin column and small-scale purification methods for dupilumab. Culture samples were collected at different time points. A column loading of 600 µg was used for spin column purification of the harvested material in a small bioreactor. The harvested samples were purified in duplicate, with 5 µL of 2 M Tris base used as a neutralization buffer. Details of the column loadings used for comparison are shown in Table 8.

[0333] Table 8. Column loading and buffer volume values ​​used in the comparative study of dupilumab.

[0334]

[0335] Time-course sample studies of aflibercept. Culture samples were collected at different time points. Each sample was purified using a theoretical column loading of 600 µg. The column loading was based on the titer of each sample, which varied across the sample set. For low-titer samples, the maximum permissible loading volume (600 µL) was used. For higher-titer samples, the volume was calculated based on a column loading of 600 µg. Purification was performed in triplicate at time points A, B, C, and D, and once at time points E, F, G, H, I, and J. Details of the column loading for the samples are shown in Table 9.

[0336] Table 9. Column loading and buffer volume values ​​used in time-process sample studies of aflibercept.

[0337]

[0338] Example 1. Optimization of protein yield of dupilumab

[0339] Purifying 400 µL of dupilumab cell culture using the initial protocol recommended by the protein A column manufacturer (centrifugation time increased from 30 seconds to 1 minute) yielded an average protein recovery of 65.3%, likely due to column overload. A volume of 400 µL (later measured at 7.903 µg / µL) was chosen without knowing the concentration of the cell culture. The resulting protein loading in the column was approximately 3.16 g, about three times the binding capacity listed by the manufacturer.

[0340] Two additional elution steps performed on the same column did not produce significant amounts of eluted protein, indicating that the average 34.7% protein loss was not retained in the column, but rather due to unavailable binding sites that prevented binding to protein A. The protein is likely lost in the flow-through during both the binding and washing steps.

[0341] Reducing the volume of cell culture used for purification to 76 µL to achieve a column loading of 600 µg resulted in an average increase in protein yield of 73%. This value was considered undesirable, so the elution buffer and neutralization buffer were subsequently optimized.

[0342] Litmus paper was used to measure the approximate pH values ​​of the elution buffer and neutralization buffer. The pH of the 2.5% acetic acid elution buffer was found to be approximately 2, lower than the optimal pH (pH 2.5–4) for eluting antibodies from protein A (Chahar et al.; Urh et al.). Furthermore, the neutralization buffer recommended by the column manufacturer was too weak to neutralize the elution buffer to the desired pH of approximately 7, even when 200 µL of the maximum possible volume of 2 M Tris base (twice the recommended molar concentration) was added to 400 µL of elution buffer. Based on this observation, another elution buffer, glycine, from the column manufacturer's buffer kit was tested using a similar litmus test series. Diluting the concentrated 1 M glycine to the recommended 0.1 M, the glycine buffer itself was found to have a pH of approximately 3. This value is consistent with the pH given in the buffer kit manual (pH 2.9). A summary of the tested buffer combinations and measured pH values ​​is shown in Table 10.

[0343] Table 10. Approximate pH measurements of the elution buffer and Tris base mixture.

[0344]

[0345]

[0346] The optimal buffer combination was found to be 400 µL of 0.1 M glycine elution buffer and 10 µL of 1 M Tris base or 5 µL of 2 M Tris base neutralization buffer. Due to the availability of 2 M Tris base, rather than 1 M Tris base which requires an additional dilution step, 5 µL of 2 M Tris base was chosen for future experiments. The pH of 2 M Tris base is approximately 10. When purifying cell culture samples using this improved buffer combination, protein yields increased to an average of 86.2%. The experiment was rerun with six replicates, where accuracy was improved by using precise volumes (weighing the elution buffer and assuming a density of 1 µg / µL), resulting in an average yield of 85.7%.

[0347] The reproducibility of the improved method was then checked by running experiments on different days and using different centrifuges, resulting in a yield of 86.6% as determined by titer assay. Additionally, the protein concentration of the runoff collected after each centrifugation step was measured using UV-Vis spectroscopy (consistent with large-scale methods), and these values ​​were used to calculate an average protein yield of 91.4%, comparable to yields obtained from large-scale purification using this antibody.

[0348] Example 2. Linearity Study of Dupilumab Column Loading

[0349] Linearity studies were conducted to determine the ability of the improved method to consistently and reliably produce high protein yields using a variety of starting protein concentrations. In the first run of the linearity studies, column loading ranges from 20 µg to 1200 µg were investigated. The mass of purified protein was calculated based on the measured eluent concentration and volume (Table 11), and then the protein mass was calculated based on the measured eluent weight, assuming a density of 1 µg / µL. In Table 12, the incoming protein or actual column loading was calculated based on the volume of the added cell culture sample and the actual cell culture concentration (7.903 µg / µL) determined by titration. The protein masses obtained in the two elution steps were combined, and the yield for each column loading was calculated.

[0350] Table 11. Measured concentrations and volumes of all eluents, and calculated purified protein masses obtained in the first run of the linearity study of column loading.

[0351]

[0352] Table 12. Protein yields obtained in the first run of the linearity study at column loading, calculated based on actual column loading and total purified protein mass.

[0353]

[0354] The protein values ​​obtained from most replicates were higher than those from the input, indicating that other proteins or components of the sample (possibly glycine buffer) were interfering with the spectroscopic measurements. Therefore, the linearity study was repeated by adding a blank sample to establish baseline protein concentrations. The blank was subtracted from each concentration value, and the protein mass was calculated using precise volume (Table 13). The actual column loading values ​​(Table 14) were calculated based on the actual cell culture sample concentration (7.960 µg / µL) measured by titration on the same day.

[0355] Table 13. Measured concentrations and volumes of all eluents, and calculated purified protein masses obtained in the second run of the linearity study of column loading.

[0356]

[0357] Table 14. Protein yield obtained in the second run of the linearity study on column loading, calculated based on actual column loading and total purified protein mass.

[0358]

[0359]

[0360] likeFigure 7A As shown, for each column loading, the protein obtained by plotting the input protein yields R. 2 The curve with a value of 0.9997 shows a highly linear relationship between the two. Subsequently, as... Figure 7B As shown, protein yield % was plotted against incoming protein values, and the column loading range of 200–1200 µg showed that the yield % remained consistently around 94.19% (0.76% RSD), while lower column loadings resulted in lower yield % and higher column loadings resulted in an unknown yield % . The high yield (94.34%) at a 1200 µg column loading was unexpected, as the manufacturer's theoretical column capacity is approximately 1 mg.

[0361] Example 3. Charge variants and aggregation assays for dupilumab.

[0362] The product quality attributes of dupilumab samples purified in triplicate according to the improved protocol and dupilumab samples purified from the same cell cultures by a large-scale method (comparative samples) were analyzed to compare their product quality attributes.

[0363] Size exclusion chromatography (SUC) was used to analyze the size variation and aggregation of dupilumab samples (Table 15). The main peak represents the monomeric form of dupilumab. The HMW peak represents dupilumab with a relatively high molecular weight (e.g., aggregates), while the LMW peak represents dupilumab with a relatively low molecular weight (e.g., fragments). Peak area is proportional to the amount of each species in the sample tested. Although the analysis yielded LMW peak area results, this method is intended for quantification of aggregates, and the quantification of fragments is considered unreliable and for reference only.

[0364] Table 15. Size exclusion chromatography results of samples purified on a large scale using triplicate elution cells and comparative compounds.

[0365]

[0366] The measured average agglomeration in the rotating column elution cell was 16.596%, approximately 1.9 times that of the large-scale purified (comparative) sample. This result was considered undesirable, and the purification protocol was repeated to prepare new samples and the agglomeration was tested again (Table 16).

[0367] Table 16. Size exclusion chromatography results of triplicate elution cells and comparative samples after large-scale purification in repeated experiments.

[0368]

[0369] The average high molecular weight peak area % of the replicates from the three elution cells was found to be 16.646%, confirming the previous aggregation results, which were approximately 1.9 times higher than those of the samples purified on a large scale. Samples purified using this method were not subjected to subsequent product quality testing. Instead, the reasons for the high aggregation rate were investigated to further optimize the method and reduce aggregate levels in samples purified on a small scale.

[0370] Aggregation optimization. Size exclusion chromatography was used to analyze dupilumab samples purified with different buffers and centrifuge settings. Internal buffers for large-scale downstream preparation processes were tested, including 20 mM sodium phosphate (pH 7.10–7.30) as binding / washing buffer and 0.24% or 40 mM acetic acid (pH 2.80–3.20) as elution buffer. Initially, the neutralization step was omitted as it was considered unnecessary for preparing samples for SEC testing. The results of the optimization experiments are summarized in Table 17 below.

[0371] Table 17. Summary of SEC results for purified cell culture samples using different buffer solutions and centrifuge settings.

[0372]

[0373] The results showed that, compared with samples eluted with 0.1 M glycine, returning to the initially used 2.5% acetic acid elution buffer almost doubled aggregation to over 30%, and increased fragmentation. When using the internal buffer, aggregation decreased to 6.4% and fragmentation decreased to 1.4%, the former comparable to samples after large-scale purification, but the latter undesirably excessive.

[0374] For samples eluted with 0.1 M glycine, reducing centrifugation time and relative centrifugal force (RCF) decreased aggregation from 16.6% to 13.4% and increased fragmentation from 0.4–0.5% to 0.6%. For samples purified with internal buffer, fragmentation decreased slightly from 1.4% to 1.1–1.3%, while aggregation increased from 6.4% to 6.6–7.3%. Therefore, it was determined that further optimization beyond centrifuge settings was desirable for improving aggregation and fragmentation.

[0375] Subsequently, it was hypothesized that the lack of a neutralization step might be the cause of the low purity. Therefore, the dopamine cell culture samples were purified again using internal buffer, with 10 µL of 2 M Tris base added to each elution buffer. The resulting aggregation (6.9–7.3%) and fragmentation (0.5–0.6%) were both comparable to the SEC results of the previously large-scale purified samples.

[0376] To compare the product quality attributes of samples purified from the same cell culture samples, the experiment was repeated using an internal buffer and 10 µL of 2 M Tris base as a neutralization buffer. On the same day, the resulting samples and those purified using the large-scale method were analyzed using size exclusion chromatography. The results summarized in Table 18 indicate that aggregation and fragmentation in the two samples were comparable, with the novel method exhibiting higher purity.

[0377] Table 18. Comparison of SEC results between samples purified using small-scale (elution cell) and large-scale (comparative) methods.

[0378]

[0379] Subsequently, the charge variant distributions of the small-scale purified and large-scale purified samples were analyzed using iCIEF. Two samples were prepared in two versions (i.e., pure and dialyzed) to investigate the effect of percolation on the charge variant distribution of dupilumab. The results (Table 19) show comparable peak area % values ​​between the dialyzed and pure samples, with the largest difference observed in the acidic variant (region 1) of the large-scale purified sample being approximately 3.5%. Therefore, it was decided to omit the additional percolation step in future analyses.

[0380] The results obtained from the spin-column purified sample and the large-scale purified sample were also comparable, with the values ​​in Zone 1 of the spin-column purified sample falling within approximately 0.28% (pure) or approximately 3.2% (dialyzed) of the large-scale purified sample, and the values ​​in Zone 2 falling within approximately 2.7% (pure) or approximately 1.3% (dialyzed) of the large-scale purified sample. The greatest difference could be seen in the basic charge variant (Zone 3), with a difference of approximately 16.1% (pure) and approximately 19.1% (dialyzed).

[0381] Table 19. Comparison of iCIEF results between samples purified using the new small-scale (elution cell) and established large-scale (comparative) methods (both dialyzed and pure samples).

[0382]

[0383] Example 4. Fracture and Glycan Profile Testing of Dupilumab

[0384] Aliquots of the plinomacloba antibody sample described in Example 3 were analyzed by CE-SDS under both reducing and non-reducing conditions. The samples were treated with sodium dodecyl sulfate (SDS) to mask the native charge of the protein, and reducing conditions were achieved by adding a reducing agent to break interchain disulfide bonds, leading to further cleavage and the appearance of light chain (LC) and heavy chain (HC) peaks on the electrophoresis pattern. The results are summarized in Table 20 below (LMW = low molecular weight; NGMP = non-glycosylated main peak; MP = main peak; HMW = high molecular weight).

[0385] Table 20. Comparison of CE-SDS results between samples purified using optimized small-scale methods and established large-scale methods.

[0386]

[0387] Analysis of non-reduced samples provides information about product purity and can be used to assess breakage caused by host cell proteases, as well as disulfide bond reduction and re-oxidation. The area percentage (%) of the LMW and HMW peaks indicates the amount of low molecular weight species (e.g., fragments) and high molecular weight species (e.g., aggregates) present in the sample, providing orthogonal information about aggregation in addition to data obtained using size exclusion chromatography. The main peak represents intact antibody species (Dadouch et al., Separations, 2021, 8:4). Reducing conditions allow for analysis of heavy chain glycosylation occupancy by quantifying non-glycosylated heavy chains, as well as analysis of the relative amounts of heavy and light chains (Wagner et al., Journal of Pharmaceutical and Biomedical Analysis, 2020, 184:113166).

[0388] For the purpose of assessing fragmentation, the LMW peak under non-reducing conditions was used. The spin-column purified sample showed a fragmentation rate of 6.3826%, which is approximately 110% of the fragmentation rate of the mass-produced sample (5.7922%). Overall, the parameters are comparable, with the main peak area percentage differing by only 0.87%. Significant differences were observed in the non-reducing HMW peak, with the area percentage of the spin-column purified sample being less than half that of the mass-produced sample. Although this method was not designed for assessing aggregation, these results are consistent with SEC results, indicating a significantly lower aggregate content in the spin-column sample.

[0389] like Figure 8As shown in the paper, hydrophilic interaction chromatography (HILIC) using an N-glycan kit was also used to analyze spin-column purified dupilumab samples and comparative-scale purified dupilumab samples to evaluate four N-glycan structures: G0F((Fuc)1(GlcNAc)2(Man)3 + (GlcNAc)2), G1F(1–6)((Fuc)1(GlcNAc)2(Man)3 + (GlcNAc)2(Gal)1), G1F(1–3)(((Fuc)1(GlcNAc)2(Man)3 + (GlcNAc)2(Gal)1) and G2F((Fuc)1(GlcNAc)2(Man)3 + (GlcNAc)2(Gal)2)) and the combined fucoidylated glycan (the sum of the four N-glycans). The results are summarized in Table 21.

[0390] Table 21. Comparison of N-glycan analysis results between antibody samples purified using the new small-scale method and established large-scale methods.

[0391]

[0392] The glycan profiles were found to be very similar, with peak area percentages between samples ranging from 0.33% to 1.35% for the four N-glycans, and within 0.47% for the combined fucoidylated glycans. These final product quality attributes confirm that the product quality profile of dupilumab produced by the developed small-scale purification method is comparable to that of samples purified using established large-scale methods.

[0393] Example 5. Yield of the optimized dupilumab purification method

[0394] The yield percentage for triplicate purification using the optimized, developed rotating column method was calculated based on two concentration measurement methods (UV-Vis spectroscopy and titer determination). Both analyses were performed on the same day using the same eluent sample.

[0395] In the former technique, a blank sample prepared by performing a purification protocol without adding any cell culture material to the column is used as the background absorbance and subtracted from the measured concentration of each sample in the corresponding elution step. This is done to improve the accuracy of the antibody concentration of interest by removing the influence of potential impurities and other light-absorbing sample components during spectroscopic measurements. Assuming a density of 1 mg / mL, the accurate sample volume is calculated based on the weight of the eluent. The mass of the obtained protein is then calculated using the concentration and volume values. The results are summarized in Table 22 below.

[0396] Table 22. Concentration of the rotating column eluent determined by UV-Vis spectroscopy and the calculated protein mass.

[0397]

[0398] The mass of protein obtained in the two elution steps for each column was then summed. The amount of incoming protein was calculated based on the volume of cell culture added to each column (81 µL) and the concentration (6.362 mg / mL) determined by titer assay on the day of purification. The percentage yield of the purification step for each column was then calculated using the incoming protein and the obtained protein values, averaging 94.50%, as shown in Table 23.

[0399] Table 23. Percentage yield of each run in a triplicate run of the developed protein purification method, based on the protein mass obtained from the concentration obtained by UV-Vis spectroscopy.

[0400]

[0401] No blank sample is required for titer determination. The obtained concentration value and accurate sample volume are used to calculate the mass of the obtained protein (Table 24).

[0402] Table 24. Concentration of the rotating column eluent determined by titration and the calculated protein mass.

[0403]

[0404] Similar to the spectral results, the protein mass of each copy was summed, and the previously calculated input protein value was used to determine the yield % (Table 25).

[0405] Table 25. Percentage yield of each run in triplicate of the developed protein purification method, based on the protein mass obtained from the concentration determined by titration.

[0406]

[0407] The two average yield percentages (94.50% and 93.28%) calculated using spectroscopic and chromatographic techniques are comparable to each other and to the step yields obtained using established large-scale purification methods. This confirms that the novel small-scale method provides delivery of purified dupilumab in similar yields and PQ profiles.

[0408] Example 6. Column loading of the optimized dupilumab purification method

[0409] The percentage of protein yield for dupilumab purification was calculated using triplicate aliquots of different theoretical column loadings (200 µg, 400 µg, 600 µg, 800 µg, and 1000 µg) obtained by UV-Vis spectroscopy and titration assays to assess the optimal column loading range. Similar to the yield study described in Example 5, a blank sample was used to measure the background absorbance of the spectroscopic measurements, and the precise volume of the eluent was determined by measuring the precise weight of the eluent, assuming a density of 1 mg / mL. See the results summarized in Table 26.

[0410] Table 26. Concentration of the rotating column eluent determined by UV-Vis spectroscopy and the calculated protein mass.

[0411]

[0412]

[0413] For each replicate, the protein mass obtained in each elution step was summed. The amount of introduced protein was calculated based on the volume of cell culture added to each column and its concentration (6.594 mg / mL) determined by titer assay. Subsequently, the introduced protein and the obtained protein values ​​were used to calculate the yield % for each column, as shown in Table 27 and Figure 9A This is a summary.

[0414] Table 27. Percentage yield of each run in a triplicate run of the developed protein purification method, based on the protein mass obtained from the concentration obtained by UV-Vis spectroscopy.

[0415]

[0416] The concentration of the eluent was also measured by titration, and the obtained protein mass was calculated based on the precise weight of the eluent (Table 28). See Table 29 and... Figure 9B As shown, the proteins obtained from the two elution steps of each column are summed, and the yield is calculated using the previously determined input protein values.

[0417] Table 28. Concentration of the rotating column eluent determined by titration and the calculated protein mass.

[0418]

[0419] Table 29. Percentage yield of each run in a triplicate run of the developed protein purification method, based on the protein mass obtained from the concentration determined by titration.

[0420]

[0421] Both spectroscopic and chromatographic analysis techniques indicate that the optimal column loading range for an average yield > 90% is 400–800 µg. A comparison of the results obtained by the two methods can be made... Figure 9C As seen in the image, the 600 µg column achieved the highest yield (93.72% by UV-Vis spectroscopy and 97.26% by titration), although the average yield of the column loading was lower than that of replica 2 (87.12% by UV-Vis spectroscopy and 87.69% by titration), likely due to sample loss prior to weighing the eluent, eluent spillage during the purification procedure or sample transport. However, it is worth noting that the purification method was developed for PQ assay purposes, so further maximizing the yield was not a priority.

[0422] Example 7. Comparison of novel small-scale methods and established large-scale methods for purifying dupilumab.

[0423] The experimental results disclosed in Examples 1-6 demonstrate that dupilumab purified from upstream CHO cell culture material using a small-scale method exhibits comparable PQ profiles and step yields compared to dupilumab samples purified using the established large-scale method. Both techniques utilize protein A affinity chromatography, with the novel method employing a small-scale rotating column and the established large-scale method using a conventional liquid chromatography system.

[0424] Cell culture samples from the same batch purified using two methods on the same day were analyzed based on their key product quality attributes: aggregation, fragmentation, charge variants, and glycan profiles. The results are summarized in... Figure 10 As shown in the image.

[0425] It was found that, except for the G2F glycoform which was within 1.4%, the product quality of the spin-column purified sample was within 1% of that of the large-scale purified sample in terms of total combined fucoidylated glycans and all N-glycan structures (G0F, G1F(1-6), G1F(1-3)). The results indicate that the glycan profiles of the samples purified using these two methods are comparable.

[0426] The charge variant distributions of the spin-column purified samples were found to be comparable, with acidic variants (Region 1) within 0.3% and “neutral” variants (variants with similar pI values, Region 2) within 2.8% of the samples purified by large-scale methods. The peak area of ​​the basic variant (Region 3) was found to be approximately 84% relative to the established large-scale method; however, the relatively large difference was not unexpected. The lower abundance of the basic variant of this particular antibody resulted in a smaller peak area and higher assay variability. Similar variability was observed in standard injections, and therefore, unlike the other two regions, the standard operating procedure for iCIEF analysis did not require a % RSD for the values ​​in Region 3. Furthermore, the percentage difference between normalized data was relatively large because the actual area % values ​​compared to the normalized data were much smaller than those of the other two regions, while the numerical difference between the actual area % values ​​in Region 3 was not significant.

[0427] The spin-column purified sample exhibited slightly lower fragmentation (90.75%) and aggregation (76.77%) results, meaning the purity of the spin-column purified sample differed slightly from that of the large-scale purified sample. This was a surprising result, as the multiple centrifugation steps of the small-scale method were expected to increase aggregation rates due to mechanical stress. However, the results showed less fragmentation and aggregation occurring during its own purification process, which may be due to differences in exposure to low pH conditions, or the small-scale method's ability to remove those impurities, which may depend on the type of protein A resin used in the column. In the small-scale method, the two elution steps totaled less than 5 minutes, and the eluent was immediately neutralized with neutralization buffer. In the large-scale purification method, antibody elution from the column required more time, and the eluent was not neutralized because the virus inactivation step required acidic conditions, which was performed after affinity chromatography in the downstream purification process. The amplifying effect of low pH conditions on aggregation and fragmentation rates has been described above (Jin et al.; Hu et al.). Subsequent purification steps in large-scale purification further refine the resulting products, removing additional impurities generated during downstream preparation processes. In the future, this hypothesis can be investigated by performing small-scale purification protocols as usual, except that the amount of time the sample is exposed to each buffer is consistent with the timeline of the established large-scale method.

[0428] Example 8. Comparison of spin column and small-scale purification methods for dupilumab.

[0429] Samples harvested from different small bioreactors with similar titers of dupilumab were purified using a spin column method. Following purification, protein concentration, aggregation, fragmentation, charge variants, and glycan profiles of the samples were analyzed. The aim of this study was to determine whether the optimized spin column method used for purifying dupilumab could be applied to purifying harvested samples obtained from small bioreactors with volumes of only a few hundred milliliters.

[0430] For data analysis purposes, all values ​​obtained from samples purified by spin column were normalized relative to values ​​obtained from samples purified by conventional small-scale purification. Normalized values ​​represent a 100% percentage of complete comparability, and values ​​above or below 100% represent a percentage change relative to values ​​obtained from small-scale purification. Table 8 shows the percentage of comparability of aggregation and charge between samples purified using the new small-scale spin column purification method and established small-scale purification methods.

[0431] The comparability percentage of HMW species between the spin column purified samples and the small-scale purified samples ranged from -38.88% to -20.62%. These results indicate that the spin column purified samples exhibited a lower HMW% than the small-scale purified samples (Table 30 and...). Figure 11 Furthermore, the comparability percentage of total major species between the spin-column purified samples and the small-scale purified samples ranged from +2.41% to +5.30%, indicating that the purity of the spin-column purified samples was comparable to that of the small-scale purified samples. (Table 30 and...) Figure 11 ).

[0432] Table 30. Comparison of CE-SDS and icIEF results between samples purified using the new rotating column and small-scale methods.

[0433]

[0434] The difference in the percentage of HMW species comparability is attributed to the fact that the small-scale samples were virus-inactivated samples. Virus inactivation is a process of exposing the protein A elution buffer to a lower pH environment for the purpose of inactivating the virus. This process may cause protein aggregation and could explain why uninactivated, spin-column purified samples showed lower aggregation levels.

[0435] The charge variant distributions of samples purified by rotating column and samples purified on a small scale were found to differ by region (Table 30 and...). Figure 12 For example, comparable results in the acidic variant (Zone 1) ranged from +3.52% to +9.39%, while comparable results in the dominant species (Zone 2) ranged from -4.46% to +2.24%, and comparable results in the basic variant (Zone 3) ranged from -27.03% to -9.57%. Samples from bioreactors 5 and 6 showed the most similar charge variant distributions across all zones (particularly Zone 3).

[0436] By analyzing the concentrations obtained from each purification, the comparability between small-scale values, the experimental conditions, and incubation times, potential explanations for the poor comparability of bioreactors 4, 7, and 8 for region 3 compared to bioreactors 5 and 6 were assessed. However, the reasons could not be determined. Furthermore, it has been previously observed that the charge variant distribution in region 3 exhibits high variability.

[0437] It must be noted that the comparability results listed above are normalized, which leads to significant variability. However, the variability between actual values ​​is small. Table 31 shows examples of differences in actual values ​​for zone 3, where the percentage difference in charge between samples purified by rotating column and those purified on a small scale ranges from 0.9% for bioreactor 5 to 3.0% for bioreactor 4. However, when these values ​​are normalized, the percentages of charge comparability for zone 3 in bioreactor 5 and bioreactor 4 are 90.43% and 72.97%, respectively, highlighting the difference between actual and normalized values.

[0438] Table 31. Percentage difference in charge values ​​obtained in zone 3 between samples purified by rotating column and samples purified on a small scale.

[0439]

[0440] \

[0441] To assess breakage, low molecular weight species (LMWs) were analyzed in samples purified by spin column under non-reducing conditions and samples purified on a small scale (Wong et al., 2023, Analytical Biochemistry, 666:115073). These LMWs included impurities from products or processes resulting from cellular enzymes, reoxidation, and disulfide bond reduction (Dadouch et al., 2021, Separation, 8(1):4). Samples under reducing conditions were used to determine the number of non-glycosylated heavy chains, LMWs, and high molecular weight species (Dadouch et al.). As shown in Table 32, the comparability percentage of LMW species between samples purified by spin column under non-reducing conditions and samples purified on a small scale across all bioreactors ranged from -21.91% to -10.69%. Under reducing conditions, the comparability percentage of LMW species between samples purified by spin column and samples purified on a small scale ranged from -36.78% to -7.02%.

[0442] Table 32. Fracture analysis of samples purified by rotating column and samples purified on a small scale from different bioreactors.

[0443]

[0444] likeFigure 13 As shown, under both reducing and non-reducing conditions, the spin-column purified samples exhibited comparable purity, NGMP (non-glycosylated main peak) % and NGHC (non-glycosylated heavy chain) % to those purified on a small scale. Interestingly, the spin-column purified samples showed lower LMW % under both reducing and non-reducing conditions. This may be due to the effect of low pH on the samples during virus inactivation, which could lead to increased fragmentation in small-scale purified samples. In fact, studies investigating the effects of different pH and buffer species on monoclonal antibodies have demonstrated that lower pH conditions cause more fragmentation than higher pH conditions (Zheng et al., 2015, AAPSPharmaSciTech, 18(1):42-48). It was found that lower pH unfolds the CH2 domain, leading to increased surface accessibility and fragmentation (Zheng et al.).

[0445] In summary, these results demonstrate that comparable product quality results can be obtained between samples purified by rotating column and samples purified on a small scale, thus indicating that the small-scale purification method for dupilumab yields comparable results when using harvested material from a small bioreactor.

[0446] Example 9. Optimization of aflibercept purification method

[0447] To develop an optimal purification method for aflibercept, aflibercept samples purified using a spin column method and a conventional small-scale column method were analyzed to compare their product quality attributes. For the spin column method, two elution steps were performed using the same elution buffer, with a second elution step performed to maximize yield. Since the same elution buffer was used for both elution steps, eluents 1 and 2 were considered comparable in terms of protein population. For product quality testing purposes, only eluent 1 was analyzed, as combining the two eluents would result in a diluted sample.

[0448] Initially, the product quality attributes of aflibercept purified using a spin column method with buffers provided in the manufacturer's manual were compared with those purified using a small-scale purification method with internal buffers. The spin column was packed with 603 µg of aflibercept harvest material and purified using the manufacturer's buffers. The eluent was then analyzed by size exclusion ultra-high performance liquid chromatography (SE-UPLC) to assess purity and aggregation levels, and charge distribution characteristics were analyzed by imaging capillary isoelectric focusing (iCIEF).

[0449] Table 33 shows the comparable percentages of aggregation and charge between samples obtained from spin-column purification using the manufacturer's buffer and samples obtained from routine small-scale purification using the internal buffer. Differences in aggregation and charge distribution were observed between the spin-column purified and small-scale purified samples. SE-UPLC analysis showed that the total main peak of the spin-column purified sample was reduced by 44.9% compared to the small-scale purified sample. Furthermore, it was found that the HMW species percentage in the small-column purified sample was 2567% higher than that in the small-scale purified sample. This indicates that a significant number of molecules in the spin-column eluent are aggregated, which is undesirable in terms of product quality. Since the technique used for aggregation experiments specifically measures high molecular weight species and the total average peak, low molecular weight results are ignored and therefore considered inaccurate.

[0450] Differences in charge distribution were observed across all three zones: zone 1 showed a 40.75% lower charge distribution after spin-column purification compared to small-scale purification; zone 2 showed a 24.85% higher charge distribution; and zone 3 showed a 9.19% higher charge distribution. Therefore, it is concluded that there are significant differences in product quality properties between samples obtained using conventional small-scale purification methods and spin-column purification methods using the column manufacturer's buffer. This indicates that spin-column purification using the manufacturer's buffer is unsuitable for purifying aflibercept.

[0451] Table 33. Comparison of CE-SDS and icIEF results between samples purified using the spin column method with the manufacturer's buffer and samples purified using the small-scale method with the internal buffer.

[0452]

[0453] Because the results obtained by spin-column purification using the manufacturer's buffer were not ideal compared to established small-scale methods, a spin-column method using an internal buffer was evaluated. Spin-columns were packed with 603 µg of aflibercept harvest material, and the aggregation and purity levels, as well as charge distribution, of the eluent were assessed. Table 34 shows the comparability percentages of aggregation and charge between samples purified from the spin-column using the internal buffer and those obtained from conventional small-scale purification. The results showed that the aggregation and charge of the spin-column purified samples were comparable to those of the small-scale purified samples. The total peak value of the spin-column purified samples was found to be 0.08% higher than that of the small-scale purified samples. Furthermore, the HMW% of the spin-column purified samples was found to be 4.4% lower than that of the small-scale purified samples, indicating lower aggregation in the aflibercept samples obtained using the new spin-column method.

[0454] Regarding charge distribution, region 2, composed of dominant or neutral species, was found to be 3.85% higher in the spin-column purified sample than in the small-scale purified sample, with the difference being only a few percentage points. Furthermore, the comparability percentages of regions 1 and 3 in the spin-column purified sample were 11.99% higher and 12.97% lower than those in the small-scale purified sample. Although the spin-column and small-scale values ​​differed to some extent, the differences were considered negligible, and the values ​​were considered comparable. In summary, these results highlight the suitability of the internal buffer with the spin-column purification method for aflibercept.

[0455] Table 34. Comparison of CE-SDS and icIEF results between samples purified using the spin column method with internal buffer and samples purified using the small-scale method.

[0456]

[0457] like Figure 14 As shown, there is a significant difference in product quality values ​​obtained using the internal buffer compared to the manufacturer's buffer. Similarities exist between values ​​obtained from small-scale purification and spin column purification using the internal buffer. These results demonstrate that the internal buffer is best suited for spin column purification of aflibercept. Therefore, a linearity study was conducted using different column loadings to determine the optimal column loading.

[0458] Example 10. Column loading of the optimized aflibercept purification method.

[0459] The aim of the linearity study was to define the maximum and minimum column loading values ​​that would produce product quality comparable to conventional small-scale purification. For this purpose, a column loading range of 100.5 µg–804 µg was investigated. The loading volume was calculated based on a harvest titer of 1.34 mg / ml and the desired column loading protein. The step yield was calculated by dividing the total pool protein by the total protein loading. The total pool protein was calculated using the elution pool volume, determined by weighing the elution tubes before and after elution, and the elution pool concentration, determined by UV-Vis spectroscopy. Table 35 shows the step yield for eluent 1. Furthermore, since three copies of eluent 1 were combined and submitted for testing, the average of the three copies was taken as the step yield.

[0460] Table 35. Average step yield results of eluent 1 (triples) from linear studies.

[0461]

[0462]

[0463] Table 36. Average step yield results of eluent 2 (three-part fraction) from the linear study.

[0464]

[0465] Table 37. Tripartite average step yields from linear studies.

[0466]

[0467]

[0468] As shown in Table 35, the lowest step yield obtained with elution buffer 1 was 66.78% when a total column loading of 804 µg was applied, and the highest step yield was 81.27% when a total column loading of 402 µg was applied. One reason why the step yield of elution buffer 1 was not 100% can be attributed to the limited incubation time required during the elution step, as the elution buffer cannot efficiently release molecules bound to protein A within a short period of time. When total column loadings of 603 µg and 100.5 µg were applied, the step yields of elution buffer 2, as seen in Table 36, ranged from 18.10% to 29.22%, indicating that most of the remaining bound molecules were successfully eluted upon the second addition of elution buffer. The average step yield ranged from 88.77% to 101.24% (Table 37). The lowest average step yield, 88.77%, was obtained when 804 µg of protein was loaded onto the rotating column. One reason for the low step yield may be the high total protein concentration used, because protein A may not have any binding sites available for binding to the molecule, causing the molecule to flow through during the binding step.

[0469] Figure 15A A graphical representation of the average step yield percentage for each column load value is shown. Figure 15B A graphical representation of the average step yield percentage of eluent 1 for each column loading value is shown. (e.g.) Figure 15A It was demonstrated that a total column loading as low as 100.5 µg yielded optimal step yields. Optimal step yields were also obtained for total column loadings of 201 µg, 402 µg, and 603 µg. Since satisfactory step yields were obtained for all tested column loading values, aggregation and fragmentation of eluent 1 were analyzed by SE-UPLC, and charge was analyzed by iCIEF and compared with eluents obtained from established small-scale purification methods.

[0470] As shown in Table 38 and Figure 16As shown, the total peak height of the spin-column purified sample is comparable to that of the small-scale purified sample. A variation in the comparability percentage of HMW species was observed, ranging from -0.35% to +24.31%. For each total protein loading except 402 µg, the spin-column purified sample showed a higher comparability percentage of HMW species compared to the small-scale purified sample. One reason for this variability may be sample manipulation due to freeze-thaw cycles, as the samples were thawed to some extent for product quality testing.

[0471] Table 38. Comparison of CE-SDS and icIEF results between samples purified using the spin column method with internal buffer and samples purified using the small-scale method.

[0472]

[0473] The distribution of charge variants in the spin-column purified samples differed across the three zones, with the dominant species (zone 2) in the spin-column purified samples being most comparable to those in the small-scale purified samples (comparability percentages across different column loading values ​​ranged from +1.78% to +5.92%). The comparability percentages of acidic (zone 1) and basic (zone 3) variants in the spin-column purified samples were higher and lower than those in the small-scale purified samples, respectively (comparability percentages in zone 2 ranged from +11.3% to +17.47%, and in zone 3 from -10.54% to -17.3%). Since the results from icIEF are based on percentage calculations, any variation in one zone will be reflected in one or both zones; therefore, an inverse relationship between zones 1 and 3 is expected.

[0474] like Figure 17 One possible reason for the variability in the charge variant distribution, as demonstrated, is the use of different protein A resins for purification, as different protein A resins may have affected the charge variants observed in eluent 1. Residual charge variants may have already eluted in eluent 2, but were not considered because only the sample from eluent 1 was sent for testing.

[0475] Overall, the developed spin column purification method exhibits high comparability in terms of step yield, aggregation, and charge. Some differences in samples purified by spin column purification compared to small-scale purification methods (particularly differences in charge variant distribution) can be attributed to the type of Protein A resin used in the spin column purification method. The spin column purification method reduces the time required to purify samples from several days to approximately 30 minutes. Furthermore, the low sample requirements facilitate the use of small-scale bioreactors or samples generated in the early stages of cell culture with very low protein concentrations, thus allowing for multi-timepoint sample studies for product quality analysis.

[0476] Example 11. Time-process sample purification of aflibercept using a rotating column.

[0477] Although a higher yield was obtained when a total column loading of 400 µg was loaded onto the rotating column (see Table 35), a column loading of 600 µg was chosen for the time-process study because it provided a higher protein concentration for elution 1 and was comparable to the product quality results under the 400 µg column loading condition.

[0478] Submit eluent 1 for each time point for aggregation and charge testing. For data analysis, product quality results from samples purified by the spin column were normalized to results from harvested material purified using the same spin column. The comparability of product quality results from time point samples to the percentage of harvest is shown in Table 39.

[0479] Table 39. Comparison of CE-SDS and icIEF results between time-point samples purified using the rotating column purification method and harvested materials purified using the rotating column purification method.

[0480]

[0481]

[0482] The percentage of total main peak in the time-point samples was comparable to the height of the harvested material, with the comparability percentage ranging from -0.81% to +1.26% (Table 38 and...). Figure 18 Except for the samples from time point A, HMW% appears to increase over time, with samples from time points I and J being most comparable to the harvested material. This is due to several factors, including increased protein concentration during protein production, physicochemical stress, and interactions with misfolded proteins (Kumari et al., 2023, Journal of Pharmaceutical and Biomedical Analysis, 18(1):126-132). However, the reason for the high HMW% at time point A is unclear.

[0483] like Figure 19 As shown, the charge variant distribution changes from time point A to J, indicating that the charge variant distribution changes during the generation process. In fact, changes in charge variants as a result of the culture process have been observed in studies optimizing CHO cell culture (Weng et al., 2020, *Cytotechnology*, 72(2):259-269). Figure 19As shown, the dominant species (zone 2) appears to decrease over time, while the acidic and basic variants (zones 1 and 3) appear to increase over time, with the acidic and basic variants at time point J being highly comparable to the harvested material. The most pronounced changes are observed starting from time point D, where the values ​​in zone 1 appear to stabilize over time, while the values ​​in zones 2 and 3 change in an inverse relationship.

[0484] Overall, these results demonstrate that multi-timepoint testing of samples can be used to understand the product quality profile of aflibercept during the production process. This provides a unique possibility for studying the impact on product quality during the early stages of the production process, as product quality testing is typically only performed post-harvest. Furthermore, this purification method can be used for rapid purification and product quality assessment should any deviations in product quality occur during the production process and process-related studies are required.

[0485] Example 12. Purification of dupilumab and aflibercept from other cell types

[0486] Alternative purification processes for dupilumab and aflibercept can be characterized by the use of other mammalian cell types, such as HEK 293 cells and young hamster kidney (BHK) cells. Non-mammalian host cell lines, such as the Sf9 insect cell line, can also be used. Different cell lines can be used with minor modifications to the cell culture medium and purification process, consistent with the description herein. Despite these minor modifications to the purification procedure, small-scale spin-column purification methods are expected to produce antibodies with a product quality profile comparable to large-scale purification methods.

[0487] For example, high-glucose growth media, such as Dulbecco's Modified Eagle's Medium (DMEM) or Eagle's Minimum Essential Medium Eagle (MEM) supplemented with 5% to 10% fetal bovine serum (FBS), or serum-free / chemically defined media, such as EX-CELL® 293 serum-free medium or Gibco™ CD BHK-21 production medium, can be used to culture HEK 293 or BHK cells. Supplemented TNM-FH Grace insect medium supplemented with glutamine and 10% FBS can be used to culture Sf9 insect cells. Binding / washing buffers can be modified with different pH and / or ionic strengths to ensure that all excess and unbound components are washed from the protein A spin column. Binding / washing buffers may include buffers such as sodium phosphate, sodium acetate, HEPES, or Tris, as well as additional salts such as sodium chloride or calcium chloride. The pH of the binding / washing buffer can be in the range of pH 6 to 8. Purification using the optimized, developed rotary column method employs cell types (such as HEK 293 cells, BHK cells, and insect cell types), where modified cell culture media and buffers should produce the protein gains listed in Table 40 below:

[0488] Table 40. Concentration of rotating column eluent determined by UV-Vis spectroscopy and calculated protein mass obtained using alternative cell types and cell culture media.

[0489]

[0490] Currently, small-scale laboratory studies of cell culture materials require setting up multiple scaled-down bioreactors to obtain samples at different stages of the upstream production process. This is because purification using established large-scale methods requires relatively large sample volumes, and the number of sampling points from small-scale bioreactors is limited. Therefore, such research currently requires not only a significant workload but also a substantial time investment.

[0491] The newly developed small-scale purification method demonstrated that the antibody product quality profile is comparable to established large-scale methods that can be performed in a shortened timeframe (approximately 30 minutes – typically within a 2-working-day process – compared to several hours), and with smaller volumes of buffer and sample material itself, depending on sample concentration. The greater efficiency of the small-scale method allows for the purification of cell culture samples daily throughout the preparation process, even in small-scale studies. Analyzing daily samples facilitates near real-time or online monitoring of the product quality properties of the resulting antibodies, thereby improving understanding of the process.

[0492] As CHO cells continue to secrete the proteins of interest, the concentration of recombinant proteins in daily cell culture samples increases significantly over time. However, typically only harvested samples with relatively high concentrations (approximately 7 mg / mL, varying between batches) are purified and analyzed. In the experiments disclosed in the examples, daily samples with lower protein concentrations were purified using the developed spin column method. The studies demonstrated that even a column loading of 200 µg (one-third of the observed optimal column loading of 600 µg) provided an acceptable average step yield. Since the purification method was performed for testing purposes, maximizing the recovery percentage was not important. However, the yield percentage affected the final concentration of the purified mAb after elution twice with 400 µL of elution buffer. Further studies were conducted to purify proteins from small-scale bioreactor harvested material using the spin column purification method, and it was found that samples obtained by this purification method yielded product quality results comparable to those of samples purified on a small scale, demonstrating the applicability of the method to small-scale bioreactor harvested material.

[0493] Future studies will explore column loading ranges of 50–200 µg to investigate protein yields achievable with small-scale methods when adding smaller sample volumes. Low-concentration samples collected early in the preparation process (particularly on days 1 and 2) may require purification volumes larger than the maximum capacity of the rotating column (600 µL) to achieve the desired column loading. In such cases, the antibody binding step may be performed more than once to allow larger sample volumes to pass through the column. Additionally, if the initial purified mAb sample concentration is lower than the optimal concentration for PQ testing due to reduced column loading and yield %, the volume of elution buffer used during the two elution steps may be reduced. Optionally or additionally, a centrifugal filter may be used to concentrate the purified sample.

[0494] The ability to analyze cell culture samples daily throughout the upstream process can also improve the response to unexpected events that may affect process performance. Studying such events early in the batch duration through daily testing after purification using a developed small-scale method can help determine the potential impact within a day of the event's occurrence. Using conventional methods, the upstream generation process (which takes approximately two weeks, depending on the mAb) needs to be completed before the purification and analysis of harvested cell culture samples. Therefore, daily testing can save time and cost by introducing the ability to monitor PQ changes associated with process inputs.

[0495] The methods and systems of this invention can also be used to construct a database of product quality attribute profiles for monoclonal antibodies throughout the upstream production process. This can be achieved by continuously collecting data from consecutive large-scale production batches. The relevant product quality data can then be correlated with process parameters such as pH, temperature, feed strategy, or concentration of cell culture components. This will allow for establishing the impact of process parameter variations on the PQ profile. Such an understanding will support process-related studies in the event of unexpected parameter changes (e.g., a decrease in pH) and may be used to adjust process parameters as needed to obtain products with the desired PQ profile.

[0496] When using protein A resin, the methods and systems of this invention can be applied to any antibody or Fc-containing protein, such as Fc-receptor fusion proteins. Corresponding affinity resins (e.g., affinity resins containing target ligands) can be used to purify other proteins of interest. Further research will include using spin column purification methods as a reference for developing purification methods for other molecules, and developing similar methods for other purification types (e.g., ion exchange chromatography).

[0497] Furthermore, the developed method is easily automated and can be adapted to reduce operator workload and facilitate the purification of large volumes of samples. The method and system of this invention can be applied as part of an end-to-end purification and product quality analysis platform. The procedure can be automated with a robotic liquid handling system, and testing of multiple samples can be performed using 96-well batch chromatography (Rathore and Bhambure; Lambiase et al., 2023, Journal of Chromatography A, 463809).

[0498] Currently, in the product quality testing described in this disclosure, only glycan profiling is automated using commercially available pipetting robots. However, fracture testing can be readily transferred to the aforementioned platform. Recently, numerous automated high-throughput purification and multi-attribute method (MAM) platforms have been designed and documented in the published literature, which incorporate the analysis of multiple cPQ attributes (Sitasuwan et al., 2021, Monoclonal Antibody, 13(1):1978131; Yang et al., 2023, Monoclonal Antibody, 15(1):2197668; Lambiase et al., 2022, Journal of Chromatography A, 1670:462944; Liu et al., 2021, Journal of Pharmaceutical Sciences, 111(2):358-367). In addition, an automated system for direct sampling from an upstream bioreactor coupled to a MAM analysis platform for online monitoring purposes is described (Liu et al.; Dahotre et al., 2022, Journal of Chromatography A, 1672:463067).

[0499] This disclosure describes a high-throughput method for purifying monoclonal antibodies from upstream cell culture material for the purpose of monitoring key product quality attributes, including aggregation, fragmentation, charge variants, and glycan profiles. Samples purified using the developed method have been found to have a product quality profile comparable to samples purified using established large-scale downstream methods.

Claims

1. A method for enriching a protein of interest from a cell culture sample, wherein the protein of interest is dupilumab or aflibercept, the method comprising: (a) Contacting a cell culture sample including dupilumab with a centrifuge column including an affinity resin to produce an immobilized sample, wherein the affinity resin specifically binds to dupilumab. (b) subjecting the immobilized sample to at least one washing step; and (c) subject the immobilized sample from (b) to at least one elution step to produce enriched dupilumab.

2. The method according to claim 1, wherein the cell culture sample is derived from mammalian cell culture or insect cell culture.

3. The method according to claim 1, wherein the cell culture sample is derived from CHO cell culture, CHO-K1 cell culture, BHK cell culture, HEK 293 cell culture, Sf9 insect cell culture or a variant thereof.

4. The method according to claim 1, wherein the cell culture sample is a clarified cell culture sample.

5. The method according to claim 1, wherein the cell culture sample is collected from the cell culture on one of the following days: day 1 to day 20, day 3 to day 15, day 3 to day 12, day 3 to day 10, day 5 to day 10, day 5 to day 12, or day 5 to day 13.

6. The method of claim 1, wherein the cell culture sample is collected from the cell culture on a day selected from the group consisting of: day 1, day 2, day 3, day 4, day 5, day 6, day 7, day 8, day 9, day 10, day 11, day 12, day 13, day 14, day 15, day 16, day 17, day 18, day 19, or day 20.

7. The method of claim 1, further comprising repeating the method at least once.

8. The method of claim 7, wherein the method is repeated using at least a first cell culture sample and a second cell culture sample collected from the same cell culture.

9. The method of claim 8, wherein the first cell culture sample is collected on the first day and the second cell culture sample is collected on the second day.

10. The method of claim 8, wherein the first cell culture sample and the second cell culture sample are collected at a time between about 3 hours, about 6 hours, about 12 hours, about 18 hours, about 24 hours, about 1 day, about 2 days, about 3 days, about 4 days, about 5 days, about 6 days, about 7 days, about 8 days, about 9 days, or about 10 days.

11. The method of claim 1, further comprising performing the method in parallel on at least two cell culture samples.

12. The method of claim 11, wherein the at least two cell culture samples are derived from two different cell cultures.

13. The method of claim 1, wherein the contacting step comprises combining the cell culture sample and the binding buffer.

14. The method of claim 13, wherein the binding buffer comprises Tris-buffered saline, sodium phosphate, HEPES, or Tris.

15. The method of claim 14, wherein the binding buffer further comprises sodium chloride or calcium chloride.

16. The method of claim 13, wherein the binding buffer comprises sodium phosphate.

17. The method of claim 16, wherein the binding buffer further comprises sodium chloride.

18. The method of claim 13, wherein the pH of the binding buffer is from about 6 to about 8.

19. The method of claim 1, wherein the contacting step comprises adding a combined volume of binding buffer and the cell culture sample to the column, the combined volume being 250 µL to 1000 µL, 300 µL to 900 µL, 400 µL to 800 µL, 500 µL to 700 µL, 550 µL to 650 µL, 590 µL to 610 µL, 599 µL to 601 µL, or about 600 µL.

20. The method of claim 1, wherein the contacting step comprises adding a volume of the cell culture sample to the column, the volume being 50 µL to 100 µL, 60 µL to 90 µL, 70 µL to 80 µL, about 70 µL, about 71 µL, about 72 µL, about 73 µL, about 74 µL, about 75 µL, about 76 µL, about 77 µL, about 78 µL, about 79 µL, or about 80 µL.

21. The method of claim 1, wherein the contacting step comprises adding a quantity of protein to the column, the quantity being 100.5 µg to 804 µg, 250 µg to 1 g, 350 µg to 900 µg, 450 µg to 804 µg, 500 µg to 700 µg, 550 µg to 650 µg, 575 µg to 625 µg, 590 µg to 610 µg, about 595 µg, about 596 µg, about 597 µg, about 598 µg, about 599 µg, about 600 µg, about 601 µg, about 602 µg, about 603 µg, about 604 µg, or about 605 µg.

22. The method according to claim 1, wherein the affinity resin is protein A resin, protein G resin, or a combination thereof.

23. The method of claim 1, wherein the at least one washing step comprises adding a washing buffer to the column and centrifuging the column to produce a washed flow solution.

24. The method of claim 23, wherein the washing buffer comprises Tris-buffered saline, sodium phosphate, sodium acetate, HEPES, or Tris.

25. The method of claim 24, wherein the washing buffer further comprises sodium chloride or calcium chloride.

26. The method of claim 23, wherein the washing buffer comprises sodium phosphate or sodium acetate.

27. The method of claim 26, wherein the washing buffer further comprises sodium chloride.

28. The method of claim 23, wherein the pH of the washing buffer is about 6 to about 8.

29. The method of claim 1, wherein the number of washing steps is one, two, or three.

30. The method according to claim 1, wherein the number of washing steps is two.

31. The method of claim 30, wherein the washing buffer in the first washing step comprises sodium phosphate and sodium chloride, and the washing buffer in the second washing step comprises sodium acetate.

32. The method of claim 23, wherein the volume of the washing buffer is about 600 µL.

33. The method of claim 23, wherein the centrifugation is performed at a relative centrifugal force (RCF) of about 100.

34. The method of claim 23, wherein the centrifugation is performed for about 1 minute.

35. The method of claim 1, wherein the at least one elution step comprises adding an elution buffer to the column and centrifuging the column to produce an eluent.

36. The method of claim 35, wherein the elution buffer comprises acetic acid or glycine.

37. The method of claim 35, wherein the concentration of acetic acid is about 0.24% or about 40 mM.

38. The method of claim 35, wherein the concentration of acetic acid is about 0.12% or about 20 mM.

39. The method of claim 35, wherein the concentration of glycine is about 0.1 M.

40. The method of claim 35, wherein the volume of the elution buffer is about 400 µL.

41. The method of claim 35, wherein the pH of the elution buffer is 1 to 4, 2 to 4, 2.5 to 3.5, 2.8 to 3.2, about 1, about 1.5, about 2, about 2.5, about 3, or about 3.

5.

42. The method of claim 1, wherein the number of elution steps is one, two, or three.

43. The method of claim 35, wherein the centrifugation is performed at approximately 100 RCF.

44. The method of claim 35, wherein the centrifugation is performed for about 1 minute.

45. The method of claim 1, wherein the at least one elution step comprises adding a neutralization buffer to the column.

46. ​​The method of claim 45, wherein the neutralizing buffer comprises a Tris base.

47. The method of claim 46, wherein the concentration of the Tris base is 1 M to 2 M, about 1 M, about 1.5 M, or about 2 M.

48. The method of claim 45, wherein the volume of the neutralizing buffer is from 5 µL to 50 µL, about 5 µL, about 10 µL, about 20 µL, about 30 µL, about 40 µL, or about 50 µL.

49. The method of claim 1, wherein the enriched protein of interest yields a yield greater than 50%, greater than 60%, greater than 70%, greater than 80%, greater than 90%, greater than 95%, greater than 99%, about 60%, about 70%, about 75%, about 80%, about 85%, about 90%, about 91%, about 92%, about 93%, about 94%, about 95%, about 96%, about 97%, about 98%, about 99%, or about 100%.

50. The method of claim 1, wherein the amount of protein in the enriched protein of interest is greater than 10 µg, greater than 20 µg, greater than 50 µg, greater than 100 µg, greater than 200 µg, greater than 300 µg, greater than 400 µg, greater than 500 µg, greater than 600 µg, greater than 700 µg, greater than 800 µg, greater than 900 µg, greater than 1000 µg, greater than 1100 µg, greater than 1200 µg, about 10 µg, about 20 µg, about 50 µg, about 100 µg, about 200 µg, about 300 µg, about 400 µg, about 500 µg, about 600 µg, about 700 µg, about 800 µg, about 900 µg, about 1000 µg, about 1100 µg, or about 1200 µg.

51. The method of claim 1, wherein the concentration of the enriched protein of interest is greater than 0.01 µg / µL, greater than 0.05 µg / µL, greater than 0.1 µg / µL, greater than 0.2 µg / µL, greater than 0.5 µg / µL, greater than 1 µg / µL, greater than 2 µg / µL, about 0.05 µg / µL, about 0.1 µg / µL, about 0.2 µg / µL, about 0.5 µg / µL, about 1 µg / µL, about 1.5 µg / µL, about 2 µg / µL, or about 2.5 µg / µL.

52. The method of claim 1, wherein the duration of the method is less than 24 hours, less than 12 hours, less than 6 hours, less than 3 hours, less than 2 hours, less than 1 hour, less than 30 minutes, about 3 hours, about 2 hours, about 1.5 hours, about 1 hour, about 50 minutes, about 45 minutes, about 40 minutes, about 30 minutes, or about 20 minutes.

53. The method of claim 1, further comprising at least one product quality attribute characterizing the enriched protein of interest.

54. The method of claim 1, further comprising subjecting the enriched protein of interest to chromatography, mass spectrometry, spectroscopy, capillary electrophoresis, gel electrophoresis and / or ligand binding assays.

55. The method of claim 1, further comprising characterizing at least one size variant of the enriched protein of interest.

56. The method of claim 1, further comprising at least one high molecular weight species characterizing the enriched protein of interest.

57. The method of claim 56, wherein the characterization comprises subjecting the enriched protein of interest to size exclusion chromatography (SEC) analysis.

58. The method of claim 1, further comprising at least one fragment characterizing the enriched protein of interest.

59. The method of claim 58, wherein the characterization comprises subjecting the enriched protein of interest to capillary electrophoresis and sodium dodecyl sulfate (CE-SDS) analysis.

60. The method of claim 1, further comprising at least one charge variant characterizing the enriched protein of interest.

61. The method of claim 60, wherein the characterization comprises subjecting the enriched protein of interest to imaging capillary isoelectric focusing electrophoresis (iCIEF).

62. The method of claim 1, further comprising at least one glycan characterizing the enriched protein of interest.

63. The method of claim 62, wherein the characterization comprises subjecting the enriched protein of interest to hydrophilic interaction chromatography (HILIC) analysis.

64. The method of claim 53, further comprising using the at least one product quality attribute to determine whether the cell culture should continue or be terminated.

65. The method of claim 53, further comprising using the at least one product quality attribute to determine whether the cell culture should be modified.

Images