Method for purifying proteins containing fc

Abstract

The present invention relates to methods for purifying Fc-containing proteins. The present invention relates to methods for purifying proteins containing an Fc region, such as antibodies and Fc fusion proteins. In particular, the present invention relates to a purification method resulting in reduced levels of aggregate proteins, the method comprising adsorbing a protein containing an Fc region to a temperature-responsive protein A resin, and eluting the protein from the resin at a temperature below about 35C with an elution buffer comprising a chaotropic agent, a sugar alcohol, and at least one amino acid. Also described is a method for separating fully assembled antibodies from their half-antibody forms using an elution buffer.

Description

[0001] This application is a divisional application. The original application was filed on July 21, 2017, with application number 201780045463.1 (PCT / US2017 / 043384) and the invention title was "Purification method for Fc-containing proteins".

[0002] Cross-references to related applications

[0003] This application claims the benefit of U.S. Provisional Application No. 62 / 365,943, filed July 22, 2016, the entire contents of which are incorporated herein by reference.

[0004] Instructions for electronically submitted text files

[0005] This application contains a sequence list that has been electronically submitted in ASCII format, and the entire contents of that sequence list are incorporated herein by reference. A computer-readable copy of the sequence list created on July 20, 2017, is named A-2036-WO-PCT_SeqList_ST25 and is 8.81 kilobytes in size. Technical Field

[0006] This invention relates to the field of biopharmaceutical manufacturing. In particular, it relates to methods for reducing or preventing the aggregation of proteins containing Fc regions during purification processes. The invention also relates to methods for isolating antibodies from difficult-to-remove contaminants (e.g., haptens). Background Technology

[0007] Aggregates remain a major problem in the production of genetically engineered biologies, particularly fusion proteins comprising an immunoglobulin Fc region, such as multispecific antigen binding proteins. Protein aggregates are unacceptable in therapeutic drugs due to their strong immunogenicity (Maggio, Journal of Excipients and Food Chemicals, Vol. 3(2): 45-53, 2012; Sauerborn et al., Trends Pharmacol. Sci., Vol. 31(2): 53-59, 2010). Protein aggregates can occur during recombinant expression of the protein due to the vigorous shaking of the cell culture required for adequate aeration of the culture medium (Vazquez-Rey and Lang, Biotechnology and Bioengineering, Vol. 108(7): 1494-1508, 2011). Furthermore, many proteins are prone to unfolding, exposing hydrophobic regions, and subsequent aggregation upon exposure to extreme pH conditions (i.e., pH > 9.0 or pH < 4.5). Typical purification methods for proteins comprising an immunoglobulin Fc region require capture of the protein with a protein A affinity resin and subsequent elution from the resin with a low pH acidic buffer (e.g., pH 2.7 to 3.7) (Hari et al., Biochemistry, Vol. 49(43): 9328-9338, 2010; Ejima et al., PROTEINS: Structure, Function, and Bioinformatics, Vol. 66: 954-962, 2007). Low pH elution not only induces protein aggregation, resulting in reduced yield, but can also compromise the long-term stability of the recovered non-aggregated protein.

[0008] To avoid the adverse effects resulting from low pH elution of protein A resins, a temperature-responsive protein A resin was developed as a replacement for conventional protein A resins. This new resin consists of a mutant form of protein A that binds to the Fc region of immunoglobulins at temperatures below 10°C, but loses affinity for the Fc region at elevated temperatures (e.g., 40°C) (Koguma et al., Journal of Chromatography A, Vol. 1305: 149-153, 2013). Initially, the temperature-responsive protein A resin was received with great enthusiasm because it allowed for elution of proteins containing Fc regions from the resin at neutral pH by simply heating the column (Koguma et al., Journal of Chromatography A, Vol. 1305: 149-153, 2013). While purifying proteins with the temperature-responsive protein A resin resulted in stable molecules, this approach was impractical because it required a change in column temperature of 30°C or more from column loading to elution, thus creating a significant delay time between the loading and elution steps. The time delay required to manipulate the column temperature added a substantial cycle time to the purification operation, making this approach undesirable for large-scale manufacturing processes and high-throughput purification of large panels of proteins during the discovery phase. In addition, the effect of large-scale temperature fluctuations on the stability of proteins eluted from the temperature-responsive protein A resin remained unknown.

[0009] Accordingly, there is a need in the art for an effective purification method for proteins containing Fc regions that minimizes or reduces the impact on the structural integrity of the proteins associated with previous protein A chromatography. SUMMARY

[0010] The present invention is based in part on the development of a buffer composition that allows for elution of bound proteins containing Fc regions from temperature-responsive protein A resins at neutral pH and constant temperature. The purification protocol employing temperature-responsive protein A resins in combination with the elution buffer compositions described herein results in reduced levels of aggregated proteins, thereby reducing the number of downstream purification steps. In addition, such purification methods are easily scalable for industrial manufacturing because elution can be performed in a temperature-independent manner using the elution buffers described herein.

[0011] Accordingly, the present invention provides a method of purifying a protein comprising an Fc region. In one embodiment, the method comprises contacting a solution comprising the protein and one or more impurities with a temperature-responsive protein A material at a temperature at which the protein binds to the material; and eluting the protein from the material at a temperature of less than about 35°C with an elution buffer described herein, wherein the protein is purified from the one or more impurities in the solution.

[0012] In certain embodiments, the present application also provides methods for reducing aggregation during purification of a protein comprising an Fc region. In one embodiment, the method comprises adsorbing the protein to a temperature-responsive Protein A material at a temperature at which the protein binds to the material; and eluting the protein from the material at a temperature of less than about 35°C with an elution buffer described herein, wherein the amount of the protein comprising an Fc region in aggregated form in the eluate from the temperature-responsive Protein A material is less than the amount in an eluate from a conventional Protein A material. In some embodiments, the amount of the protein comprising an Fc region in aggregated form in the eluate from the temperature-responsive Protein A material is less than 30%. In related embodiments, at least 70% of the protein comprising an Fc region in the eluate from the temperature-responsive Protein A material is in monomeric form.

[0013] The elution buffer used in the methods of the present application has a neutral range of pH (e.g., a pH of about 6.5 to about 7.5) and comprises a chaotropic agent and a sugar alcohol. In some embodiments, the chaotropic agent and the sugar alcohol are present in a molar concentration ratio of about 0.4 to about 4.5. In one embodiment, the molar concentration ratio of the chaotropic agent to the sugar alcohol is about 1.1 to about 1.8. Depending on the particular chaotropic agent used, the concentration of the chaotropic agent in the elution buffer can be from about 0.4 M to about 5 M. In some embodiments, the chaotropic agent can be urea, guanidinium chloride, or a thiocyanate salt (e.g., sodium thiocyanate, potassium thiocyanate, or ammonium thiocyanate). In certain embodiments, the chaotropic agent is urea. In other embodiments, the chaotropic agent is guanidinium chloride. The concentration of the sugar alcohol in the elution buffer can be from about 1 M to about 4.5 M. In certain embodiments, the sugar alcohol is sorbitol, mannitol, or glycerol. In one embodiment, the sugar alcohol is sorbitol. In another embodiment, the sugar alcohol is mannitol.

[0014] In some embodiments, the elution buffer employed in the methods of the present application further comprises one or more amino acids. The amino acid can be a non-polar amino acid, such as alanine, cysteine, glycine, isoleucine, leucine, methionine, phenylalanine, proline, tryptophan, or valine, and / or a basic amino acid, such as histidine, lysine, ornithine, or arginine. In certain embodiments, the elution buffer comprises at least one non-polar amino acid and at least one basic amino acid. For example, in one embodiment, the elution buffer comprises proline and arginine. The concentration of the amino acid in the elution buffer can be from about 0.25 M to about 1 M.

[0015] In various embodiments, the elution buffer can further comprise a salt, such as a sodium salt or a chloride salt. The salt can be present in the elution buffer at a concentration of about 0.1 M to about 1 M. In some embodiments, the salt included in the elution buffer is sodium chloride.

[0016] In certain embodiments, the elution buffer used in the methods of the application comprises about 2 M to about 4.5 M of a chaotropic agent, about 1 M to about 4.5 M of a sugar alcohol, about 0.25 M to about 1 M of a basic amino acid, about 0.25 M to about 1 M of a non-polar amino acid, and about 0.25 M to about 0.8 M of a salt. In some such embodiments, the chaotropic agent is urea, the sugar alcohol is sorbitol, the basic amino acid is arginine, the non-polar amino acid is proline, and the salt is sodium chloride.

[0017] Elution of Fc region-containing proteins from temperature-responsive Protein A material according to the methods of the application can be performed at a temperature of from about 1 °C to about 25 °C. In some embodiments, elution of the protein is performed at the same or similar temperature at which the protein was adsorbed or bound to the temperature-responsive Protein A material. For example, in one embodiment, elution of the Fc region-containing protein is performed at a temperature of less than about 10 °C, e.g., from about 1 °C to about 6 °C. In other embodiments, elution of the Fc region-containing protein is performed at room temperature, e.g., at a temperature of from about 20 °C to about 25 °C.

[0018] Fc region-containing proteins that can be purified according to the methods described herein include antibodies, Fc-fusion proteins, and multi-specific antigen binding proteins. In some embodiments, the Fc region-containing protein is an antibody. In other embodiments, the Fc region-containing protein is an Fc-fusion protein, e.g., an Fc-fusion protein comprising at least one single chain Fv fragment. The Fc region-containing protein can be produced recombinantly, e.g., in mammalian cells. In such embodiments, the Fc region-containing protein can be purified from cell culture supernatant or cell lysate, e.g., those resulting from a harvest operation of a bioreactor.

[0019] The application is also based in part on the discovery that size exclusion chromatography can effectively separate fully assembled antibodies from half antibody contaminants when used with a buffer comprising a chaotropic agent, a sugar alcohol, and at least one amino acid as the mobile phase. Accordingly, the application also provides methods for separating an antibody from its half antibody form. In one embodiment, the method comprises contacting a solution comprising an antibody and its half antibody form with a gel filtration matrix using a mobile phase having a pH of from about 6.5 to about 7.5 and comprising a chaotropic agent, a sugar alcohol, a non-polar amino acid, and a basic amino acid; and collecting elution fractions from the gel filtration matrix, wherein the antibody is eluted in one set of elution fractions and its half antibody form is eluted in another set of elution fractions, thereby separating the antibody from its half antibody form. In certain embodiments of the method, the antibody is a multi-specific, e.g., bispecific, heterodimeric antibody.

[0020] Any of the elution buffers described herein for eluting a protein containing an Fc region from a temperature-responsive Protein A resin at neutral pH and constant temperature can be used as a mobile phase in size exclusion chromatography for separating an antibody from its half-antibody form. In some embodiments, the mobile phase comprises about 2 M to about 4.5 M of a chaotropic agent, about 1 M to about 4.5 M of a sugar alcohol, about 0.25 M to about 1 M of a basic amino acid, and about 0.25 M to about 1 M of a non-polar amino acid. In these and other embodiments, the mobile phase comprises urea, sorbitol, arginine, and proline. In certain embodiments, the mobile phase comprises about 2 M to about 4.5 M urea, about 1 M to about 4.5 M sorbitol, about 0.25 M to about 1 M arginine, about 0.25 M to about 1 M proline, and 0.25 M to about 0.8 M sodium chloride. In some embodiments, the mobile phase comprises about 4 M urea, about 2.2 M sorbitol, about 0.5 M arginine, about 0.5 M proline, and about 0.75 M sodium chloride. The pH of the mobile phase can be from about 6.5 to about 7.5, or in particular embodiments, from about 7.0 to about 7.4.

[0021] In certain embodiments of the method, the mobile phase is passed through the gel filtration matrix at a flow rate of about 0.01 ml / min to about 0.2 ml / min. In other embodiments, the mobile phase is passed through the gel filtration matrix at a flow rate of about 0.02 ml / min to about 0.06 ml / min. The gel filtration matrix can be composed of cross-linked agarose and dextran, and can have a fractionation range of about 10 kDa to about 600 kDa. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 Structure of IgG-scFv binding protein. This figure depicts a schematic of an IgG-scFv binding protein, which is an example of an Fc-containing protein that can be purified by the methods described herein. The IgG-scFv binding protein is composed of two single chain variable fragments (scFv), each containing a variable domain from a first antibody, linked together by a peptide linker, fused by another peptide linker to the carboxy terminus of the heavy chain of a second antibody.

[0023] Figure 2 IgG-scFv binding protein eluted from a conventional Protein A column. This figure shows the SEC-UPLC chromatogram of a sample from a Protein A eluate pool of IgG-scFv binding protein. The binding protein was eluted from a conventional Protein A chromatography column using 1% acetic acid buffer, pH 2.7, at 4°C. The eluate outside of the SEC-UPLC column contained an aggregate peak (indicated with a black arrow) and a monomer peak (indicated with a black star). Each of the four graphs is a separate experiment.

[0024] Figure 3IgG-scFv binding protein eluted from temperature-responsive protein A chromatography column at 4°C. This plot shows the SE-UPLC chromatogram for a sample from the protein A eluate pool of IgG-scFv binding protein. The binding protein was eluted from the temperature-responsive protein A chromatography column at 4°C using an elution buffer with a pH of 7.2 and containing 50 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea. The monomer peak is indicated by the black star. Each of the three plots is a separate experiment.

[0025] Figure 4A SDS-PAGE of IgG-scFv binding protein during purification on temperature-responsive protein A chromatography column. The samples in each lane on the gel are as follows: Stds = protein standards; Feed = sample of clarified cell culture supernatant prior to column loading; FT = sample of flow-through fraction of the column; and Eluate = sample of the eluate pool after eluting the binding protein from the column at room temperature with an elution buffer containing 25 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea at a pH of 7.2.

[0026] Figure 4B IgG-scFv binding protein eluted from temperature-responsive protein A chromatography column at room temperature. This plot shows the SE-HPLC chromatogram for a sample from the temperature-responsive protein A eluate pool of IgG-scFv binding protein. The binding protein was eluted from the temperature-responsive protein A chromatography column at room temperature using an elution buffer with a pH of 7.2 and containing 25 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea. The peak with a retention time of about 14 minutes is the monomeric form of the binding protein. The inset provides the retention time and peak area and height for each peak shown on the chromatogram. Approximately 99% of the IgG-scFv binding protein present in the eluate pool was in the monomeric form.

[0027] Figure 5 Structure of a single-chain bispecific Fv-Fc binding protein. This plot depicts a schematic of a single-chain bispecific Fv-Fc binding protein, which is an example of an Fc-containing protein that can be purified by the methods described herein. The single-chain bispecific Fv-Fc binding protein comprises a first scFv fragment containing heavy and light chain variable domains from a first antibody fused to a second scFv fragment containing heavy and light chain variable domains from a second antibody, and an Fc region fused at its N-terminus to the first scFv fragment by a peptide linker.

[0028] Figure 6A SE-HPLC chromatogram of bispecific Fv-Fc binding protein eluted from a conventional Protein A column using 174 mM acetic acid. Bispecific Fv-Fc binding protein was eluted from a conventional Protein A chromatography column at room temperature using 174 mM (1%) acetic acid, pH 2.7. The black arrow indicates the peak corresponding to binding protein aggregates. The black star indicates the peak corresponding to monomeric form of binding protein. The inset provides the retention time and peak area and height for each peak shown on the chromatogram. Approximately 39% of the bispecific Fv-Fc binding protein present in the eluate pool was in monomeric form.

[0029] Figure 6B SE-HPLC chromatogram of bispecific Fv-Fc binding protein eluted from a conventional Protein A column using 33 mM acetic acid. Bispecific Fv-Fc binding protein was eluted from a conventional Protein A chromatography column at room temperature using 33 mM (0.06%) acetic acid, pH 3.7. The black arrow indicates the peak corresponding to binding protein aggregates. The black star indicates the peak corresponding to monomeric form of binding protein. The inset provides the retention time and peak area and height for each peak shown on the chromatogram. Approximately 53% of the bispecific Fv-Fc binding protein present in the eluate pool was in monomeric form.

[0030] Figure 7A SE-HPLC chromatogram of bispecific Fv-Fc binding protein eluted from a temperature-responsive Protein A column. Bispecific Fv-Fc binding protein was eluted from a temperature-responsive Protein A chromatography column at room temperature using an elution buffer having a pH of 7.2 and comprising 25 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 2.5 M urea. The inset provides the retention time and peak area and height for the peak shown on the chromatogram. The black arrow indicates the peak corresponding to binding protein aggregates. The peak having a retention time of about 15.9 minutes corresponds to monomeric form of Fv-Fc binding protein and is annotated with a black star. Approximately 75% of the bispecific Fv-Fc binding protein present in the eluate pool was in monomeric form.

[0031] Figure 7B. SE-HPLC chromatogram of bispecific Fv-Fc binding protein eluted from a temperature-responsive Protein A column. The bispecific Fv-Fc binding protein was eluted from a temperature-responsive Protein A column at room temperature using an elution buffer having a pH of 7.2 and comprising 25 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea. The inset provides the retention time and peak area and height for each peak shown on the chromatogram. The black arrow indicates the peak corresponding to the binding protein aggregate. The peak having a retention time of about 15.9 minutes corresponds to the monomeric form of the Fv-Fc binding protein and is annotated with a black star. Approximately 88% of the bispecific Fv-Fc binding protein present in the eluate pool was in the monomeric form.

[0032] Figure 7C . SDS-PAGE of bispecific Fv-Fc eluate from a temperature-responsive Protein A column buffer. Different samples were taken during purification of the bispecific Fv-Fc binding protein using a temperature-responsive Protein A column. The samples in each lane on the gel were as follows: Stds = protein standards; Feed = sample of clarified cell culture supernatant prior to column loading; FT = sample of column flow-through fraction; and Eluate = sample of the eluate pool after elution of the binding protein from the column at room temperature using an elution buffer comprising 25 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea at a pH of 7.2.

[0033] Figure 8A . SE-HPLC chromatogram of a monoclonal antibody eluted from a temperature-responsive Protein A column at 4°C. The antibody was eluted from a temperature-responsive Protein A column at 4°C using an elution buffer having a pH of 7.2 and comprising 20 mM HEPES, 2 M guanidinium chloride, and 2.2 M sorbitol. The peak corresponding to the monomeric form of the antibody is indicated with a black star.

[0034] Figure 8B . Table summarizes characteristics of peaks shown in the chromatogram in Figure 8A approximately 90% of the monoclonal antibody was recovered in the monomeric form in the eluate pool.

[0035] Figure 9A and 9B . Figure depicts LCMS chromatograms of recombinant bispecific heterodimeric antibody from two different lots after conventional Protein A purification. The fully assembled heterodimeric antibody having a predicted mass of 148351 Daltons was detected in both lots, and a half antibody having a predicted mass of 74355 Daltons. No other species of half antibody having a predicted mass of 74004 Daltons was detected. The level of half antibody varied from lot to lot.

[0036] Figure 10A The figure shows an SE-HPLC chromatogram of a sample from the standard protein A elution cell of a bispecific heterodimeric antibody (“heterodimeric antibody A”). The fully assembled heterodimeric antibody (“whole antibody”) eluted before the hemiantibody on the size exclusion column. The table below the chromatogram provides retention times, peak areas, widths, and heights for the two peaks shown on the chromatogram. Approximately 72% of the hemiantibody was present in the elution cell.

[0037] Figure 10B SDS-PAGE analysis of samples from the standard protein A elution buffer pool of the bispecific heterodimeric antibody (“heterodimeric antibody A”). Indicator sample volumes under reducing (left side of the gel) or non-reducing (right side of the gel) conditions were loaded onto Tris-glycine SDS gels in a 4%–20% gradient. Protein standards were loaded into the middle lane. Significant amounts of the hapten were present after purification by standard protein A chromatography.

[0038] Figure 11 Preparative SEC chromatogram of the bispecific heterodimeric antibody formulation (“heterodimeric antibody A”). Preparative SEC gel filtration was performed using a mobile phase containing PBS, pH 7.0, relative to the bispecific heterodimeric antibody formulation. SEC performed under standard conditions cannot separate the half-antibody from the fully assembled heterodimeric antibody and other incomplete fragments.

[0039] Figure 12 Elution profile of a bispecific heterodimeric antibody (“heterodimeric antibody A”) from a preparative SEC gel filter column (2 x 80 ml Superdex 200) using standard protein A elution buffer. SEC was performed at a flow rate of 0.02 ml / min using a mobile phase containing 50 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea at pH 7.2. Three major protein peaks were observed.

[0040] Figure 13A and 13B SDS-PAGE analysis of the elution fraction from SEC, the curves are shown in... Figure 12 Collected through each of the three peaks. “L” = Material loaded before SEC. In non-reduction ( Figure 13A ) or restore ( Figure 13B) under non-reducing (left side of gel) or reducing (right side of gel) conditions. Peak 2 contains mostly fully assembled antibody ("F"), while peak 3 contains mostly half-antibodies ("H"). Peak 1 corresponds to higher molecular weight aggregates. SEC with a mobile phase comprising 50 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea at pH 7.2 effectively separates fully assembled heterodimeric antibody from half-antibodies.

[0041] Figure 14A Figure 3 shows analytical SE-HPLC chromatograms from a sample of loading material comprising bispecific heterodimeric antibody before SEC and from the final pool after SEC. The sample of loading material was loaded onto a preparative SEC gel filtration column (2 x 80 ml Superdex 200). SEC was performed with a mobile phase comprising 50 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea at pH 7.2 at a flow rate of 0.02 ml / min. Fully assembled heterodimeric antibody elutes primarily in peak 2, while half-antibodies elute primarily in peak 3. Higher molecular weight aggregates elute in peak 1. SEC with this mobile phase can effectively separate fully assembled heterodimeric antibody from half-antibody contaminants. Figure 12

[0042] Figure 14B Figure 4 shows SDS-PAGE analysis of a sample of recombinant heterodimeric antibody from a conventional protein A eluate pool, loading material before SEC, and the final pool after SEC. The samples were loaded onto a 4-20% gradient Tris-glycine SDS gel under non-reducing (left side of gel) or reducing (right side of gel) conditions.

[0043] Figure 14C Figure 5 shows LCMS chromatograms of recombinant bispecific heterodimeric antibody after SEC purification with a mobile phase comprising 50 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea at pH 7.2. Only fully assembled heterodimeric antibody is detectable. Neither of the two half-antibodies is detectable. SEC effectively removes half-antibody contaminants. Compare with Figure 9A and 9B .

[0044] Figure 15A and 15B ​Figure depicts LCMS chromatograms of recombinant bispecific heterodimeric antibody B before (left panel) and after (right panel) SEC purification with a mobile phase of pH 7.2 containing 50 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea. Prior to SEC purification, the preparation contained fully assembled antibody and half antibody. After SEC purification, the half antibody could not be detected. Figure 15A Figure 15B

[0045] Figure 16A 16B Figure depicts LCMS chromatograms of recombinant bispecific heterodimeric antibody C before (left panel) and after (right panel) SEC purification with a mobile phase of pH 7.2 containing 50 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea. Prior to SEC purification, the preparation contained fully assembled antibody and half antibody. After SEC purification, the half antibody could not be detected. Figure 16A Figure 16B

[0046] Figures 17A-17F SDS-PAGE analysis of recombinant bispecific heterodimeric antibodies from elution fractions of preparative SEC gel filtration columns using a mobile phase of pH 7.2 containing 50 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea. SEC was performed at different flow rates: 0.02 ml / min (left panel), Figure 17A 0.04 ml / min (middle left panel), Figure 17B 0.06 ml / min (middle right panel), Figure 17C 0.08 ml / min (top right panel), Figure 17D 0.1 ml / min (top left panel), and Figure 17E 0.2 ml / min (bottom panel). "F" = fully assembled heterodimeric antibody. "H" = half antibody. Samples were loaded on 4-20% gradient Tris-glycine SDS gels under non-reducing conditions. Half antibody separation efficiency decreased with increasing flow rate. Figure 17F DETAILED DESCRIPTION

[0047] ​​​​​​The present invention is based, in part, on the development of purification methods for Fc region-containing proteins (e.g., antibodies and Fc fusion proteins) that minimize the level of aggregated protein and other contaminants (e.g., half antibodies). Conventional methods for purifying Fc region-containing proteins typically employ a protein A affinity chromatography column that requires an extremely acidic solution to elute the bound protein from the column. This acidic solution often induces aggregation, instability, and (frequently) precipitation of the protein. While a modified temperature-responsive protein A resin has been developed that allows elution of bound protein at neutral pH using a temperature shift, this modified resin is not suitable for large-scale manufacturing operations due to the large amount of time required for the resin temperature to increase.

[0048] The present inventors have designed a purification method for Fc region-containing proteins that avoids the use of the low pH elution buffer of traditional protein A chromatography and the elevated temperature required for modified temperature-responsive protein A resins. The purification method of the present invention employs an elution buffer that comprises a chaotropic agent, a sugar alcohol, and optionally one or more amino acids. The composition of the elution buffer allows for the removal or elution of proteins bound to temperature-responsive protein A resins from the resin at neutral pH without increasing the temperature of the resin above 35°C. It was also surprisingly found that the elution buffer can also be used as a mobile phase in size exclusion chromatography to effectively separate antibodies from their half antibody forms.

[0049] Thus, in one embodiment, the present invention provides a method of purifying an Fc region-containing protein from a solution comprising the protein and one or more impurities, the method comprising contacting the solution with a temperature-responsive protein A material at a temperature at which the protein binds to the material; and eluting the protein from the material at a temperature of less than about 35°C with an elution buffer described herein.

[0050] In another embodiment, the present invention provides a method of separating an antibody from its half antibody form, the method comprising contacting a solution comprising the antibody and its half antibody form with a gel filtration matrix using one of the elution buffers described herein as a mobile phase, and collecting elution fractions from the gel filtration matrix, wherein the antibody elutes in one set of elution fractions and its half antibody form elutes in another set of elution fractions, thereby separating the antibody from its half antibody form.

[0051] The methods of the present application are particularly suitable for purifying proteins comprising an immunoglobulin Fc region. As used herein, a "protein comprising an Fc region" or a "protein containing an Fc region" refers to a protein or polypeptide comprising a contiguous amino acid sequence that corresponds to the amino acid sequence of an immunoglobulin Fc region. The term "Fc region" refers to a C-terminal region of an immunoglobulin heavy chain, which can be produced by papain digestion of an intact antibody. The Fc region of an immunoglobulin generally comprises two constant domains, a CH2 domain and a CH3 domain, and optionally a CH4 domain. In certain embodiments, the Fc region is an Fc region from an IgGl, IgG2, IgG3, or IgG4 immunoglobulin. In some embodiments, the Fc region comprises CH2 and CH3 domains from a human IgGl or human IgG2 immunoglobulin.

[0052] Proteins containing an Fc region that can be purified according to the methods of the present application include, but are not limited to, antibodies, Fc-fusion proteins, and multi-specific antigen binding proteins. As used herein, the term "antibody" refers to a tetrameric immunoglobulin comprising two light chain polypeptides (each about 25 kDa) and two heavy chain polypeptides (each about 50-70 kDa). The term "light chain" or "immunoglobulin light chain" refers to a polypeptide comprising, from amino-terminus to carboxyl-terminus, a single immunoglobulin light chain variable region (VL) and a single immunoglobulin light chain constant region (CL). The immunoglobulin light chain constant domain (CL) can be kappa (K) or lambda (l). The term "heavy chain" or "immunoglobulin heavy chain" refers to a polypeptide comprising, from amino-terminus to carboxyl-terminus, a single immunoglobulin heavy chain variable region (VH), an immunoglobulin heavy chain constant domain 1 (CHI), an immunoglobulin hinge region, an immunoglobulin heavy chain constant domain 2 (CH2), an immunoglobulin heavy chain constant domain 3 (CH3), and optionally an immunoglobulin heavy chain constant domain 4 (CH4). The heavy chains are classified as mu (m), delta (d), gamma (g), alpha (a), and epsilon (e), and define the isotype of an antibody as IgM, IgD, IgG, IgA, and IgE, respectively. The IgG and IgA classes of antibodies are further divided into subclasses, i.e., IgGl, IgG2, IgG3, and IgG4, and IgAl and IgA2, respectively. The heavy chains in the IgG, IgA, and IgD antibodies have three domains (CHI, CH2, and CH3), while the heavy chains in the IgM and IgE antibodies have four domains (CHI, CH2, CH3, and CH4). The immunoglobulin heavy chain constant domains can be from any immunoglobulin isotype, including subtypes. The antibody chains are linked together via inter-polypeptide disulfide bonds between the CL and CHI domains (i.e., between the light and heavy chains) and between the hinge regions of the antibody heavy chains.

[0053] Antibodies that can be purified according to the methods of the present application include, but are not limited to, polyclonal antibodies, monoclonal antibodies, recombinant antibodies, chimeric antibodies, humanized antibodies, and human antibodies. In some embodiments, the antibody is an acid-sensitive antibody. As used herein, “acid-sensitive antibody” refers to an antibody that is unstable, aggregates, or loses structural integrity under acidic conditions, e.g., at a pH below about 6. In certain embodiments, the antibody is a multispecific (e.g., bispecific) heterodimeric antibody. A multispecific heterodimeric antibody refers to an antibody that is capable of specifically binding two or more different antigens and comprises two different light chains and two different heavy chains. For example, in some embodiments, a bispecific heterodimeric antibody comprises a light chain and a heavy chain from a first antibody that specifically binds a first antigen, and a light chain and a heavy chain from a second antibody that specifically binds a second antigen (see Figure 9A and 9B the right, where the filled symbols represent light and heavy chains that form a binding site with the first antigen, and the unfilled symbols represent light and heavy chains that form a binding site with the second antigen). Bispecific heterodimeric antibodies can be produced by co-expressing two light chains and two heavy chains in the same cell or by separately expressing the polypeptide chains and subsequently assembling them. To facilitate heterodimer formation, the polypeptide chains can be engineered using, for example, a “knobs-into-holes” approach or a charge-pairing approach. These approaches are known to those of skill in the art and are described in WO 96 / 027011; Ridgway et al., Protein Eng., vol. 9: 617-621, 1996; Merchant et al., Nat, Biotechnol., vol. 16: 677-681, 1998; WO 2009 / 089004; WO 2014 / 081955; and Gunasekaran et al., J. Biol. Chem., vol. 285: 19637-19646, 2010, which are incorporated by reference herein in their entireties.

[0054] In certain embodiments, the antibody purified according to the methods of the present application is a monoclonal antibody.

[0055] As used herein, the term "monoclonal antibody" (or "mAb") refers to an antibody obtained from a population of substantially homogeneous antibodies, i.e., the individual antibodies comprising the population are identical except for possible naturally occurring mutations that can be present in minor amounts. Monoclonal antibodies are highly specific, being directed against a single antigenic site or epitope, as opposed to polyclonal antibody preparations, which typically include different antibodies directed against different epitopes. Monoclonal antibodies can be produced using any technique known in the art, for example, by immortalizing spleen cells harvested from a transgenic animal after completion of an immunization regimen. The spleen cells can be immortalized using any technique known in the art, for example, by fusing them with myeloma cells to produce hybridomas. The myeloma cells used in the fusion process to produce hybridomas are preferably non-antibody producing, have high fusion efficiency, and enzyme deficiencies, and are such that they can be selected by their inability to grow in certain selective media unique to the absence of neomycin. Examples of suitable cell lines for mouse fusion include Sp-20, P3-X63 / Ag8, P3-X63-Ag8.653, NS1 / 1.Ag 41, Sp210-Ag14, FO, NSO / U, MPC-11, MPC11-X45-GTG 1.7, and S194 / 5XXO Bul; examples of cell lines for rat fusion include R210.RCY3, Y3-Ag 1.2.3, IR983F, and 4B210. Other cell lines for cell fusion are U-266, GM1500-GRG2, LICR-LON-HMy2, and UC729-6.

[0056] In some embodiments, the monoclonal antibody can be a humanized antibody. A "humanized antibody" refers to an antibody in which regions (e.g., framework regions) have been modified to comprise corresponding regions from a human immunoglobulin. Typically, a humanized antibody can be generated from a monoclonal antibody that was originally produced in a non-human animal. Certain amino acid residues in the monoclonal antibody, typically from the non-antigen recognition portion of the antibody, are modified to be homologous to the corresponding residues in a corresponding antibody of the same isotype in humans. For example, humanization can be performed using various methods, such as by substituting at least a portion of the variable region of a rodent with the corresponding region of a human antibody (see, e.g., U.S. Patent Nos. 5,585,089 and 5,693,762; Jones et al., Nature, vol. 321:522-525, 1986; Riechmann et al., Nature, vol. 332:323-27, 1988; Verhoeyen et al., Science, vol. 239:1534-1536, 1988). The CDRs of the light and heavy chain variable regions of an antibody produced in another species can be grafted into a consensus human framework region (FR). To generate a consensus human FR, the FRs from several human heavy or light chain amino acid sequences can be aligned to identify consensus amino acid sequences.

[0057] In other embodiments, the monoclonal antibody can be a fully human antibody. A "fully human antibody" is an antibody comprising variable and constant regions derived from or indicative of human germline immunoglobulin sequences. Fully human antibodies can be produced by immunizing a transgenic animal, typically a mouse, that is capable of producing a human antibody repertoire in the absence of endogenous immunoglobulin production. The antigen for this purpose is typically of six or more contiguous amino acids and is optionally conjugated to a carrier, such as a hapten. See, e.g., Jakobovits et al., 1993, Proc. Natl. Acad. Sci. USA 90:2551-2555; Jakobovits et al., 1993, Nature, 362:255-258; and Bruggermann et al., 1993, Year in Immunol. 7:33. In one example of this approach, a transgenic animal is produced by disabling the endogenous mouse immunoglobulin loci encoding mouse heavy and light immunoglobulin chains and inserting large fragments of human genomic DNA containing loci encoding human heavy and light chain proteins into the mouse genome. The animal with less than the full complement of human immunoglobulin loci is then crossed with a modified animal to obtain an animal with all the desired modifications of the immune system. When given an immunogen, these transgenic animals produce antibodies that are immunospecific for the immunogen but have human rather than murine amino acid sequences, including the variable regions. For further details of such methods, see, e.g., WO 96 / 33735 and WO 94 / 02602. Additional methods related to transgenic mice for making human antibodies are described in U.S. Patent Nos. 5,545,807; 6,713,610; 6,673,986; 6,162,963; 5,939,598; 5,545,807; 6,300,129; 6,255,458; 5,877,397; 5,874,299; and 5,545,806; PCT publications WO 91 / 10741, WO 90 / 04036, WO 94 / 02602, WO 96 / 30498, WO 98 / 24893, and EP 546073B1 and EP 546073A1.

[0058] Human-derived antibodies can also be generated using phage display technology. Phage display is described, for example, in Dower et al., WO 91 / 17271, McCafferty et al., WO 92 / 01047, and Caton and Koprowski, Proc. Natl. Acad. Sci. USA, 87:6450-6454 (1990), each of which is incorporated by reference herein in its entirety. Antibodies generated by phage technology are typically generated as antigen-binding fragments, e.g., Fv or Fab fragments, in bacteria and thus lack effector functions. Effector functions can be introduced by one of two strategies: fragments can be engineered into whole antibodies for expression in mammalian cells, if desired, or into bispecific antibody fragments with a second binding site capable of triggering effector functions. Typically, the Fd fragment (VH-CH1) and light chain (VL-CL) of an antibody are cloned separately by PCR and randomly recombined in combinatorial phage display libraries, which can then be selected for binding to a particular antigen. Antibody fragments are expressed on the phage surface and Fv or Fab selected by antigen binding and reamplification through several rounds of antigen binding and reamplification, a process called panning. Antibody fragments specific for the antigen are enriched and eventually isolated. Phage display technology can also be used for a method of humanization of rodent monoclonal antibodies, called "guided selection" (see Jespers, L.S., et al., Bio / Technology 12, 899-903 (1994)). For this, the Fd fragment of a mouse monoclonal antibody is displayed in combination with a human light chain library, and the resulting hybrid Fab library can then be selected with antigen. Thus, the mouse Fd fragment provides a template for guided selection. Subsequently, the selected human light chains are combined with a library of human Fd fragments. The resulting library generates human Fab entirely.

[0059] In certain embodiments, the protein comprising an Fc region to be purified according to the methods of the application is an Fc fusion protein. An "Fc fusion protein" is a protein that contains an Fc region fused or linked to a heterologous polypeptide. Typically, the fusion protein is expressed from a fusion gene in which a nucleotide sequence encoding a polypeptide sequence from one protein (e.g., an Fc region) is attached in-frame with a nucleotide sequence encoding a polypeptide sequence, and optionally separated by a linker, from a different protein. The fusion gene can then be expressed by a recombinant host cell to produce a single fusion protein. The heterologous polypeptide fused to the Fc region can be a polypeptide from a protein other than an immunoglobulin. For example, the heterologous polypeptide can be a ligand polypeptide, a receptor polypeptide, a hormone, a cytokine, a growth factor, an enzyme, or other polypeptide that is not a component of an immunoglobulin. Such Fc fusion proteins can comprise an Fc region fused to a receptor or fragment thereof or a ligand from a receptor, including but not limited to any of the following receptors: the two forms of TNFR (referred to as p55 and p75), Type I and Type II interleukin 1 receptors (as described in European Patent No. 0460846, U.S. Patent No. 4,968,607, and U.S. Patent No. 5,767,064, which are incorporated by reference in their entireties herein), interleukin 2 receptor, interleukin 4 receptor (as described in European Patent No. 0367566, U.S. Patent No. 5,856,296, which are incorporated by reference in their entireties herein), interleukin 15 receptor, interleukin 17 receptor, interleukin 18 receptor, granulocyte macrophage colony stimulating factor receptor, granulocyte colony stimulating factor receptor, oncostatin-M and leukemia inhibitory factor receptor, NF-κΒ receptor activator (RANK, as described in U.S. Patent No. 6,271,349, which is incorporated by reference in its entirety herein), VEGF receptor, EGF receptor, FGF receptor, TRAIL receptors (including TRAIL receptors 1, 2, 3, and 4), and receptors comprising a death domain, such as Fas or apoptosis-inducing receptor (AIR). Fc fusion proteins also include peptibodies, such as those described in WO 2000 / 24782, which is incorporated by reference in its entirety herein.

[0060] In other embodiments, the Fc region fused or linked heterologous polypeptide can be a polypeptide from an immunoglobulin or fragment thereof, rather than an immunoglobulin derived from an Fc region. For example, the heterologous polypeptide can be a heavy and / or light chain variable region from a different antibody than the antibody from which the Fc region was obtained. In certain embodiments, the Fc fusion protein comprises at least one single chain Fv fragment (scFv fragment). A “single chain variable antibody fragment” or “scFv fragment” comprises the VH and VL regions of an antibody, wherein these regions are present in a single polypeptide chain, and optionally, a peptide linker between the VH and VL regions, such that the Fv is capable of forming the desired structure for antigen binding (see, e.g., Bird et al., Science, vol. 242:423-426, 1988; and Huston et al., Proc. Natl. Acad. Sci. USA, vol. 85:5879-5883, 1988). In some embodiments, the Fc fusion protein comprises two scFv fragments. In other embodiments, the Fc fusion protein comprises three scFv fragments. In still other embodiments, the Fc fusion protein comprises four scFv fragments.

[0061] Fc fusion proteins comprising one or more scFv fragments or other fragments of antibody variable regions can have multiple binding sites for one or more antigens. Accordingly, such Fc fusion proteins can include multispecific, multivalent antigen binding proteins, such as the small modular immunopharmaceuticals described in U.S. Pub. No. 20030133939; the single chain multivalent binding proteins described in WO 2007 / 146968, the bispecific bivalent scFv-Fc molecules described in WO 2014144722, and various bispecific antibody molecules such as IgG-scFv, IgG-Fab, 2scFv-IgG, 4scFv-IgG, VH-IgG, IgG-VH, and Fab-scFv-Fc, and others described in Spiess et al., Mol Immunol., vol. 67:95-106, 2015, and Kontermann, mAbs, vol. 4:182-197, 2012. In one embodiment, the Fc fusion protein purified according to the methods of the application is an IgG-scFv binding protein. As used herein, an “IgG-scFv binding protein” is a binding protein comprising two single chain variable fragments (scFv), each fragment comprising variable domains from a first antibody, linked together by a peptide linker, fused by another peptide linker to the carboxy terminus of a heavy chain of a second antibody. An example of an IgG-scFv binding protein is depicted in Figure 1In another embodiment, the Fc fusion protein purified according to the method of the invention is an Fv-Fc binding protein. An “Fv-Fc binding protein” comprises at least one scFv fragment optionally fused to the Fc region via a linker. In some embodiments, the Fv-Fc binding protein comprises two scFv fragments fused to each other, optionally fused sequentially to the N-terminus or C-terminus of the Fc region via a linker peptide. Such bivalent Fv-Fc molecules are described in WO 2014144722, the entire contents of which are incorporated herein by reference. An example of such a single-chain, bispecific Fv-Fc binding protein is shown in… Figure 5 middle.

[0062] Antibodies and Fc fusion proteins that can be purified using the methods described herein can bind one or more proteins, including but not limited to CD2, CD3, CD4, CD8, CD11a, CD14, CD18, CD19, CD20, CD22, CD23, CD28, CD25, CD33, CD40, CD44, CD52, CD80 (B7.1), CD86 (B7.2), CD147, IL-1a, IL-1b, IL-4, IL-5, IL-8, IL-10, IL-13, IL-2 receptor, IL-4 receptor, IL-6 receptor, IL-13 receptor, IL-I8 receptor subunit, Angiogenin (e.g., Angiogenin-1, Angiogenin-2, or Angiogenin-4), PDGF-b, VEGF, TGF, TGF-b2, TGF-b1, EGF receptor, VEGF receptor, FGF receptor, C5 complement, b-klotho, Calcitonin Gene-Related Peptide (CGRP), CGRP receptor, Pituitary Adenylate Cyclase-Activating Polypeptide (PACAP), Pituitary Adenylate Cyclase-Activating Polypeptide Type 1 Receptor (PAC1 receptor), IgE, tumor antigens, e.g., tumor antigen CA125, tumor antigen MUC1, PEM antigen, PD-1, LCG, which is a lung cancer-associated expressed gene product, HER-2, tumor-associated glycoprotein TAG-72, SK-l antigen, integrin a4b7, integrin VLA-4, B2 integrin, TRAIL receptors 1, 2, 3, and 4, RANK, RANK ligand, Sclerostin, Dickkopf-1 (DKK-1), TLA1, TNF-a, adhesion molecule VAP-l, epithelial cell adhesion molecule (EpCAM), intercellular adhesion molecule-3 (ICAM-3), leukocyte integrin adhesion, platelet glycoprotein gp IIb / IIIa, cardiac myosin heavy chain, PCSK9, parathyroid hormone, rNAPc2, MHC I, carcinoembryonic antigen (CEA), alpha-fetoprotein (AFP), tumor necrosis factor (TNF), CTLA-4, which is a cytotoxic T-lymphocyte-associated antigen, Fc-gamma-l receptor, HLA-DR 10b, HLA-DR antigen, L-selectin, IPN-gamma, respiratory syncytial virus, human immunodeficiency virus (HIV), hepatitis B virus (HBV), Streptococcus mutans, and Staphylococcus aureus.

[0063] Other exemplary Fc region-containing proteins that can be purified according to the methods described herein include, but are not limited to, aflibercept Alemtuzumab Bevacizumab Cetuximab Panitumumab gemtuzumab evolocumab alirocumab denosumab rituximab tocilizumab tilmanocept trastuzumab eculizumab adalimumab infliximab etanercept dalizumab basiliximab palivizumab omalizumab abciximab efalizumab pembrolizumab nivolumab natalizumab brentuximab vedotin filgrastim

[0064] The Fc region-containing protein to be purified can be produced by recombinant methods, i.e., by a living host cell that has been genetically engineered to produce the protein. Methods of genetically engineering cells to produce proteins are well known in the art. See, e.g., Ausabel et al., eds., (1990), Current Protocols in Molecular Biology (Wiley, New York). Such methods include the introduction of nucleic acids that encode and allow for the expression of the protein into a living host cell. These host cells can be prokaryotic cells, yeast cells, or higher eukaryotic cells grown in culture. Prokaryotic host cells include eubacteria, such as Gram-negative or Gram-positive organisms, such as Enterobacteriaceae, like Escherichia, e.g., E. coli, Enterobacter, Erwinia, Klebsiella, Proteus, Salmonella, e.g., S. typhimurium, Serratia, e.g., Serratia marcescans, and Shigella, also Bacillus, e.g., B. subtilis and B. licheniformis, Pseudomonas, and Streptomyces. Saccharomyces cerevisiae or common baker's yeast is the most commonly used of lower eukaryotic host microorganisms.However, many other genera, species, and strains are generally useful herein, such as Pichia, e.g., P. pastoris, Schizosaccharomyces pombe; Kluyveromyces, Yarrowia; Candida; Trichoderma reesia; Neurospora crassa; Schwanniomyces, e.g., Schwanniomyces occidentalis; and filamentous fungi, such as, for example, Neurospora, Penicillium, Tolypocladium, and Aspergillus hosts, e.g., A. nidulans and A. niger. In particular embodiments, recombinant proteins are produced in animal cells, particularly mammalian cells.Mammalian cell lines that are useful as expression hosts are well known in the art and include, without limitation, immortalized cell lines available from the American Type Culture Collection (ATCC), including, without limitation, Chinese hamster ovary (CHO) cells, including CHOK1 cells (ATCC CCL 61), DXB-11, DG-44, and Chinese hamster ovary cells / -DHFR (CHO, Urlaub et al., Proc. Natl. Acad. Sci. USA 77:4216, 1980); monkey kidney CV1 line transformed by SV40 (COS-7, ATCC CRL 1651); human embryonic kidney line (293 cells, or subclones thereof, e.g., 293 cells grown in suspension culture (Graham et al., J. Gen Virol. 36:59, 1977); baby hamster kidney cells (BHK, ATCC CCL 10); mouse Sertoli cells (TM4, Mather, Biol. Reprod. 23:243-251, 1980); monkey kidney cells (CV1 ATCC CCL 70); African green monkey kidney cells (VERO-76, ATCC CRL-1587); human cervical carcinoma cells (HELA, ATCC CCL 2); canine kidney cells (MDCK, ATCC CCL 34); buffalo rat liver cells (BRL 3A, ATCC CRL 1442); human lung cells (W138, ATCC CCL 75); human hepatoma cells (Hep G2, HB 8065); mouse mammary tumor (MMT 060562, ATCC CCL51); TRI cells (Mather et al., Annals N.Y Acad. Sci. 383:44-68, 1982); MRC 5 cells or FS4 cells; mammalian

[0065] The methods of the present application can be used to purify or isolate a target protein (e.g., a protein comprising an Fc region) from one or more impurities in solution. As used herein, "purifying a protein" refers to a method that reduces the amount of material that is not the target protein and desirably excluded from the final protein composition. These impurities or contaminants can include proteins (e.g., soluble or insoluble proteins, or protein fragments, including undesired fragments of the protein of interest, such as half-antibodies), lipids (e.g., cell wall material), endotoxins, viruses, nucleic acids (e.g., chromosomal or extrachromosomal DNA, t-RNA, rRNA, or mRNA), or combinations thereof, or any other material that is not the target protein of interest comprising an Fc region. In some embodiments, the impurities or contaminants can be from the host cell that produces the protein of interest comprising an Fc region. For example, in some embodiments, the impurities or contaminants are host cell proteins, host cell nucleic acids (DNA or RNA), or other cellular components of the prokaryotic or eukaryotic host cell that expresses the protein of interest comprising an Fc region. In some embodiments, the impurities or contaminants are not derived from the host cell, e.g., the impurities or contaminants can be proteins or other materials from the cell culture or growth media, buffers or media additives. In other embodiments, the impurities or contaminants can be undesired forms of the protein comprising an Fc region, such as proteolytic fragments or unassembled components (e.g., light chains, heavy chains, half-molecules, or fragments thereof) of the protein. The term "impurities" as used herein can include a single undesired material, or a combination of several undesired materials. Suitable methods for detecting contaminating proteins and nucleic acids (e.g., host cell proteins and nucleic acids) are known to those of skill in the art. Such methods include, but are not limited to, enzyme-linked immunosorbent assays (ELISAs), gel electrophoretic methods, and quantitative polymerase chain reaction methods. Size exclusion high performance liquid chromatography and capillary electrophoresis methods can be used to measure high molecular weight species (e.g., aggregates) and low molecular weight species (fragments, unassembled components) of the protein comprising an Fc region.

[0066] The solution containing a protein having a target Fc region that can be purified can be any solution containing the protein and one or more impurities or contaminants whose presence is undesirable. The solution containing the protein and one or more impurities can include any solution derived from a cell culture in which the target protein has been produced. For example, when a host cell that has been modified by recombinant methods to produce a protein is cultured, the protein having an Fc region can be produced intracellularly, in the periplasmic space, or directly secreted into the culture medium. If the protein is produced intracellularly, the cells can be lysed according to methods known to those of skill in the art to produce a cell lysate containing the protein having an Fc region. Thus, in one embodiment, the solution comprising the protein having an Fc region to be purified is a cell culture lysate. In some embodiments, prior to the purification methods described herein, host cells, lysed fragments, and other large particulates can be removed from the cell culture lysate, for example, by centrifugation, microfiltration, or ultrafiltration. In other embodiments, the recombinant protein is secreted by the host cell into the culture medium. In such embodiments, the recombinant host cells and other particulate matter can be separated from the cell culture medium containing the protein having an Fc region, for example, by tangential flow filtration or centrifugation to produce a cell culture supernatant. Thus, in some embodiments, the solution containing the protein having an Fc region to be purified is a cell culture supernatant. The cell culture lysate, cell culture supernatant, or other solution containing the protein having an Fc region can be further clarified to remove fine particulate matter and soluble aggregates prior to the purification methods of the present application. In some embodiments, clarification of the solution can be accomplished by filtering the solution with a membrane having a pore size of between about 0.1 μιη to about 0.5 μιη, preferably a membrane having a pore size of about 0.22 μιη.

[0067] In certain embodiments, the solution containing the protein having an Fc region and one or more impurities is a harvest stream or pool from a bioreactor in which host cells expressing the protein are being cultured. By "harvest stream" or "harvest pool" is meant a solution that has been processed by one or more operations to separate cells, cell fragments, or other large particulates from the protein having an Fc region. Such operations include standard operations known to those of skill in the art for harvesting recombinant proteins from host cell cultures, such as flocculation, centrifugation, and various forms of filtration (e.g., depth filtration, tangential flow microfiltration, and tangential flow ultrafiltration). In some embodiments, the solution containing the protein and one or more impurities is a harvest stream or pool from an industrial scale bioreactor (e.g., a production bioreactor). Industrial scale bioreactors typically produce recombinant proteins in volumes exceeding 500 liters, particularly 2,000 liters to 20,000 liters. In a particular embodiment, the solution containing one or more impurities and the protein to be purified according to the methods of the present application is a harvest stream or pool from a production bioreactor having a volume of about 2,000 liters or greater.

[0068] In certain embodiments, the methods of the present application comprise contacting a solution containing a protein to be purified with a temperature-responsive Protein A material. By "temperature-responsive Protein A material" is meant a mutant form of Protein A, a cell wall protein found in strains of Staphylococcus aureus, that has been altered such that its ability to bind the Fc region of a protein varies with temperature. Such temperature-responsive Protein A mutants are described in U.S. Patent No. 8,198,409, which is incorporated by reference herein in its entirety. In some embodiments, the temperature-responsive Protein A material comprises a mutant Protein A that has different Fc region binding ability at a low temperature, e.g., about 0 °C to about 15 °C, than at a higher temperature, e.g., about 35 °C or greater. In certain embodiments, the temperature-responsive Protein A material comprises a mutant Protein A comprising an amino sequence of any of the sequences listed in Table 1 below. In one embodiment, the mutant Protein A comprises the amino acid sequence of SEQ ID NO: 1. In another embodiment, the mutant Protein A comprises the amino acid sequence of SEQ ID NO: 2. In another embodiment, the mutant Protein A comprises the amino acid sequence of SEQ ID NO: 4. In still another embodiment, the mutant Protein A comprises the amino acid sequence of SEQ ID NO: 6.

[0069] Table 1. Amino acid sequences of mutant Protein A proteins

[0070]

[0071]

[0072] The temperature-responsive Protein A material can be prepared synthetically, e.g., by peptide synthesis or recombinant techniques. Alternatively, the temperature-responsive Protein A material can be obtained commercially, e.g., from Nomadic Bioscience Co., Ltd. (Byzen The temperature-responsive Protein A material is immobilized on a solid phase. The solid phase can include, but is not limited to, beads, resins, gels, particles, membranes, tubes, plates, and films. In one embodiment, the solid phase on which the temperature-responsive Protein A is immobilized is a bead, particularly a magnetic bead. In another embodiment, the solid phase on which the temperature-responsive Protein A is immobilized is a resin. In certain embodiments, the temperature-responsive Protein A material is immobilized to a bead, resin, particle, or other solid phase suitable for packing into a container, and the solid phase containing the temperature-responsive Protein A material is packed or loaded into a container, e.g., a column.

[0073] Suitable materials that can be made into a solid phase include glass, silica (e.g., silica gel), polysaccharides (e.g., polysaccharide matrices), such as agarose, dextran, and cellulose, organic polymers, such as polyacrylamide, polymethyl methacrylate, and styrene-divinylbenzene copolymer. Methods of immobilizing temperature-responsive Protein A materials to various solid phases are known to those of skill in the art and can include activating the solid phase material with a functional coupling group (e.g., a carboxyl or thiol group) and other methods described in U.S. Patent Publication Nos. 2015 / 0218208 and 2013 / 0317172.

[0074] As used herein, contacting a solution comprising a protein comprising an Fc region with a temperature-responsive Protein A material or adsorbing a protein comprising an Fc region to a temperature-responsive Protein A material means combining the protein with the temperature-responsive Protein A material under conditions in which the protein binds to the material. In particular, the protein comprising an Fc region binds to the temperature-responsive Protein A material at the temperature of the protein-binding material, e.g., at a temperature from about 0 °C to about 15 °C, from about 1 °C to about 12 °C, or from about 2 °C to about 8 °C. In some embodiments, the protein comprising an Fc region is contacted with or adsorbed to the temperature-responsive Protein A material at a temperature of about 10 °C or less. In certain embodiments, the protein comprising an Fc region is contacted with or adsorbed to the temperature-responsive Protein A material at a temperature of about 1 °C to about 6 °C. In one embodiment, the protein comprising an Fc region is contacted with or adsorbed to the temperature-responsive Protein A material at a temperature of about 4 °C.

[0075] In certain embodiments, the temperature-responsive Protein A material can be equilibrated with a suitable buffer prior to contact with a solution comprising a protein comprising an Fc region to be purified. One such suitable equilibration buffer is phosphate buffered saline at pH 7.2. Other suitable equilibration buffers include Tris, BIS, and HEPES at a concentration from about 0.5 mM to about 100 mM, the equilibration buffer comprising a physiological salt concentration (e.g., 150 mM NaCl) at a pH of about 5 to about 9, preferably at a pH of about 6.0 to about 8.0, and more preferably at a pH of about 6.5 to about 7.5.

[0076] Once the Fc region-containing protein is bound to the temperature-responsive Protein A material, the bound material can be optionally washed with one or more wash solutions prior to elution from the material. The one or more wash solutions are typically buffers of neutral pH (e.g., about 6.5 to about 7.5) that include a salt, such as sodium acetate, sodium citrate, or sodium chloride. Suitable salt concentrations in the wash solutions are from about 0.1 M to about 2 M, about 0.5 M to about 2 M, about 0.75 M to about 1.5 M, or about 0.2 M to about 0.6 M. In certain embodiments, the one or more wash solutions include sodium chloride at a concentration of about 0.5 M to about 2 M. In some embodiments, the one or more wash solutions include sodium chloride at a concentration of about 0.1 M to about 0.8 M. In one embodiment, the one or more wash solutions include about 10 to about 25 mM phosphate buffer and about 0.1 M to about 0.8 M NaCl at a pH of about 7 to about 7.5. In another embodiment, the one or more wash solutions include about 20 to about 50 mM Tris buffer and about 0.1 M to about 0.8 M NaCl at a pH of about 7 to about 7.5.

[0077] The one or more wash buffers can also include additional components that aid in the removal of impurities from the temperature-responsive Protein A material without significantly affecting the binding interaction of the Fc region-containing protein and the temperature-responsive Protein A material. These additional components can include divalent cations (e.g., calcium, magnesium, and nickel), detergents (e.g., polysorbate 20 or polysorbate 80), or polymers (e.g., polyethylene glycol). In certain embodiments, the temperature of the one or more wash solutions is the same as the temperature at which the Fc region-containing protein is adsorbed to the temperature-responsive Protein A material (e.g., 0 °C to 15 °C). In other embodiments, the temperature of the one or more wash solutions is about 15 °C to 25 °C. In still other embodiments, the temperature of the one or more wash solutions is no more than 25 °C, i.e., the temperature of the one or more wash solutions is 25 °C or less, such as between about 1 °C to about 25 °C.

[0078] Once the Fc region-containing protein is bound to the temperature-responsive Protein A material and the bound material is optionally washed as described above, the Fc region-containing protein is removed from the temperature-responsive Protein A material using an elution buffer as described above. Typically, elution of the Fc region-containing protein from temperature-responsive Protein A resin requires the temperature of the resin to be raised to 35°C or higher. See U.S. Patent No. 8,198,409 and Koguma et al., Journal of Chromatography A [Journal of Chromatography A], Vol. 1305: 149-153, 2013. However, the composition of the elution buffers described in detail below allows for the removal or elution of proteins bound to temperature-responsive Protein A resin from the resin at neutral pH without raising the temperature of the resin above 35°C. Thus, in certain embodiments, the methods of the application include eluting the bound Fc region-containing protein from the temperature-responsive Protein A material at a temperature of less than about 35°C. For example, in some embodiments, the protein is eluted from the temperature-responsive Protein A material at a temperature of from about 1°C to about 34°C, from about 4°C to about 32°C, from about 10°C to about 30°C, or from about 15°C to about 25°C. In certain embodiments, the protein is eluted from the temperature-responsive Protein A material at a temperature of less than about 30°C, for example, from about 1°C to about 25°C. In some embodiments, the protein is eluted from the temperature-responsive Protein A material at a temperature of from about 20°C to about 25°C. In other embodiments, the protein is eluted from the temperature-responsive Protein A material at a temperature of from about 15°C to about 22°C. In certain embodiments, the protein is eluted from the temperature-responsive Protein A material at a temperature of less than about 10°C, for example, from about 1°C to about 8°C. In some embodiments, the protein is eluted from the temperature-responsive Protein A material at a temperature of from about 1°C to about 6°C. In other embodiments, the protein is eluted from the temperature-responsive Protein A material at a temperature of from about 2°C to about 8°C. In some embodiments, the protein is eluted from the temperature-responsive Protein A material at the same temperature at which the protein was bound to the material (i.e., the temperature of the material is not changed between the adsorption and elution steps).

[0079] In some embodiments, the Fc region-containing protein can be eluted with the elution buffers detailed below isocratically with the temperature-responsive Protein A material. In alternative embodiments, the Fc region-containing protein can be eluted from the temperature-responsive Protein A material with a linear gradient, for example, starting with a 100% solution having a similar composition as the wash solutions described herein and ending with a 100% of the elution buffers detailed below.

[0080] The elution buffer employed to remove the Fc region-containing protein from the temperature-responsive protein A material is typically a buffered solution having a pH of about 6.5 to about 7.5. In some embodiments, the elution buffer has a pH of about 6.8 to about 7.5. In other embodiments, the elution buffer has a pH of about 7.2 to about 7.5. In a particular embodiment, the elution buffer has a pH of about 7.0 to about 7.4. Any buffer can be used provided that the buffer is capable of maintaining the pH of the solution within the target pH range (e.g., a buffer having a pKa value between 6 and 8). Suitable buffers that are buffered in the neutral pH range can be used as components of the elution buffer in the methods of the present application, including, but not limited to, HEPES (N-[2-hydroxyethyl]piperazine-N'-[2-ethanesulfonic acid]), Tris, phosphate, citrate, MES (2-(N-morpholino)ethanesulfonic acid), BES (N,N-bis[2-hydroxyethyl]-2-aminoethanesulfonic acid), PIPES (piperazine-N,N-bis(2-ethanesulfonic acid)), Tricine (N-tris[hydroxymethyl]methylglycine), diglycine (N,N-bis(2-hydroxyethyl)glycine), TES (N-tris[hydroxymethyl]methyl-2-aminoethanesulfonic acid), TAPSO (3-[N-tris(hydroxymethyl)methylamino]-2-hydroxypropanesulfonic acid), Bis-Tris (bis(2-hydroxyethyl)amino-tris(hydroxymethyl)methane), and MOPS (3-[N-morpholino]propanesulfonic acid). The buffer can be present at a concentration of from about 5 mM to about 200 mM, from about 10 mM to about 150 mM, from about 15 mM to about 100 mM, from about 20 mM to about 75 mM, or from about 25 mM to about 50 mM. In some embodiments, the elution buffer comprises a HEPES buffer, for example, at a concentration of about 15 mM to about 100 mM. In other embodiments, the elution buffer comprises a Tris buffer, for example, at a concentration of about 15 mM to about 50 mM.

[0081] In various embodiments, the elution buffer comprises a chaotropic agent. A "chaotropic agent" is a substance that disrupts the hydrogen-bond network of water and can lower the order of a macromolecular structure by affecting intramolecular interactions mediated by non-covalent forces such as hydrogen bonds, van der Waals forces, and hydrophobic interactions. Without being bound by theory, it is believed that the presence of a chaotropic agent in the elution buffer serves to relax the structure of the Fc region-containing protein to facilitate its disengagement from the temperature-responsive Protein A material. Suitable chaotropic agents that can be included in the elution buffer include, but are not limited to, butanol, ethanol, propanol, guanidinium chloride, lithium acetate or lithium perchlorate, magnesium chloride, phenol, sodium dodecyl sulfate, urea, thiourea, and thiocyanate (e.g., sodium thiocyanate, ammonium thiocyanate, or potassium thiocyanate). In certain embodiments, the elution buffer comprises urea, guanidinium chloride, or sodium dodecyl sulfate as a chaotropic agent. In other embodiments, the elution buffer comprises urea, guanidinium chloride, or thiocyanate as a chaotropic agent. In still other embodiments, the elution buffer comprises urea or guanidinium chloride as a chaotropic agent. The chaotropic agent can be present in the elution buffer in a concentration of from about 0.4 M to about 5.0 M, from about 0.5 M to about 2.5 M, from about 0.8 M to about 1.2 M, from about 2 M to about 4.5 M, or from about 3 M to about 4.2 M, depending on the particular chaotropic agent used. In certain embodiments, the elution buffer comprises urea, for example, in a concentration of from about 2 M to about 4.5 M.

[0082] In other embodiments, the elution buffer comprises urea in a concentration of from about 3 M to about 4.2 M. In a particular embodiment, the elution buffer comprises urea in a concentration of about 4 M. In certain embodiments, the elution buffer comprises guanidinium chloride, for example, in a concentration of from about 0.5 M to about 2.5 M. In other embodiments, the elution buffer comprises guanidinium chloride in a concentration of from about 0.8 M to about 1.2 M. In certain embodiments, the elution buffer comprises guanidinium chloride in a concentration of about 1 M. In other particular embodiments, the elution buffer comprises guanidinium chloride in a concentration of about 2 M.

[0083] In certain embodiments, the elution buffer comprises a sugar alcohol in addition to the chaotropic agent. A "sugar alcohol" is an organic compound derived from a sugar and has the general formula HOCH2(CHOH)nCH2OH, where n is an integer from 1 to 5. In certain embodiments, the elution buffer comprises a sugar alcohol in a concentration of from about 0.1 M to about 1.0 M. In other embodiments, the elution buffer comprises a sugar alcohol in a concentration of from about 0.2 M to about 0.8 M. In certain embodiments, the elution buffer comprises a sugar alcohol in a concentration of about 0.5 M. In other embodiments, the elution buffer comprises a sugar alcohol in a concentration of about 0.4 M. In still other embodiments, the elution buffer comprises a sugar alcohol in a concentration of about 0.6 M. nCH2OH, wherein n typically varies from 1 to 22 or greater. Exemplary sugar alcohols that can be included in the elution buffer include, but are not limited to, glycerol, erythritol, threitol, arabitol, xylitol, ribitol, mannitol, sorbitol, galactitol, fucitol, iditol, volemitol, isomaltitol, maltitol, lactitol, maltotriitol, and maltotetraitol. In certain embodiments, the elution buffer comprises sorbitol, mannitol, xylitol, or glycerol as the sugar alcohol. In one embodiment, the sugar alcohol in the elution buffer is sorbitol. In another embodiment, the sugar alcohol in the elution buffer is mannitol. Depending on the particular sugar alcohol selected, the sugar alcohol can be present in the elution buffer at a concentration from about 1 M to about 4.5 M, from about 1.5 M to about 4 M, or from about 2 M to about 2.5 M. In certain embodiments, the elution buffer comprises sorbitol, for example, at a concentration of about 1 M to about 4.5 M, more preferably about 2 M to about 2.5 M. In one embodiment, the elution buffer comprises sorbitol at a concentration of about 2.2 M.

[0084] Also, without being bound by theory, it is believed that the presence of the sugar alcohol prevents the Fc region-containing protein from completely unfolding in the presence of the chaotropic agent. Thus, in some embodiments, the elution buffer comprises a particular concentration ratio of the chaotropic agent and the sugar alcohol that is balanced between relaxing the structure of the Fc region-containing protein to disengage it from the temperature-responsive Protein A material, but preventing the protein from completely unfolding and losing its essential native structure. In such embodiments, the molar concentration ratio of the chaotropic agent to the sugar alcohol is about 0.4 to about 4.5, about 0.8 to about 4, about 1 to about 2, about 1.5 to about 2.5, or about 1.8 to about 2.2. In certain embodiments, the elution buffer comprises urea and sorbitol, wherein the molar concentration ratio of the urea to the sorbitol is about 1 to about 2.5, or more preferably about 1.1 to about 1.8. In other embodiments, the elution buffer comprises guanidinium chloride and sorbitol, wherein the molar concentration ratio of the guanidinium chloride to the sorbitol is about 0.5 to about 1.5, or more preferably about 0.5 to about 0.9.

[0085] In certain embodiments, the elution buffer can further comprise one or more amino acids. For example, in some embodiments, the elution buffer can further comprise a basic amino acid, a non-polar amino acid, or both a basic amino acid and a non-polar amino acid. Without being bound by theory, it is believed that basic amino acids facilitate the dissociation of Fc region-containing proteins from temperature-responsive Protein A materials by modulating the electrostatic interactions of the proteins with the material, while non-polar amino acids facilitate the detachment of Fc region-containing proteins from temperature-responsive Protein A materials by modulating the hydrophobic interactions of the proteins with the material. As used herein, a "basic amino acid" is a polar amino acid in either the D or L form that is hydrophilic and positively charged at pH values below its pKa. Exemplary basic amino acids suitable for use in the elution buffer include, but are not limited to, arginine, ornithine, lysine, and histidine. In some embodiments, the elution buffer comprises arginine. In other embodiments, the elution buffer comprises lysine. The basic amino acid can be present in the elution buffer at a concentration of from about 0.1 M to about 1.5 M, from about 0.25 M to about 1 M, or from about 0.3 M to about 0.8 M. In certain embodiments, the elution buffer comprises a basic amino acid (e.g., arginine) at a concentration of about 0.5 M.

[0086] A "non-polar amino acid," as used interchangeably herein with "non-polar amino acid," refers to an amino acid in either the D or L form that contains a hydrophobic functional group and is uncharged at neutral pH. Exemplary non-polar amino acids suitable for use in the elution buffer include, but are not limited to, alanine, cysteine, glycine, isoleucine, leucine, methionine, phenylalanine, proline, tryptophan, and valine. In certain embodiments, the elution buffer comprises proline. The non-polar amino acid can be present in the elution buffer at a concentration of from about 0.1 M to about 1.5 M, from about 0.25 M to about 1 M, or from about 0.3 M to about 0.8 M. In certain embodiments, the elution buffer comprises a non-polar amino acid (e.g., proline) at a concentration of about 0.5 M. In some embodiments, the elution buffer comprises at least one basic amino acid and at least one non-polar amino acid. In such embodiments, the basic amino acid and the non-polar amino acid can be present in the elution buffer at the same concentration. For example, the basic amino acid and the non-polar amino acid can be present in the elution buffer at a concentration of from about 0.25 M to about 1 M, more preferably from about 0.3 M to about 0.8 M. In certain embodiments, the basic amino acid and the non-polar amino acid are each present in the elution buffer at a concentration of about 0.5 M. In one embodiment, the elution buffer comprises arginine and proline. In another embodiment, the elution buffer comprises lysine and proline.

[0087] In some embodiments, the elution buffer employed in the methods of the application can further comprise one or more salts. A "salt" refers to an ionic compound resulting from the neutralization reaction of an acid and a base. A salt typically consists of equal numbers of cations and anions, such that the overall net charge of the salt is zero. Suitable salts for inclusion in the elution buffer include, but are not limited to, sodium salts, such as sodium acetate, sodium citrate, sodium chloride, and sodium sulfate; potassium salts, such as potassium acetate, potassium citrate, potassium chloride, and potassium sulfate; and chloride salts, such as sodium chloride, magnesium chloride, nickel chloride, potassium chloride, and ammonium chloride. In certain embodiments, the elution buffer comprises sodium chloride. In other embodiments, the elution buffer comprises potassium chloride. The salt can be included in the elution buffer at a concentration of from about 0.1 M to about 1 M, from about 0.25 M to about 0.8 M, from about 0.5 M to about 1 M, or from about 0.5 M to about 0.8 M. In one embodiment, the salt (e.g., sodium chloride) is present in the elution buffer at a concentration of about 0.75 M.

[0088] In certain embodiments, the elution buffer used in the methods of the application has a pH of about 6.5 to about 7.5 and comprises about 5 mM to about 200 mM buffer, about 0.4 M to about 5 M chaotropic agent, about 1 M to about 4.5 M sugar alcohol, about 0.1 M to about 1.5 M non-polar amino acid, about 0.1 M to about 1.5 M basic amino acid, and about 0.1 M to about 1 M salt. In some embodiments, the elution buffer has a pH of about 7.0 to about 7.4 and comprises about 15 mM to about 100 mM buffer, about 2 M to about 4.5 M chaotropic agent, about 1 M to about 4.5 M sugar alcohol, about 0.25 M to about 1 M non-polar amino acid, about 0.25 M to about 1 M basic amino acid, and about 0.25 M to about 0.8 M salt. In other embodiments, the elution buffer has a pH of about 7.0 to about 7.4 and comprises about 15 mM to about 100 mM buffer, about 0.5 M to about 2.5 M chaotropic agent, about 1 M to about 4.5 M sugar alcohol, about 0.25 M to about 1 M non-polar amino acid, about 0.25 M to about 1 M basic amino acid, and about 0.25 M to about 0.8 M salt. In certain embodiments, the elution buffer has a pH of about 7.0 to about 7.4 and comprises about 15 mM to about 100 mM buffer, about 2 M to about 4.5 M chaotropic agent, about 2 M to about 2.5 M sugar alcohol, about 0.25 M to about 1 M non-polar amino acid, about 0.25 M to about 1 M basic amino acid, and about 0.25 M to about 0.8 M salt. In some embodiments, the elution buffer has a pH of about 7.0 to about 7.4 and comprises about 15 mM to about 100 mM buffer, about 0.5 M to about 2.5 M chaotropic agent, about 2 M to about 2.5 M sugar alcohol, about 0.25 M to about 1 M non-polar amino acid, about 0.25 M to about 1 M basic amino acid, and about 0.25 M to about 0.8 M salt.

[0089] For any of the above elution buffer compositions, the buffer can be HEPES or Tris, the chaotropic agent can be urea or guanidinium chloride, the sugar alcohol can be sorbitol or mannitol, the non-polar amino acid can be proline, the basic amino acid can be arginine or lysine, and the salt can be a sodium salt, such as sodium chloride. For example, in certain embodiments, the elution buffer has a pH of about 6.5 to about 7.5 and comprises 15 mM to about 100 mM HEPES, about 2 M to about 4.5 M urea, about 1 M to about 4.5 M sorbitol, about 0.25 M to about 1 M proline, about 0.25 M to about 1 M arginine, and about 0.25 M to about 0.8 M sodium chloride. In some embodiments, the elution buffer has a pH of about 7.0 to about 7.4 and comprises about 20 mM to about 75 mM HEPES, about 3 M to about 4.2 M urea, about 2 M to about 2.5 M sorbitol, about 0.3 M to about 0.8 M proline, about 0.3 M to about 0.8 M arginine, and about 0.5 M to about 1 M sodium chloride. In one embodiment, the elution buffer has a pH of about 7.2 and comprises about 25 mM HEPES, about 4 M urea, about 2.2 M sorbitol, about 0.5 M proline, about 0.5 M arginine, and about 0.75 M sodium chloride. In another embodiment, the elution buffer has a pH of about 7.2 and comprises about 50 mM HEPES, about 4 M urea, about 2.2 M sorbitol, about 0.5 M proline, about 0.5 M arginine, and about 0.75 M sodium chloride. In some embodiments, the elution buffer has a pH of about 6.5 to about 7.5 and comprises 15 mM to about 100 mM Tris, about 2 M to about 4.5 M urea, about 1 M to about 4.5 M sorbitol, about 0.25 M to about 1 M proline, about 0.25 M to about 1 M arginine, and about 0.25 M to about 0.8 M sodium chloride. In other embodiments, the elution buffer has a pH of about 6.5 to about 7.5 and comprises 15 mM to about 100 mM HEPES, about 2 M to about 4.5 M urea, about 1 M to about 4.5 M mannitol, about 0.25 M to about 1 M proline, about 0.25 M to about 1 M arginine, and about 0.25 M to about 0.8 M sodium chloride. In still other embodiments, the elution buffer has a pH of about 6.5 to about 7.5 and comprises 15 mM to about 100 mM Tris, about 2 M to about 4.5 M urea, about 1 M to about 4.5 M mannitol, about 0.25 M to about 1 M proline, about 0.25 M to about 1 M arginine, and about 0.25 M to about 0.8 M sodium chloride.

[0090] In certain embodiments, the elution buffer has a pH of about 6.5 to about 7.5 and comprises 15 mM to about 100 mM HEPES, about 0.5 M to about 2.5 M guanidine chloride, about 1 M to about 4.5 M sorbitol, about 0.25 M to about 1 M proline, about 0.25 M to about 1 M arginine, and about 0.25 M to about 0.8 M sodium chloride. In some embodiments, the elution buffer has a pH of about 6.5 to about 7.5 and comprises 15 mM to about 100 mM Tris, about 0.5 M to about 2.5 M guanidine chloride, about 1 M to about 4.5 M sorbitol, about 0.25 M to about 1 M proline, about 0.25 M to about 1 M arginine, and about 0.25 M to about 0.8 M sodium chloride. In other embodiments, the elution buffer has a pH of about 6.5 to about 7.5 and comprises 15 mM to about 100 mM HEPES, about 0.5 M to about 2.5 M guanidine chloride, about 1 M to about 4.5 M mannitol, about 0.25 M to about 1 M proline, about 0.25 M to about 1 M arginine, and about 0.25 M to about 0.8 M sodium chloride. In still other embodiments, the elution buffer has a pH of about 6.5 to about 7.5 and comprises 15 mM to about 100 mM Tris, about 0.5 M to about 2.5 M guanidine chloride, about 1 M to about 4.5 M mannitol, about 0.25 M to about 1 M proline, about 0.25 M to about 1 M arginine, and about 0.25 M to about 0.8 M sodium chloride.

[0091] In some embodiments, the methods of the application reduce or eliminate aggregation of the Fc region-containing protein that would otherwise occur during the purification procedure. Thus, the application also includes methods for reducing aggregation during purification of a Fc region-containing protein. In one embodiment, the method comprises adsorbing the Fc region-containing protein to a temperature-responsive Protein A material at a temperature at which the protein binds to the material; and eluting the protein from the material at a temperature of less than about 35°C with an elution buffer having a pH of about 6.5 to about 7.5 and comprising a chaotropic agent, a sugar alcohol, a non-polar amino acid, and a basic amino acid, wherein the amount of the Fc region-containing protein in aggregated form in the eluate is less than the amount in an eluate from a conventional Protein A material.

[0092] As used herein, an "aggregated form" of an Fc region-containing protein refers to a multimeric form of the protein that is composed of multiple molecules or monomers of the protein held together by non-covalent interactions. A "monomeric form" of an Fc region-containing protein refers to a form of the protein that comprises a single, complete molecule of the protein, including all components and chains. For example, an antibody monomer or monomeric form of an antibody is composed of two light chains and two heavy chains connected by disulfide bonds. Similarly, a monomeric form of an IgG-scFv binding protein comprises two light chains and two modified heavy chains connected by disulfide bonds, where each modified heavy chain comprises a single chain variable fragment fused to its carboxy terminus. A monomeric form of an IgG-scFv binding protein is shown in Figure 1 For proteins containing a single chain Fc region, such as the single chain bispecific Fv-Fc binding depicted in Figure 5 A monomeric form is a single polypeptide chain.

[0093] In some embodiments of the purification methods described herein, a majority of the Fc region-containing proteins in the eluate from the temperature-responsive Protein A material are in monomeric form. For example, at least 60%, at least 65%, at least 70%, at least 75%, at least 80%, at least 85%, at least 90%, at least 95%, at least 96%, at least 97%, at least 98%, or at least 99% of the Fc region-containing proteins in the eluate from the temperature-responsive Protein A material eluted with the elution buffer described herein are in monomeric form. In certain embodiments, at least 70% of the Fc region-containing proteins in the eluate from the material are in monomeric form. In some embodiments, at least 80% of the Fc region-containing proteins in the eluate from the material are in monomeric form. In other embodiments, at least 90% of the Fc region-containing proteins in the eluate from the material are in monomeric form.

[0094] In certain embodiments of the purification methods described herein, the amount of aggregated Fc region-containing proteins in the eluate from the temperature-responsive Protein A material using the elution buffers described herein is less than the amount of aggregated Fc region-containing proteins in the eluate from a conventional Protein A material. As used herein, a "conventional Protein A material" refers to native or wild-type Protein A found in S. aureus strains that loses its affinity for protein Fc regions under acidic conditions (e.g., at a pH below about 6). Conventional Protein A materials include commercially available Protein A resins, such as MabSelect SuRe TM resin (GE Healthcare), resins, such as those described in the Examples. In some embodiments, the amount of Fc region-containing protein in aggregated form in the eluate from the temperature-responsive Protein A material using the elution buffer described herein is less than the amount of Fc region-containing protein in aggregated form in the eluate from conventional Protein A material using an acidic elution buffer, such as an acetic acid buffer at a pH of about 2.5 to about 4, e.g., 1% acetic acid.

[0095] The amount of Fc region-containing protein in aggregated form in the eluate from the temperature-responsive Protein A material using the elution buffer described herein is preferably less than 50%, less than 45%, less than 40%, less than 35%, less than 30%, less than 25%, less than 20%, less than 15%, less than 10%, less than 5%, less than 4%, less than 3%, less than 2%, or less than 1%. In certain embodiments, less than 30% of the Fc region-containing protein in the eluate from the temperature-responsive Protein A material is in aggregated form. In some embodiments, less than 20% of the Fc region-containing protein in the eluate from the temperature-responsive Protein A material is in aggregated form. In other embodiments, less than 10% of the Fc region-containing protein in the eluate from the temperature-responsive Protein A material is in aggregated form.

[0096] Methods of detecting and quantifying protein aggregation and monomeric form are known to those of skill in the art and can include size exclusion high performance liquid chromatography methods, such as those described in the Examples. Other suitable methods include sedimentation velocity analytical ultracentrifugation, asymmetrical flow field flow fractionation, and dynamic light scattering. See, e.g., Gabrielson et al., Journal of Pharmaceutical Sciences, Vol. 96: 268-279, 2007.

[0097] In certain embodiments of the methods of the application, the Fc region-containing protein can be subjected to further purification steps after elution from the temperature-responsive Protein A material. For example, in some embodiments, the Fc region-containing protein eluted from the temperature-responsive Protein A material is subjected to one or more additional chromatography steps. These additional chromatography steps can include ion exchange chromatography (e.g., cation exchange chromatography or anion exchange chromatography), hydrophobic interaction chromatography, mixed mode chromatography, size exclusion chromatography (e.g., gel filtration chromatography), hydroxyapatite chromatography, metal affinity chromatography, or a combination thereof. The additional chromatography steps can be performed in bind-and-elute mode, in which the target protein binds to the chromatography material and the impurities flow through, or flow-through mode, in which the impurities bind to the chromatography material and the target protein flows through. In some embodiments, the Fc region-containing protein eluted from the temperature-responsive Protein A material is subjected to cation exchange chromatography. In other embodiments, the Fc region-containing protein eluted from the temperature-responsive Protein A material is subjected to anion exchange chromatography. In certain embodiments, the Fc region-containing protein eluted from the temperature-responsive Protein A material is subjected to hydrophobic interaction chromatography. In still other embodiments, the Fc region-containing protein eluted from the temperature-responsive Protein A material is subjected to mixed mode chromatography. In certain embodiments, the Fc region-containing protein eluted from the temperature-responsive Protein A material is subjected to size exclusion chromatography (e.g., gel filtration chromatography), such as the size exclusion chromatography described in further detail below.

[0098] The chromatography steps of the methods of the application can be followed by additional steps, such as viral inactivation, viral filtration, and / or ultrafiltration / diafiltration steps. For example, in some embodiments, the eluate pool or the effluent from the temperature-responsive Protein A material can be used in a viral inactivation step using a detergent or UV inactivation method. In one embodiment, the Fc region-containing protein eluted from the temperature-responsive Protein A material is subjected to a detergent viral inactivation step. A detergent, such as Triton X-100 (e.g., at a concentration of 1% v / v) can be added to the eluate pool or the effluent from the temperature-responsive Protein A material and incubated at neutral pH for about 30 minutes to about 60 minutes to inactivate any viruses present.

[0099] The present application also provides methods for separating antibodies from their half-antibody forms. As described in Example 5, it was surprisingly found that the elution buffers described herein for eluting Fc region-containing proteins from temperature-responsive Protein A resins at neutral pH and constant temperature can be used as mobile phases in size exclusion chromatography for efficiently separating antibodies from their half-antibody forms. Thus, in certain embodiments, the present application includes methods for separating antibodies from their half-antibody forms, the methods comprising contacting a solution comprising the antibodies and their half-antibody forms with a gel filtration matrix using a mobile phase having a pH of about 6.5 to about 7.5, and comprising a chaotropic agent, a sugar alcohol, a non-polar amino acid, and a basic amino acid; and collecting elution fractions from the gel filtration matrix, wherein the antibodies elute in one set of elution fractions and their half-antibody forms elute in another set of elution fractions, thereby separating the antibodies from their half-antibody forms.

[0100] A "half-antibody" refers to a form of an antibody of interest that typically comprises a single light chain polypeptide and a single heavy chain polypeptide (see Figure 9A and 9B the schematic on the left). Half-antibodies (used interchangeably with "half-molecules") are typically due to incomplete assembly or disruption of interactions between two heavy chain polypeptides of an antibody (e.g., disruption of inter-polypeptide disulfide bond formation between the hinge regions of two heavy chains). In certain embodiments, the antibodies to be separated from their half-antibody forms are multispecific (e.g., bispecific) heterodimeric antibodies. In such embodiments, two half-antibodies can be generated from the recombinant production of a bispecific heterodimeric antibody: one half-antibody that binds a first antigen and another half-antibody that binds a second antigen. The methods of the present application can separate one or both half-antibodies from the fully assembled antibody.

[0101] In some embodiments of the method of isolating an antibody from its half-antibody form, the method comprises contacting a solution containing the antibody to be purified and its half-antibody form with a gel filtration matrix. Gel filtration matrices typically consist of porous beads made of cross-linked polymers, which can be packed in a column or other container. Various types of gel filtration matrices suitable for use in the methods of the application are commercially available, including but not limited to, dextran-based gels, such as SEPHADEX (cross-linked dextran and epichlorohydrin); polyacrylamide-based gels, such as SEPHACRYL (cross-linked copolymer of allyl dextran and N,N'-methylenebisacrylamide); agarose-based gels, such as SUPEROSE (highly cross-linked agarose) or SEPHAROSE (cross-linked agarose); and composite gels prepared from two gels, such as SUPERDEX (cross-linked dextran and agarose). In certain embodiments, the gel filtration matrix comprises cross-linked agarose and dextran. For example, in one embodiment, the gel filtration matrix is a SUPERDEX gel filtration matrix (GE Healthcare), such as SUPERDEX 200 gel filtration matrix. The fractionation range of the gel filtration matrix can be from about 5 kDa to about 5000 kDa, about 10 kDa to about 1500 kDa, or about 10 kDa to about 600 kDa. In some embodiments, the fractionation range of the gel filtration matrix employed in the methods of the application is from about 10 kDa to about 600 kDa.

[0102] The mobile phase for size exclusion chromatography to separate fully assembled antibodies from half-antibodies can be any of the elution buffers described herein for eluting Fc region-containing proteins from temperature-responsive Protein A resins at neutral pH and constant temperature. In certain embodiments, the mobile phase has a pH of about 6.5 to about 7.5 and comprises a chaotropic agent, a sugar alcohol, a non-polar amino acid, and a basic amino acid. The mobile phase is typically a buffered solution having a pH of about 6.5 to about 7.5. In some embodiments, the mobile phase has a pH of about 6.8 to about 7.5. In other embodiments, the mobile phase has a pH of about 7.2 to about 7.5. In one particular embodiment, the mobile phase has a pH of about 7.0 to about 7.4. Suitable buffers and concentrations buffered within this pH range are described in detail above. In some embodiments, the mobile phase comprises a HEPES buffer, such as at a concentration of about 15 mM to about 100 mM. In other embodiments, the mobile phase comprises a Tris buffer, such as at a concentration of about 15 mM to about 50 mM.

[0103] The chaotropic agent used in the mobile phase can be any chaotropic agent described above at any concentration. For example, the chaotropic agent in the mobile phase can be urea, guanidinium chloride, sodium thiocyanate, potassium thiocyanate, or ammonium thiocyanate. In one particular embodiment, the mobile phase comprises urea as the chaotropic agent. In such embodiments, the urea is present in the mobile phase at a concentration of about 2 M to about 4.5 M. In some embodiments, the mobile phase comprises urea at a concentration of about 3 M to about 4.2 M. In other embodiments, the urea is present in the mobile phase at a concentration of about 4 M.

[0104] The mobile phase also preferably comprises a sugar alcohol, which can be any of those described above included in the elution buffer. In some embodiments, the mobile phase comprises sorbitol, mannitol, xylitol, or glycerol as the sugar alcohol. In one embodiment, the sugar alcohol in the mobile phase is sorbitol. In another embodiment, the sugar alcohol in the mobile phase is mannitol. Depending on the particular sugar alcohol selected, the sugar alcohol can be present in the mobile phase at a concentration of from about 1 M to about 4.5 M, from about 1.5 M to about 4 M, or from about 2 M to about 2.5 M. In certain embodiments, the mobile phase comprises sorbitol, for example at a concentration of about 1 M to about 4.5 M, more preferably about 2 M to about 2.5 M. In one embodiment, the mobile phase comprises sorbitol at a concentration of about 2.2 M.

[0105] In certain embodiments, the mobile phase can further comprise one or more amino acids, such as those described above included in the elution buffer of the present application. For example, in some embodiments, the mobile phase can further comprise a basic amino acid, a non-polar amino acid, or both a basic amino acid and a non-polar amino acid. In certain embodiments, the mobile phase comprises a basic amino acid selected from histidine, lysine, ornithine, and arginine. In some embodiments, the mobile phase comprises arginine. In other embodiments, the mobile phase comprises lysine. The basic amino acid can be present in the mobile phase at a concentration of from about 0.1 M to about 1.5 M, from about 0.25 M to about 1 M, or from about 0.3 M to about 0.8 M. In certain embodiments, the mobile phase comprises a basic amino acid (e.g., arginine) at a concentration of about 0.5 M.

[0106] In some embodiments, the mobile phase comprises a non-polar amino acid selected from alanine, cysteine, glycine, isoleucine, leucine, methionine, phenylalanine, proline, tryptophan, and valine. In particular embodiments, the mobile phase comprises proline. The non-polar amino acid can be present in the mobile phase at a concentration from about 0.1 M to about 1.5 M, from about 0.25 M to about 1 M, or from about 0.3 M to about 0.8 M. In certain embodiments, the mobile phase comprises a non-polar amino acid (e.g., proline) at a concentration of about 0.5 M. In some embodiments, the mobile phase comprises at least one basic amino acid and at least one non-polar amino acid. In such embodiments, the basic amino acid and the non-polar amino acid can be present in the mobile phase at the same concentration. For example, the basic amino acid and the non-polar amino acid can be present in the mobile phase at a concentration from about 0.25 M to about 1 M, more preferably from about 0.3 M to about 0.8 M. In certain embodiments, the basic amino acid and the non-polar amino acid are each present in the mobile phase at a concentration of about 0.5 M. In one embodiment, the mobile phase comprises arginine and proline. In another embodiment, the mobile phase comprises lysine and proline.

[0107] In some embodiments, the mobile phase employed in the size-exclusion chromatography-based method to isolate full antibodies from half antibodies can further comprise one or more salts, such as any of those described above that are included in the elution buffer of the present application. In certain embodiments, the mobile phase comprises sodium chloride. In other embodiments, the mobile phase comprises potassium chloride. The salt can be included in the mobile phase at a concentration from about 0.1 M to about 1 M, from about 0.25 M to about 0.8 M, from about 0.5 M to about 1 M, or from about 0.5 M to about 0.8 M. In one embodiment, the salt (e.g., sodium chloride) is present in the mobile phase at a concentration of about 0.75 M.

[0108] In certain embodiments, the mobile phase used in the methods of the application has a pH of about 6.5 to about 7.5 and comprises about 5 mM to about 200 mM buffer, about 0.4 M to about 5 M chaotropic agent, about 1 M to about 4.5 M sugar alcohol, about 0.1 M to about 1.5 M non-polar amino acid, about 0.1 M to about 1.5 M basic amino acid, and about 0.1 M to about 1 M salt. In some embodiments, the mobile phase has a pH of about 7.0 to about 7.4 and comprises about 15 mM to about 100 mM buffer, about 2 M to about 4.5 M chaotropic agent, about 1 M to about 4.5 M sugar alcohol, about 0.25 M to about 1 M non-polar amino acid, about 0.25 M to about 1 M basic amino acid, and about 0.25 M to about 0.8 M salt. In other embodiments, the mobile phase has a pH of about 7.0 to about 7.4 and comprises about 15 mM to about 100 mM buffer, about 2 M to about 4.5 M chaotropic agent, about 2 M to about 2.5 M sugar alcohol, about 0.25 M to about 1 M non-polar amino acid, about 0.25 M to about 1 M basic amino acid, and about 0.25 M to about 0.8 M salt. For any of the above described mobile phase compositions, the buffer can be HEPES or Tris, the chaotropic agent can be urea or guanidinium chloride, the sugar alcohol can be sorbitol or mannitol, the non-polar amino acid can be proline, the basic amino acid can be arginine or lysine, and the salt can be a sodium salt, such as sodium chloride. For example, in certain embodiments, the mobile phase has a pH of about 6.5 to about 7.5 and comprises 15 mM to about 100 mM HEPES, about 2 M to about 4.5 M urea, about 1 M to about 4.5 M sorbitol, about 0.25 M to about 1 M proline, about 0.25 M to about 1 M arginine, and about 0.25 M to about 0.8 M sodium chloride. In some embodiments, the mobile phase has a pH of about 7.0 to about 7.4 and comprises about 20 mM to about 75 mM HEPES, about 3 M to about 4.2 M urea, about 2 M to about 2.5 M sorbitol, about 0.3 M to about 0.8 M proline, about 0.3 M to about 0.8 M arginine, and about 0.5 M to about 1 M sodium chloride. In one embodiment, the mobile phase has a pH of about 7.2 and comprises about 25 mM HEPES, about 4 M urea, about 2.2 M sorbitol, about 0.5 M proline, about 0.5 M arginine, and about 0.75 M sodium chloride. In another embodiment, the mobile phase has a pH of about 7.2 and comprises about 50 mM HEPES, about 4 M urea, about 2.2 M sorbitol, about 0.5 M proline, about 0.5 M arginine, and about 0.75 M sodium chloride.

[0109] In some embodiments, the mobile phase has a pH of about 6.5 to about 7.5 and comprises 15 mM to about 100 mM Tris, about 2 M to about 4.5 M urea, about 1 M to about 4.5 M sorbitol, about 0.25 M to about 1 M proline, about 0.25 M to about 1 M arginine, and about 0.25 M to about 0.8 M sodium chloride. In other embodiments, the mobile phase has a pH of about 6.5 to about 7.5 and comprises 15 mM to about 100 mM HEPES, about 2 M to about 4.5 M urea, about 1 M to about 4.5 M mannitol, about 0.25 M to about 1 M proline, about 0.25 M to about 1 M arginine, and about 0.25 M to about 0.8 M sodium chloride. In still other embodiments, the mobile phase has a pH of about 6.5 to about 7.5 and comprises 15 mM to about 100 mM Tris, about 2 M to about 4.5 M urea, about 1 M to about 4.5 M mannitol, about 0.25 M to about 1 M proline, about 0.25 M to about 1 M arginine, and about 0.25 M to about 0.8 M sodium chloride.

[0110] The flow rate of the mobile phase through the gel filtration matrix can be adjusted to further enhance the separation between the antibody and its half antibody form. As described in Example 5, increasing the flow rate of the mobile phase results in a loss of efficiency of the separation between the fully assembled antibody and the half antibody. Thus, in certain embodiments, a slower flow rate is preferred. In some embodiments, the flow rate of the mobile phase is applied to the gel filtration matrix at a flow rate of about 0.01 ml / min to about 0.2 ml / min. In other embodiments, the flow rate of the mobile phase is applied to the gel filtration matrix at a flow rate of about 0.02 ml / min to about 0.06 ml / min.

[0111] As the solution comprising the antibody and its half antibody form is moved through the gel filtration matrix having the mobile phase described herein, elution fractions are collected. The protein content of the fractions can be monitored using UV absorption, for example, at 280 nm, and elution fractions comprising the fully assembled antibody can be collected, while fractions containing higher molecular weight aggregates and half antibody can be discarded. As shown in Figure 12 Example 5, samples from the elution fractions can be analyzed by SDS-PAGE and / or analytical SE-HPLC to verify the removal of fractions enriched for the fully assembled antibody and the half antibody.

[0112] Size exclusion chromatography-based methods, e.g., gel filtration chromatography-based methods, can be performed after one or more purification procedures or other unit operations to isolate antibodies from half-antibodies. For example, a size exclusion chromatography-based method can be a second or third polishing chromatography in an antibody purification process, particularly for multispecific heterodimeric antibodies. In some embodiments, a size exclusion chromatography-based method is performed after a protein A affinity chromatography purification step. Thus, the solution containing the antibody and its half-antibody form is an eluate pool or effluent stream from the protein A chromatography. The protein A chromatography can be a conventional protein A chromatography. Alternatively, the protein A chromatography can be a protein A chromatography as described herein. Because the elution buffer employed in the method of the application removes proteins containing Fc regions, e.g., antibodies, from the temperature-responsive protein A material, these can be used sequentially with the same composition of the mobile phase used in the gel filtration chromatography-based method. Thus, in certain embodiments, the application provides a method of purifying an antibody, comprising: (i) contacting a solution comprising an antibody and one or more impurities, e.g., half-antibody forms thereof, with a temperature-responsive protein A material at a temperature at which the antibody binds to the material; (ii) eluting the protein from the material with an elution buffer at a temperature of less than about 35°C, the elution buffer having a pH of about 6.5 to about 7.5, and comprising a chaotropic agent, a sugar alcohol, a non-polar amino acid, and a basic amino acid; (iii) contacting the eluate from the temperature-responsive protein A material with a gel filtration matrix using the elution buffer as the mobile phase; and (iv) collecting elution fractions from the gel filtration matrix comprising the antibody. In certain embodiments, the antibody to be purified is a multispecific heterodimeric antibody. The temperature-responsive protein A chromatography step and the gel filtration chromatography step can be operated in a continuous fashion such that the eluate stream from the temperature-responsive protein A chromatography is loaded directly onto the gel filtration matrix without any intervening holding tanks. In some embodiments, a detergent or UV viral inactivation step can be optionally incorporated between the temperature-responsive protein A chromatography step and the gel filtration chromatography step.

[0113] The following examples, including the experiments performed and results achieved, are provided for illustrative purposes only, and should not be construed as limiting the scope of the appended claims.

[0114] Example

[0115] Example 1. Purification of IgG-scFv binding proteins

[0116] This example describes the purification of one type of Fc region containing protein, an IgG-scFv binding protein, using conventional protein A chromatography or the affinity chromatography method of the present application. The IgG-scFv binding protein comprises two single chain variable fragments (scFv), each containing the heavy and light chain variable domains from a first antibody, fused via a peptide linker to the carboxy-terminal end of a second antibody heavy chain. The resulting molecule is a tetravalent binding protein with two antigen binding domains for a first target located on the amino-terminal side of the immunoglobulin Fc region and two antigen binding domains for a second target located on the carboxy-terminal side of the Fc region. The monomeric form of the IgG-scFv is shown in Figure 1

[0117] For comparison purposes, the IgG-scFv binding protein was purified using conventional protein A affinity chromatography. Between 100-250 ml of cell culture media containing cells expressing the IgG-scFv binding protein was centrifuged at low speed (600 rpm) for 15 minutes at 4°C to sediment the cells and cell debris. The resulting clear supernatant was passed through a 0.22 micron filter to remove fine particulates and soluble aggregates. The clear and filtered solution was loaded onto a column containing MabSelect SuRe TM resin (GE Healthcare) at a flow rate of 1 ml / min and a temperature of 4°C. After washing the column with a saline solution containing phosphate buffer with 0.5 M NaCl at pH 7.2, the bound IgG-scFv binding protein was eluted from the conventional protein A resin using a 174 mM (1%) acetic acid solution at pH 2.7 and a temperature of 4°C. Samples of the eluate pool were analyzed by size exclusion- super high performance liquid chromatography (SE-UPLC) using a Superdex 200 analytical gel filtration column. The results of four independent experiments are shown in Figure 2 In one experiment, the protein was eluted with low pH acetic acid buffer without neutralization to pH 7.2 Figure 2 In a second experiment, the protein was eluted with low pH acetic acid buffer and immediately neutralized to pH 7.2 Figure 2 In a third experiment, the protein was eluted with low pH acetic acid buffer and stored for eight weeks in low pH acetic acid buffer without neutralization Figure 2 In a fourth experiment, the protein was eluted with low pH acetic acid buffer and stored for eight weeks in low pH acetic acid buffer and then neutralized to pH 7.2 Figure 2 In a fourth experiment, the protein was eluted with low pH acetic acid buffer and stored for eight weeks in low pH acetic acid buffer and then neutralized to pH 7.2

[0118] ​As illustrated by the SE-UPLC graphs, the IgG-scFv binding proteins have a high tendency to aggregate during low pH elution from a conventional protein A affinity column, as evidenced by the multiple peaks eluting before the monomer peak (indicated by black star) in the chromatogram. Both acid exposure and a pH jump from low pH to neutral pH induce aggregation of the IgG-scFv binding proteins, with the pH jump having a greater side effect than acid exposure alone. Loading and elution of the conventional protein A column also was performed at room temperature, and similar results (data not shown) were obtained. Figure 2

[0119] In a second series of experiments, purification of the IgG-scFv binding proteins using the temperature-responsive protein A resin and a specific elution buffer allowed elution of the binding proteins from the temperature-responsive protein A resin without increasing the temperature of the column above room temperature. To prepare a temperature-responsive protein A (TR-ProA) column, approximately 25 ml of suspended TR-ProA resin (Byzen The column was washed with five column volumes of Solution A (phosphate buffered saline (PBS), pH 7.2). The column was then washed with Solution B (PBS with 0.5 M NaCl), followed by a ten column volume rinse with water. The rinsed column was then washed with fifteen column volumes of elution buffer containing 25 or 50 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea at pH 7.2. The column was then equilibrated with Solution A.

[0120] Various volumes (10 ml - 250 ml) of clarified and filtered media containing the IgG-scFv binding proteins were loaded onto the TR-ProA column at a flow rate of 1 ml / min at 4°C. After washing with ten column volumes of Solution B, the bound IgG-scFv binding proteins were eluted with five to ten column volumes of elution buffer at 4°C or room temperature. After elution, the column was regenerated according to the manufacturer's instructions. Samples from the eluate pool were analyzed by SE-UPLC using a Superdex 200 analytical gel filtration column. The results from three independent experiments in which the elution was performed at 4°C are shown in Figure 3 In one experiment, the protein was eluted with elution buffer at 4°C and stored at 37°C for 7 days prior to SE-UPLC analysis (top graph in Figure 3 In a second experiment, the protein was eluted with elution buffer at 4°C and stored at -80°C for 7 days prior to SE-UPLC analysis (bottom graph in Figure 3 ​(The second figure in the image). In the third experiment, the protein was eluted with elution buffer at 4°C and stored at room temperature for 7 days prior to SE-UPLC analysis. Figure 3 (See the third figure in the table). The results show that IgG-scFv binding proteins can be eluted in monomeric form with TR-ProA resin at low temperature using elution buffer. Aggregation of the binding protein was completely eliminated, as evidenced by the absence of a peak (indicated by a black star) eluted from the gel filtration column before the monomer peak. Furthermore, the elution conditions did not affect the subsequent temperature stability of the eluted IgG-scFv protein. No aggregation or degradation of the eluted protein was observed after storing it at various temperatures for 7 days.

[0121] The TR-ProA column was used to purify IgG-scFv-binding protein under the same loading and elution conditions described above, except that elution was performed at room temperature instead of 4°C. The clarified culture medium sample before loading was analyzed by SDS-PAGE, during which the column flowed through the solution and elution buffer pools. The SDS-PAGE results showed that IgG-scFv-binding protein was enriched in the elution buffer pool, and many contaminating proteins had been removed. Figure 4A The sample in the TR-ProA eluent cell was also analyzed by size exclusion-high performance liquid chromatography (SE-HPLC). Figure 4B As shown in the SE-HPLC chromatogram, 99% of the IgG-scFv binding protein present in the TR-ProA eluent pool was in monomeric form, with no detectable aggregates.

[0122] In summary, the experimental results in this example demonstrate that proteins containing the Fc region, such as multi-chain IgG-scFv binding proteins, can be eluted from temperature-responsive protein A resins at neutral pH, reducing or eliminating protein aggregation typically occurring under the low pH elution conditions of conventional protein A chromatography. Furthermore, the results show that elution buffers containing a dissociative agent, sugar alcohol, and amino acids allow for the elution of Fc-containing proteins from temperature-responsive protein A resins without raising the temperature above 35°C, which is typically required for protein elution from temperature-responsive protein A resins.

[0123] Example 2. Purification of single-stranded bispecific Fv-Fc binding protein

[0124] This example describes the purification of a second type of Fc region containing protein, a single chain bispecific Fv-Fc binding protein, such as those described in WO 2014144722, which is incorporated herein by reference in its entirety, using conventional protein A chromatography or the affinity chromatography method of the present application. The single chain bispecific Fv-Fc binding protein comprises a first scFv fragment containing heavy and light chain variable domains from a first antibody fused to a second scFv fragment containing heavy and light chain variable domains from a second antibody, and an Fc region fused at its N-terminus to the C-terminus of the first scFv fragment via a peptide linker. The monomeric form of the bispecific Fv-Fc binding protein is shown in Figure 5

[0125] In a first series of experiments, the bispecific Fv-Fc binding protein was purified using conventional protein A affinity chromatography. Cell culture media containing cells expressing the bispecific Fv-Fc binding protein was centrifuged at 4°C at 600 rpm for 15 minutes. The supernatant was removed and filtered with a 0.22 micron filter. The clarified cell culture supernatant was then loaded onto a column containing MabSelect SuRe TM Protein A resin (GE Healthcare) and washed according to the method described in Example 1. After washing the column, the bound bispecific Fv-Fc binding protein was eluted from the conventional protein A resin using a 174 mM (1%) acetic acid solution at pH 2.7 or a 33 mM (0.06%) acetic acid solution at pH 3.7 at room temperature. Samples of the eluate pool were analyzed by SE-HPLC.

[0126] Figure 6A The SE-HPLC plot of the eluate pool is shown when a 174 mM (1%) acetic acid solution was used as the elution buffer, while Figure 6B The SE-HPLC plot of the eluate pool is shown when a lower concentration acetic acid solution was used as the elution buffer. Under both elution conditions, a substantial amount of aggregation of the bispecific Fv-Fc binding protein was observed, as evidenced by multiple peaks (indicated by black arrows) with shorter retention times than the peak of the monomeric binding protein (indicated by black star). When a 174 mM (1%) acetic acid solution was used as the elution buffer, only 39% of the bispecific Fv-Fc binding protein was recovered in monomeric form. Although reducing the concentration of acetic acid in the elution buffer improved the recovery of the monomeric form of the binding protein, only 53% of the binding protein in the eluate pool was in monomeric form and significant aggregation was still observed. Thus, the bispecific Fv-Fc binding protein is particularly susceptible to aggregation under the typical low pH elution conditions required for conventional protein A chromatography.

[0127] ​In a second series of experiments, temperature-responsive Protein A resin (Byzen Bioscience Co., LTD.) and elution buffers containing chaotropes, sugar alcohols, non-polar amino acids, and basic amino acids as described in Example 1 were used to purify bispecific Fv-Fc binding proteins. This elution buffer allowed for elution of the binding proteins from the temperature-responsive Protein A (TR-ProA) resin without increasing the temperature of the column, a step typically required for elution of proteins from temperature-responsive Protein A resin. Specifically, clarified cell culture supernatant containing bispecific Fv-Fc binding proteins was loaded onto a TR-ProA column at a flow rate of 1 ml / min and a temperature of 4°C. After washing with ten column volumes of PBS containing 0.5 M NaCl, the bound bispecific Fv-Fc binding proteins were eluted with five to ten column volumes of Elution Buffer 1 or Elution Buffer 2 at room temperature. Elution Buffer 1 contained 25 mM HEPES, pH 7.2, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 2.5 M urea. Elution Buffer 2 contained 25 mM HEPES, pH 7.2, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea. Samples from the eluate pool were analyzed by SE-HPLC and SDS-PAGE.

[0128] Figure 7A SE-HPLC profile of the eluate pool obtained from elution of the binding proteins with Elution Buffer 1 at room temperature is shown, while Figure 7B SE-HPLC profile of the eluate pool obtained from elution of the binding proteins with Elution Buffer 2 at room temperature is shown. As can be seen from both figures, the bispecific Fv-Fc binding proteins can be eluted from the TR-ProA resin in essentially monomeric form with elution buffers having temperatures of less than 30°C, similar to the results obtained with the IgG-scFv binding proteins described in Example 1. In both elution conditions, the aggregation of the Fv-Fc binding proteins was significantly reduced compared to the aggregation obtained from elution from a conventional Protein A chromatography column with a low pH buffer (compare Figure 6A and 6B the peaks indicated by the black arrows in Figure 7A and 7B ). In addition, compare the figures in Figure 7A to Figure 7BFigure 2 in the Examples section, shows that increasing the concentration of the chaotropic agent (e.g., urea) in the elution buffer from 2.5 M to 4 M increased the percentage of monomeric Fv-Fc binding protein recovered in the eluate pool from 75% to 88% and further reduced the amount of aggregated binding protein present in the eluate pool. The results of SDS-PAGE analysis of samples before, during, and after purification of a bispecific Fv-Fc binding protein using TR-ProA resin and elution buffer 2 are shown in Figure 2 in the Examples section. The results show that the Fv-Fc binding protein was enriched in the eluate pool. Figure 7C

[0129] The results of the experiments described in this Example show that single chain Fc fusion proteins, such as bispecific Fv-Fc binding proteins, have a tendency to aggregate under the low pH conditions required to elute the binding protein from conventional protein A resin. By employing a temperature-responsive protein A resin and elution buffers containing chaotropic agents, sugar alcohols, non-polar amino acids, and basic amino acids, the aggregation of such binding proteins is significantly reduced. Importantly, the composition of the elution buffer enables the binding protein to be removed from the temperature-responsive protein A resin in substantially monomeric form without increasing the temperature above 35°C.

[0130] Example 3. Buffers for eluting Fc region-containing proteins from temperature-responsive protein A resin

[0131] The experiments described in this Example were designed to explore the ability of different elution buffers to remove bound Fc region-containing proteins from temperature-responsive protein A (TR-ProA) resin without increasing the temperature of the resin above room temperature. Cell clarified cell culture supernatant expressing IgG-scFv binding proteins as described in Example 1 or single chain bispecific Fv-Fc binding proteins as described in Example 2 was loaded onto a temperature-responsive protein A resin column (Byzen Nomadic Bioscience Co., LTD.) at 4°C and washed with a PBS solution containing 0.5 M NaCl. One of the elution buffers listed in Table 2 below was used to elute the bound binding protein from the column at 4°C. The percentage of binding protein recovered as monomer in the eluate pool was determined using SE-HPLC. The results are shown in Table 2 below.

[0132] Table 2. Recovery of monomeric binding protein after elution from TR-ProA resin with different elution buffers

[0133]

[0134]

[0135] ​The results show that inclusion of a chaotropic agent (e.g., urea) significantly enhances elution and recovery of the monomeric form of the bound protein.

[0136] Example 4. Purification of Monoclonal Antibodies

[0137] This example describes purification of a monoclonal antibody using a temperature-responsive Protein A resin. Clarified cell culture supernatant from cells expressing the antibody was loaded onto a temperature-responsive Protein A resin column (Byzen at 4°C with an elution buffer comprising 20 mM HEPES, 2 M guanidinium chloride, and 2.2 M sorbitol and having a pH of 7.2. A sample of the eluate pool was analyzed by SE-HPLC, and the results are shown in Figure 8A and 8B Monoclonal antibodies can be eluted from temperature-responsive Protein A resin at low temperatures using an elution buffer comprising only a chaotropic agent (e.g., guanidinium chloride) and a sugar alcohol (sorbitol) at a molar ratio of about 0.9. Nearly 90% of the antibody was recovered in monomeric form in the eluate pool.

[0138] Example 5. Separation of Whole Antibodies from Half Antibodies

[0139] Half antibodies are formed due to incomplete assembly or disruption of the interaction between two heavy chain polypeptides of an antibody (e.g., disruption of the inter-polypeptide disulfide bond between the hinge regions of the two heavy chains). Half antibodies typically consist of a single light chain polypeptide and a single heavy chain polypeptide. It is important to remove half antibodies from preparations containing desired whole antibodies for several reasons. The presence of half antibodies reduces the concentration of the whole antibody product, reduces dose reproducibility, and reduces the homogeneity of the product. In addition, half antibodies can compete with whole antibodies for binding to a target and can reduce the efficacy of the whole antibody.

[0140] Separation of half antibodies from whole antibodies can be challenging because half antibodies have similar properties to whole antibodies. For example, both half and whole antibodies contain similar Fc regions and thus cannot be effectively separated using Protein A affinity chromatography. Half antibodies have similar isoelectric points, axial ratios, and hydrodynamic radii to whole antibodies and thus separation methods based on these characteristics are often unsuitable. In addition, half antibodies tend to self-associate and bind to whole antibodies, making their removal more difficult.

[0141] The presence of half antibodies is often observed in recombinant preparations of antibodies, particularly heterodimeric antibodies, even after purification by conventional Protein A affinity chromatography Figure 9A and 9B). Heterodimeric antibodies are antibodies comprising light and heavy chains from a first antibody that binds a first target and light and heavy chains from a second antibody that binds a second target. Thus, two half antibodies can be produced from the recombination of different batches of these bispecific heterodimeric antibodies: one half antibody that binds the first target and the other half antibody that binds the second target. The presence and level of half antibodies can vary from batch to batch of recombination of heterodimeric antibodies Figure 9A and 9B ), and due to the number and nature of steps required to remove the half antibodies, the overall yield of fully assembled heterodimeric antibodies can be very low (data not shown).

[0142] Cell culture media containing cells expressing bispecific heterodimeric antibodies ("heterodimeric antibodies A") were centrifuged at low speed (600 rpm) for 15 minutes at 4°C to sediment the cells and cell debris. The resulting clear supernatant was passed through a 0.22 micron filter to remove fine particulates and soluble aggregates. The clear and filtered solution was loaded onto a column containing MabSelect SuRe TM resin (GE Healthcare) at a flow rate of 1 ml / min and a temperature of 4°C. After washing the column with a saline solution containing phosphate buffer with 0.5 M NaCl at pH 7.2, the bound antibodies were eluted from the conventional Protein A resin using a 174 mM (1%) acetic acid solution at pH 2.7 and a temperature of 4°C. Samples of the eluate pool were analyzed by SE-HPLC Figure 10A ) and SDS-PAGE Figure 10B ) as shown by the SE-HPLC and SDS-PAGE analysis, the Protein A eluate pool contained a significant amount (about 72%) of half antibodies.

[0143] To remove the half antibodies remaining in the Protein A eluate pool, a preparative size exclusion chromatography (SEC) step operating under conventional conditions was evaluated. Specifically, the Protein A eluate pool was loaded onto a Superdex 200 preparative gel filtration column using a mobile phase comprising phosphate buffered saline at pH 7.0 at a flow rate of 1 ml / min. As shown in Figure 11 , SEC operating under these conditions was not able to separate the fully assembled heterodimeric antibodies from the half antibodies. Reducing the flow rate through the preparative SEC column containing PBS mobile phase to rates as low as 0.01 ml / min did not improve the separation (data not shown).

[0144] The experiment was repeated but the regular mobile phase containing PBS was replaced with a mobile phase having a similar composition as the unique elution buffer described in Example 1. A Superdex 200 preparative gel filtration column (2 x 80 ml) was loaded with a protein A eluate pool containing fully assembled heterodimeric antibodies and half antibodies using a mobile phase containing 50 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea at pH 7.2. Three different protein peaks were observed during elution from the preparative gel filtration column Figure 12 ). Various fractions were collected during elution and analyzed by analytical SE-HPLC and SDS-PAGE. SDS-PAGE analysis showed that peak 2 contained mainly fully assembled antibodies, while peak 3 contained mainly half antibodies Figure 13A and 13B . Peak 1 corresponded to higher molecular weight aggregates. Surprisingly, the use of this unique mobile phase in SEC allowed the separation of half antibodies from fully assembled antibodies as well as from higher molecular weight aggregates Figures 14A-14C

[0145] The preparative SEC process using the unique mobile phase described above was repeated using two other bispecific heterodimeric antibodies specific for different target antigens. Preparations of both heterodimeric antibody B and heterodimeric antibody C contained half antibody contaminants prior to SEC purification Figure 15A and 16A . However, after SEC purification using a mobile phase containing 50 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea at pH 7.2, fully assembled heterodimeric antibodies could be separated from half antibodies such that only the desired fully assembled heterodimeric antibodies remained Figure 15B and 16B .

[0146] In another series of experiments, the effect of flow rate on the efficiency of half antibody separation using SEC was evaluated. SEC was performed on a preparation of recombinant bispecific heterodimeric antibodies using a Superdex 200 preparative gel filtration column (2 x 80 ml) and a mobile phase containing 50 mM HEPES, 0.75 M NaCl, 0.5 M arginine, 0.5 M proline, 2.2 M sorbitol, and 4 M urea at pH 7.2 at a flow rate ranging from 0.02 ml / min to 0.2 ml / min Figures 17A-17F . The results showed that half antibodies were more efficiently separated from fully assembled antibodies at slower flow rates, with optimal separation occurring at flow rates of about 0.02 ml / min to about 0.06 ml / min Figures 17A-17F ​).

[0147] In summary, the experimental data described in this example show that the use of a mobile phase comprising a chaotropic agent, a sugar alcohol, and an amino acid in a SEC gel filtration column can effectively separate half antibody contaminants and high molecular weight aggregates from fully assembled antibody. This method provides a robust one-step process for removing these problematic contaminants from recombinant antibody preparations, particularly multispecific heterodimeric antibody preparations.

[0148] All publications, patents, and patent applications discussed and cited herein are incorporated by reference in their entirety. It is to be understood that the disclosed application is not limited to the particular methodology, reagents, and materials described, as these can vary. It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only, and is not intended to limit the scope of the appended claims.

[0149] Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the application described herein. Such equivalents are intended to be encompassed by the following claims.

Claims

1. A method for separating an antibody from its hapten form, the method comprising: A mobile phase is used to contact a solution containing antibodies and their hapten forms with a gel filtration matrix. The mobile phase has a pH of 6.5 to 7.5 and contains urea, sorbitol, proline, arginine, and sodium chloride. and Elution fractions are collected from the gel filtration matrix, wherein the antibody is eluted in one set of elution fractions and its half-antibody form is eluted in another set of elution fractions, thereby separating the antibody from its half-antibody form.

2. The method of claim 1, wherein the urea is present at a concentration of 2M to 4.5M, or wherein the urea is present at a concentration of 4M.

3. The method of claim 1, wherein the sorbitol is present at a concentration of 1M to 4.5M, or wherein the sorbitol is present at a concentration of 2M to 2.5M.

4. The method of claim 1, wherein the arginine and / or the proline are present at a concentration of 0.25M to 1M, or wherein the arginine and / or the proline are present at a concentration of 0.5M.

5. The method of claim 1, wherein the sodium chloride is present at a concentration of 0.25 M to 0.8 M.

6. The method of claim 1, wherein the mobile phase comprises 2M to 4.5M urea, 1M to 4.5M sorbitol, 0.25M to 1M arginine, 0.25M to 1M proline, and 0.25M to 0.8M sodium chloride, or wherein the mobile phase comprises 4M urea, 2.2M sorbitol, 0.5M arginine, 0.5M proline, and 0.75M sodium chloride.

7. The method of claim 1, wherein the mobile phase has a pH of 7.0 to 7.

4.

8. The method of claim 1, wherein the mobile phase is applied to the gel filtration matrix at a flow rate of 0.02 ml / min to 0.06 ml / min.

9. The method of claim 1, wherein the gel filtration matrix has a gradation range of 10 kDa to 600 kDa.

10. The method of claim 1, wherein the gel filtration matrix comprises cross-linked agarose and dextran.

11. The method of claim 1, wherein the antibody is recombined and generated in mammalian cells.

12. The method of claim 11, wherein the mammalian cell is a Chinese hamster ovary cell.

13. The method of claim 1, wherein the antibody is a multispecific heterodimer antibody.

14. The method of claim 1, wherein the solution containing the antibody and its half-antibody form is an eluent from a protein A chromatography eluent pool or effluent, or wherein the solution containing the antibody and its half-antibody form is a cell culture supernatant or cell lysis buffer.

Images