Compositions and methods for removing detergents from aqueous solutions

Abstract

Compositions and methods for removing a first detergent or surfactant from an aqueous solution by adding a second detergent or surfactant at a concentration exceeding the critical micelle concentration (CMC) of the second detergent or surfactant are described. These compositions and methods are particularly suitable for solutions containing proteins. Typical first detergents / surfactants include polysorbate 20, polysorbate 80, and Triton X-100. Suitable second detergents or surfactants can be ionic, nonionic, or zwitterionic. Typical second detergents / surfactants include, but are not limited to, galactoside detergents (e.g., octyl-β-galactoside), glucoside detergents (e.g., MEGA8, MEGA9, MEGA10), cholamide detergents (e.g., CHAPS, CHAPSO, BIGCHAPS), and sulfobetaine detergents (e.g., sulfobetaine 3-10).

Description

[0001] This application claims the benefit of U.S. Provisional Patent Application No. 62 / 818,554, filed on March 14, 2019. These and all other referenced external materials are incorporated herein by reference in their entirety. If the definition or use of a term in a reference incorporated by reference is inconsistent or contrary to the definition of the term set forth herein, the definition of the term set forth herein shall prevail. Technical Field

[0002] The field of the invention is the removal of detergents or surfactants, in particular from pharmaceutical preparations. Background Art

[0003] The following description includes information that may be helpful in understanding the present invention. It is not an admission that any of the information provided herein is prior art or relevant to the presently claimed invention, nor is it an admission that any publication specifically or implicitly referenced is prior art.

[0004] Excipients are frequently added to small molecule (e.g., <500D) and / or protein solutions (e.g., pharmaceutical formulations) to reduce aggregation, prevent nonspecific binding to container surfaces, and otherwise improve stability. Traditional excipients include proteins (e.g., human or bovine serum albumin, ovalbumin, immunoglobulins, etc.), sugars and polysaccharides, and soluble polymers (e.g., polyvinylpyrrolidone). To provide non-immunogenic formulations, nonionic and zwitterionic detergents and surfactants are increasingly being used for this purpose. However, many detergents and surfactants interfere with analytical methods used for characterization and / or quality control of such solutions (e.g., immunoassays, peptide reporter-based assays, cell-based assays, mass spectrometry, etc.).

[0005] Unfortunately, once detergents and surfactants are introduced, they are difficult to remove from solution. Ion exchange can be used to remove charged (such as cationic or anionic) detergents and surfactants, but many commonly used surfactants are not formally charged. Hydrophobic media can bind large amounts of detergents and / or surfactants, however, such hydrophobic media can also bind large amounts of target proteins (which may be present at low concentrations). Dialysis is used to remove detergents or surfactants, but is only partially successful because the self-association of detergents or surfactants with micelles reduces the amount of "solvated" surfactants or detergents that can be removed by dialysis. Therefore, the concentration gradient drives the removal of detergents and surfactants by dialysis at a very slow rate. The formation of micelles also interferes with the removal of surfactants and detergents by more active size-based separation methods (such as gel filtration and ultrafiltration) because their size leads to retention by ultrafiltration membranes and exclusion from the internal volume of the gel filtration medium.

[0006] Commercial products are available for the removal of surfactants and detergents from aqueous solutions. For example, ThermoFisher's HiPPR resin claims to remove 95% of surfactants or detergents from low-concentration protein / peptide solutions while maintaining the protein / peptide content. G Biosciences offers a detergent resin that is described as having a high affinity for most surfactants and a low affinity for most proteins and peptides. Calbiochem offers a hydrophobic CALBIOSORB TM Resins for batch removal of surfactants are recommended and should be used in dialysis buffer to avoid nonspecific binding of proteins of interest. However, it is unclear whether these products are effective in removing polysorbate surfactants, which are increasingly used as excipients, or whether all proteins / peptides remain in solution after treatment.

[0007] In some applications, peptide solutions are extracted using organic solvents such as ethyl acetate. In such extractions, the relatively hydrophobic detergent is transferred to the relatively insoluble organic layer of the extraction mixture. However, due to the potential for irreversible denaturation, such methods are often followed by characterization of the physical properties of the protein (e.g., by mass spectrometry), which may not provide information about activity.

[0008] Therefore, there remains a need for a simple and convenient method to remove detergents and / or surfactants from aqueous solutions while retaining functional proteins and / or peptides in solution. Summary of the invention

[0009] The compositions and methods contemplated by the present invention provide for the addition of a second surfactant or detergent (e.g., a galactoside detergent, such as octyl-β-galactoside; a glucoside detergent, such as N-octanoyl-N-methylglucamine (MEGA 8), N-nonanoyl-N-methylglucamine (MEGA 9) and / or N-decanoyl-N-methylglucosamine (MEGA10); cholamid detergents, such as 3-[3-(cholamidopropyl)dimethylamino]propanesulfonic acid inner salt (CHAPS), 3-[3-(cholamidopropyl)dimethylamino]-2-hydroxy-1-propanesulfonate (CHAPSO) and / or (3a,5b,7a,12a)-N,N-bis[3-(D-glucose amido)propyl]-3,7,12-trihydroxycholestane-24-amine (BIGCHAPS); and / or sulfobetaine detergents, such as sulfobetaine 3-10) to remove surfactants or detergents (e.g., polysorbate 20, polysorbate 80 and / or polyethylene glycol octylphenyl ether (Triton X-100)) method so that the concentration of the addition of the second surfactant or detergent is at least equal to the critical micelle concentration of the second surfactant or detergent, allowing the formation of mixed micelles including two surfactants / detergents, and using a size-based separation method to separate the protein from the resulting mixed micelles (such as ultrafiltration and / or gel filtration). Some embodiments include the step of sealing the ultrafiltration membrane used in the ultrafiltration separation before or when separating the protein from the mixed micelles. Some embodiments include the step of sealing the gel filtration medium used in the gel filtration separation before or when separating the protein from the mixed micelles. Some embodiments of the present inventive concept include the step of collecting an analytical fraction containing the protein from the size-based separation method, and the analytical fraction can be analyzed using a cell-based assay. The protein can be a botulinum neurotoxin (e.g., botulinum neurotoxin type A or BoNT / A). BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 : Figure 1 An exemplary process for removing detergent from a protein-containing solution is schematically depicted.

[0011] Figure 2 : Figure 2 Shown is the CMC 535 TM Typical results for characterizing the detergent content of aqueous solutions treated with different first and second detergents.

[0012] FIG. 3A to FIG. 3C : Figure 3A Typical results are shown for the removal of polysorbate 20 from solution by diafiltration in the presence of different secondary detergents. Figure 3BTypical results are shown for the removal of polysorbate 80 from solution by diafiltration in the presence of different secondary detergents. Figure 3C Shown are typical results for the removal of Triton X-100 from solution by diafiltration in the presence of different secondary detergents.

[0013] FIG. 4A to FIG. 4C : Figure 4A Typical results are shown for botulinum neurotoxin type A (BoNT / A) obtained from solution using the methods of the present inventive concepts. The upper panel shows the BoNT / A recovered from a BoNT / A sample containing polysorbate 20. The lower panel shows the data normalized to a control sample containing BoNT / A and not containing the first detergent (polysorbate 20). Figure 4B Typical results are shown for botulinum neurotoxin type A (BoNT / A) obtained from solution using the methods of the present inventive concepts. The upper panel shows the BoNT / A recovered from a BoNT / A sample containing polysorbate 80. The lower panel shows the data normalized to a control sample containing BoNT / A and not containing the first detergent (polysorbate 80). Figure 4C Typical results are shown for botulinum neurotoxin type A (BoNT / A) obtained from solution using the methods of the present inventive concepts. The upper panel shows the BoNT / A recovered from a BoNT / A sample containing Triton X-100. The lower panel shows the data normalized to a control sample containing BoNT / A without the first detergent (Triton X-100).

[0014] Figure 5A and Figure 5B : Figure 5A Shown with CMC 535 TM Typical results from a study of the effect of additional BigCHAP wash solution on the removal of polysorbate 80. Figure 5B shows the use of CMC535 TM Typical results from a study of the effect of additional buffer washes on the removal of polysorbate 80.

[0015] Fig. 6A and Figure 6B : Fig. 6A Shown are typical results from a study of the recovery of BoNT / A from samples containing polysorbate 80, which was subsequently extracted with mixed micelles containing BigCHAP and removed by repeated washing with buffer containing BigCHAP. The upper panel shows the results relative to BoNT / A spiked into the sample. The lower panel shows the results relative to the recovery of BoNT / A samples without polysorbate 80. Figure 6BShown are typical results from a study of the recovery of BoNT / A from samples containing polysorbate 80 that was subsequently extracted with micelles mixed with BigCHAP and removed by repeated washing with buffer without a secondary detergent. The left panel shows the results relative to BoNT / A spiked into the sample. The right panel shows the results relative to the recovery of BoNT / A samples without polysorbate 80.

[0016] 7A to 7C : Fig. 7A Figure 2 shows typical results of polysorbate 20 extraction from BoNT / A samples using different concentrations of octyl-β-galactoside (OBG), BigCHAP or CHAPS, as shown in Figure 2. TM Represented by law. Figure 7B Figure 2 shows typical results of polysorbate 80 extraction from BoNT / A samples using different concentrations of octyl-β-galactoside (OBG), BigCHAP or CHAPS, as shown in Figure 2. TM Method described. Figure 7C Figure 2 shows typical results of extracting Triton X-100 from BoNT / A samples using different concentrations of octyl-β-galactoside (OBG), BigCHAP or CHAPS, as shown in Figure 2. TM Represented by law.

[0017] FIG. 8A to FIG. 8C : Fig. 8A Shown are typical results from a study of the recovery of BoNT / A from samples containing polysorbate 20 following mixed micellar extraction with varying concentrations of octyl-β-galactoside (OBG), BigCHAP, or CHAPS, as characterized using Biosentel BoTestA / E reagent. The upper panel shows the BoNT / A recovery as a percentage of the amount spiked. The lower panel shows the BoNT / A recovery as a percentage of the control BoNT / A sample without polysorbate 20. Figure 8B Shown are typical results from a study of the recovery of BoNT / A from samples containing polysorbate 80 following mixed micellar extraction with varying concentrations of octyl-β-galactoside (OBG), BigCHAP, or CHAPS, as characterized using the BioSentinel BoTestA / E reagent. The upper panel shows the BoNT / A recovery as a percentage of the amount spiked. The lower panel shows the BoNT / A recovery as a percentage of the control BoNT / A sample without polysorbate 80. Figure 8CFigure 2 shows typical results from a study of the recovery of BoNT / A from samples containing Triton X-100 after mixed micellar extraction with different concentrations of octyl-β-galactoside (OBG), BigCHAP, or CHAPS, as measured with BoTestA / E TM The upper panel shows the BoNT / A recovery as a percentage of the added amount. The lower panel shows the BoNT / A recovery as a percentage of the control BoNT / A sample without Triton X-100.

[0018] 9A to 9D : Fig. 9A Shown are typical results obtained from cell-based BoNT / A assays using BoNT / A prepared in medium without surfactant or in medium containing polysorbate 80 excipient. Fig. 9B Shown with Fig. 9A Typical results from a similar study where BoNT / A samples were dialyzed prior to assay. Fig. 9C Typical results from a study of the removal of polysorbate 80 excipient from BoNT / A containing samples using a secondary surfactant are shown, specifically the effect of polysorbate 80 excipient and residual surfactant in samples extracted with the secondary surfactant on the bioscontinel BoTest A / E TM Effect of cell morphology on the assay. Fig.9D The Biosontinel BoTest A / E was performed with BoNT / A prepared in medium containing surfactant and with BoNT / A prepared in medium containing polysorbate 80 excipient and subjected to mixed micelle extraction. TM Typical results of EC50 determination of BoNT / A in the assay.

[0019] Fig.10 : Fig.10 The typical results of the retention study of polysorbate 80 and BigCHAP mixed micelles by Amicon regenerated cellulose ultrafiltration membranes with different molecular weight cutoffs (MWC) are shown.

[0020] Fig.11 : Fig.11 Shown is a comparison of the rejection of mixed micelles of polysorbate 80 and BigCHAP by PES (Pierce, Sartoris) ultrafiltration membranes of different MWC.

[0021] Fig.12 : Fig.12Shown are typical results of BoNT / A recovery studies after removal of mixed micelles of polysorbate 80 with BigCHAPS using regenerated cellulose (Amicon) and PES (Pierce, Sartoris) ultrafiltration membranes of different molecular weight cutoffs.

[0022] Fig.13 : Fig.13 Shown are typical results investigating the effect of human serum albumin (HSA) pre-blocking and combined pre-blocking and sample spiking on first detergent and BoNT / A recoveries after mixed micelle separation.

[0023] Fig.14 : Fig.14 Biosentel BoCell in vivo TM Typical results from studies investigating the effects of human serum albumin (HSA) pre-blocking and combined pre-blocking and sample spiking on cells in the assay. DETAILED DESCRIPTION

[0024] The present subject matter provides compositions and methods in which a first detergent or surfactant (e.g., a detergent or surfactant added as an excipient to a protein or peptide solution) is at least partially removed by directly adding a second detergent or surfactant at a concentration exceeding the critical micelle concentration (CMC) of the second detergent or surfactant. The second detergent or surfactant can be selected to form micelles having a low molecular weight (e.g., less than about 500 kD, 200 kD, 150 kD, or 100 kD) and / or a small hydrodynamic radius (e.g., less than the hydrodynamic radius of the target protein). In a preferred embodiment, the second detergent of the surfactant has a CMC of about 3.5 mM to about 40 mM and / or forms micelles having an average molecular weight of about 8 kD to about 40 kD. Typical first detergents / surfactants include, but are not limited to, polysorbate 20, polysorbate 80, and Triton X-100. Suitable second detergents or surfactants can be ionic, nonionic, or zwitterionic. Typical second detergents / surfactants include, but are not limited to, galactoside detergents (e.g., octyl-β-galactoside), glucosamine detergents (e.g., MEGA 8, MEGA 9, MEGA 10), choline detergents (e.g., CHAPS, CHAPSO, BIGCHAPS), and sulfobetaine detergents (e.g., sulfobetaine 3-10).

[0025] In this application, the term "about" defines a range of 20% of a nominal value.

[0026] The second detergent or surfactant may be selected so as to not interfere with or provide an acceptable level of interference (e.g., a level of interference that does not interfere with accurate characterization) to a subsequent assay (e.g., an immunoassay and / or a cell-based assay) of the protein or peptide. When the second detergent or surfactant is added directly to a solution containing the first detergent or surfactant, molecules of the first detergent or surfactant are incorporated into the micelles of the second detergent or surfactant to form mixed micelles containing two surfactant / detergent species. Surprisingly, the inventors have discovered that the second detergent / surfactant species and their concentrations can be selected to form relatively small mixed micelles. Specifically, the second detergent or surfactant may be selected so as to be added directly to a solution containing the first detergent or surfactant, resulting in the formation of mixed micelles having an effective hydrodynamic radius smaller than that of the protein or peptide to be analyzed (e.g., botulinum neurotoxin) and present in the initial solution. It should be understood that the surfactant and / or detergent used may be present as an excipient for reducing aggregation of protein therapeutics in many pharmaceutical formulations.

[0027] The resulting small mixed micelles can be effectively separated and / or separated from the target protein or analyte by a size-based separation method (e.g., ultrafiltration, gel filtration, etc.), thereby removing at least a portion of the first and second detergents from the protein-containing solution. It should be understood that in some embodiments, such as those in conjunction with ultrafiltration, during the detergent / surfactant removal process, the target protein or peptide can be concentrated with reference to the concentration in the original sample volume. In a preferred embodiment of the present invention, sufficient first detergent / surfactant and second detergent / surfactant are removed to reduce or eliminate interference with downstream analytical methods that characterize protein quality and / or activity. The resulting high molecular weight portion (e.g., retentate or flow-through portion) comprising the protein or peptide can then be characterized by methods that may be interfered with by the first and / or second surfactant or detergent.

[0028] The grouping of the alternative elements or embodiments of the present invention disclosed herein should not be construed as limiting. The member of each grouping may be mentioned and advocated individually or together with other members of the grouping or other elements found herein. For reasons of convenience and / or patentability, one or more members of a group may be included in a group or deleted therefrom. When any such inclusion or deletion occurs, this specification is deemed to include the modified group, thereby satisfying the written description of all Markush groups used in the appended claims.

[0029] In some embodiments, the mixed micelles of the present invention can pass through a membrane that retains botulinum neurotoxin (e.g., an ultrafiltration or dialysis membrane) to remove the first surfactant from the botulinum neurotoxin solution while retaining the active protein. Similarly, the mixed micelles of the present invention can enter the internal volume of an appropriately selected size exclusion chromatography medium, where the target protein or peptide in the solution is excluded, and emerge as an excluded (i.e., flow-through) volume and / or early fraction during size exclusion chromatography column chromatography.

[0030] In some embodiments, for example, when the concentration of the protein or peptide to be characterized is low, the surface of the membrane or gel filtration medium can be blocked before and / or during the separation step. For example, such surfaces can be blocked by adding one or more exogenous proteins that do not interfere with downstream characterization methods (e.g., human serum albumin, bovine serum albumin, avian serum albumin, acetylated albumin, ovalbumin, mammalian immunoglobulin, avian immunoglobulin, mammalian or teleost gelatin, casein, etc.). In some embodiments, such surfaces can be blocked by adding long-chain polymers (e.g., polyvinyl pyrrolidone or polyethylene glycol). Such blocking can reduce or eliminate losses caused by denaturation of analytes on the separation surface and / or losses caused by nonspecific binding to the separation surface. Blocking proteins and / or polymers can be applied in a concentration range of 0.1% to 10% (w / v) and can be applied before, during, or both before and during the separation step. In some embodiments, blocking proteins and / or polymers can be applied to the surface of the size separation membrane or medium before use, blocking the surface for a given time, and removing excess material by rinsing or washing before applying a sample containing the analyte.

[0031] Figure 1 An example of a method of the present invention is provided in the schematic diagram shown, which describes the separation of mixed micelles from a solution containing botulinum neurotoxin (BoNT) using an ultrafiltration membrane spin column. As shown, a second detergent is added directly to a solution containing BoNT (also containing a first detergent) to form mixed micelles, which have a reduced hydrodynamic radius relative to micelles formed by the first detergent alone. Centrifugation in a spin column equipped with an ultrafiltration membrane of an appropriate molecular weight cutoff causes the mixed micelles to pass through the membrane and retain the BoNT. The trapped BoNT-containing solution can be easily recovered by reversing the direction of the sample-containing portion of the spin column and briefly centrifuging. It should be appreciated that recovery of BoNT in a smaller volume than the applied sample can provide a BoNT concentration that can facilitate analysis (e.g., using a cell-based assay or immunoassay).

[0032] The inventors have found that the methods conceived by the present invention can effectively remove the first surfactant that interferes with biochemical characterization methods and cell-based assay methods. The inventors believe that the methods conceived by the present invention can be similarly applied to physical methods such as mass spectrometry and capillary electrophoresis.

[0033] In a typical surfactant / detergent removal protocol, an Amicon UFC spin adsorption column is pre-blocked with a blocking buffer containing 0.5% w / v human serum albumin (HSA), and residual blocking buffer is removed by centrifugation. Then 100 μL of sample or drug (e.g., 50 mM HEPES, 140 mM NaCl, 0.5% HSA (human serum albumin), and 0.1% 100 μM botulinum neurotoxin type A (BoNT / A) in a first detergent) is mixed with 400 μL of a buffer containing a second detergent, the concentration of which is selected to match or exceed the critical micelle concentration (CMC) of the second detergent (e.g., about 1% w / v or higher). HSA may be omitted in some embodiments, as HSA has been found to reduce the recovery of some BoNTs (e.g., BoNT / A).

[0034] After a short incubation period (e.g., about 1 minute, 2 minutes, 3 minutes, 5 minutes, 10 minutes, 15 minutes, 30 minutes, or 1 hour), the mixture is applied to a centrifugal adsorption column and centrifuged. In some embodiments, the centrifugal adsorption column can be washed by adding a 500 μL volume of a buffer containing a second detergent and applying centrifugal force. After washing the centrifugal adsorption column with 500 μL of wash buffer, the centrifugal adsorption column is inverted and briefly centrifuged to recover a volume of about 15 μL representing the detergent extracted sample. This volume can be diluted for analysis and / or split between different analytical methods.

[0035] like Figure 2 As shown, the inventors found that a commercial product (CMC 535 TM ) is unreactive with many detergents that have been found to be useful as secondary detergents. As shown, CMC 535 TM The assay generated dose / response curves for polysorbate 20 (P20), polysorbate 80 (P80), and Triton X-100 (X100), but had no significant responses to many potential secondary detergents within the concentration range tested.

[0036] Typical results for first detergent removal using the above protocol and using CMC 535 TM Characterize as FIG. 3A to FIG. 3CAs shown, octyl-β-glucoside, BigCHAP, and CHAPS were able to remove polysorbate detergents well. Removal of Triton X-100 seemed to be independent of the choice of the second detergent.

[0037] Samples containing BoNT / A in a buffer containing a first surfactant were similarly treated and assayed using the BioSentinel BoTest TM Activity assay to characterize BoNT / A content. BoTest TM The activity assay utilizes a reporter peptide that contains an analog of the BoNT / A substrate and carries a pair of FRET fluorophores separated by the substrate peptide. Proteolysis results in separation of the FRET pair and a distinct change in fluorescence. 50 μL of the extracted sample is mixed with 100 μL of 0.25 μM BoTest TM A / E reporter peptide substrate mix and quantification of in vitro BoTest using dose / response curves TM The results are as follows FIG. 4A to FIG. 4C As shown. It is understood that in some cases, the first detergent can activate BoNT / A to some extent. Therefore, a control sample that retains the first detergent may show a false increase in BoNT / A activity, giving the impression that the recovery of the mixed micelles in the test sample after the first detergent is removed is low. Similarly, if a control is provided without the first detergent, the recovery may be overestimated. However, the use of certain second detergents can significantly increase the recovery of BoNT / A. As shown in the figure, the choice of second detergent can have an impact on the recovery of BoNT / A. For example, detergent SB-10 may be different from BoTest TM Assay incompatible.

[0038] In some embodiments, additional volumes of buffer containing a second detergent (BigCHAP) can be added to the sample containing BoNT / a and a first detergent (polysorbate 80), followed by additional rounds of centrifugation. Figure 5A As shown in CMC35 TM This was shown to be effective in further removing the first detergent. Figure 5B The effect of a similar study with a buffer containing no second surfactant is shown. As shown, additional washes with a buffer containing a second detergent can remove additional first detergent that was not removed in the first mixed micelle extraction. Additional washes with a buffer do not produce this effect.

[0039] It should be understood that improved initial detergent extraction does not necessarily indicate improved functional analyte recovery performance. Figure 5A and 5B The samples evaluated in were also characterized using BoTestTM A / E reagents as described above. Fig. 6A and Figure 6B . As shown, the recovery of functional BoNT / A decreases with additional washes. This may be due to non-specific binding leading to denaturation and / or loss of BoNT / A. Therefore, the mixed micelle extraction protocol may be suitable for optimal recovery of functional analytes, which may or may not be related to the most efficient protocol for removing the primary detergent. In some embodiments, additional washes after the initial mixed micelle extraction can be eliminated.

[0040] 7A to 7C The results of the optimized mixed micellar extraction protocol developed for BoNT / A samples containing polysorbate 20, polysorbate 80, and Triton X-100 as the first detergent, and octyl-β-galactoside, BigCHAP, and CHAPS as the second detergent, as shown in Figure 5. TM The spin column was pre-blocked with human serum albumin before use; human serum albumin was also added to the BoNT / A preparation. Surprisingly, the inventors found that some first detergents (e.g., Triton X-100) appear to be able to be removed from the sample by diafiltration without forming mixed micelles. As shown in the figure, the CMC 535 TM It was determined that there was no significant difference in the removal of the first detergent using 1% (w / v) and 5% (w / v) of the second detergent.

[0041] FIG. 8A to FIG. 8C Results of an optimized mixed micelle extraction protocol developed for BoNT / A samples containing polysorbate 20, polysorbate 80, and Triton X-100 as primary detergents, and using octyl-β-galactoside, BigCHAP, and CHAPS as secondary detergents, as characterized using the BoTestA / E assay, are shown. The spin adsorption columns were pre-blocked with human serum albumin prior to use; human serum albumin was also added to the BoNT / A preparations. Surprisingly, the inventors found that some primary detergents (e.g., Triton X-100) appear to be able to be removed from the sample by diafiltration without the formation of mixed micelles. As shown, for most first / second detergent combinations, the recovery of functional BoNT / A increased with increasing concentration of the second detergent.

[0042] Although BoTest TMThe A / E reagent is a solvated synthetic peptide used as a biochemical assay to characterize the recovery of BoNT / A in the above studies, but in certain applications, a cell-based assay can be used to characterize samples containing the analyte where the first detergent has been removed by mixed micellar extraction. Such a cell-based assay has the advantage of providing an additional selectivity step in the form of selective uptake of the analyte. For example, botulinum neurotoxins such as BoNT / A are selectively taken up by cultured motor neurons and some neuron-derived cells. The first detergent used as an excipient can adversely affect such cells.

[0043] Fig. 9A An example of this interference is shown in Figure 2. A sample of purified BoNT / A holotoxin in surfactant-free medium and a commercial product of similar concentration of BoNT / A in medium containing polysorbate 80 (P80) excipient were applied to the BioSentinel BoCell TM A / E cell-based assays utilize genetically modified cells expressing BoNT / A sensitive reporter peptides containing a FRET fluorophore pair. The emission ratio of these fluorophores relative to the logarithm of the BoNT / A concentration is shown in Figure 2. Fig. 9A As shown, the cell-based assay had little response to the BoNT / A samples containing surfactant. Fig. 9B shows the results of a similar study where the samples were dialyzed before application to cells. As shown, dialysis produced little improvement. Although EC 50 , but when applied to the same assay, the estimated values ​​were approximately 10-fold higher than BoNT / A presented in various surfactant-free media.

[0044] Therefore, using BoCell TM The utility of mixed micellar extraction in reducing the first detergent effect on cells in culture was characterized using cells and media used in the A / E assay, as well as BoNT / A samples in iBAM2 cell culture media containing 0.1% polysorbate 80 surfactant excipient. Aliquots (100 μL) of this BoNT / A formulation were extracted with 400 μL of 2% MEGA8 or 1% of other secondary surfactants and analyzed using an Amicon Ultra TM The resulting mixture was concentrated by centrifugal filter. The resulting concentrate was washed with an additional 450 μL of the second surfactant solution by an additional round of centrifugation, followed by an additional wash with 450 μL of iBAM2 medium. The resulting sample (approximately 15 μL) was then adjusted to a volume of 100 μL with iBAM2 medium and added to the BoCell TM A / E cells. The cells were maintained at 37°C and 5% CO 2Characterization and imaging were performed after 24 and 48 hours. Fig. 9C As shown. Fig. 9C In the figure, cells that are visually similar to control cells are represented by stars. As shown in the figure, the results of MEGA 10, BIGCHAPS and CHAPS are similar to the untreated control cells and cells treated with BAM2 medium alone. The inventors believe that the maintenance of normal cell distribution and morphology indicates the successful removal of the polysorbate 80 excipient and maintenance of cell health.

[0045] The samples were studied with different concentrations of BoNT / A in medium containing polysorbate 80, treated in the same manner with BIGCHAPS as a second surfactant and applied to the BioSentinel BoCell TM BoNTA / E cell-based assay. The results were compared with those obtained with BoNT / A in medium without surfactant, as Fig.9D As shown, samples containing surfactant provided a dose / response curve in the cell-based BoNT / A assay after mixed micelle extraction that was very similar to the curve obtained from samples without surfactant. Table 1 provides a comparison of the EC50 values ​​obtained.

[0046]

[0047] Table 1

[0048] In general, the average EC obtained from the samples containing surfactant was 50 is the average EC obtained from the control samples 50 The value of 50.5% (CV 4.8%) can be corrected by an adjustment factor (eg, 0.5).

[0049] In the above studies, the ultrafiltration membrane used had a molecular weight cutoff (MWC) of 100 kDa. However, it is understood that proteins with slightly higher molecular weights (e.g., 150 kDa BoNT / A) may pass through such membranes due to the shape of the protein molecules and the distribution of membrane pore sizes. Therefore, further studies were performed to characterize the effective range of MWC for surfactants in mixed micelle removal solutions. The passage of polysorbate 80 (as the first detergent) in the presence or absence of BIGCHAPS (as the second detergent) was studied. CMC 535 was used to remove the surfactants in the presence of BIGCHAPS. TM The assay characterized the material that passed through the regenerated cellulose ultrafiltration membrane and was responsive to polysorbate 80 but not to BigCHAP. Fig.10As shown in the figure, both 100 kDa and 50 kDa molecular weight cutoff (MWC) regenerated cellulose ultrafilters allowed mixed micelles to pass through, while the retentate showed an increased amount of polysorbate 80 when 30 kDa and 10 kDa MWC ultrafilters were used.

[0050] It should be understood that while regenerated cellulose filters are commonly used due to their high porosity and relatively low nonspecific binding, other materials used in ultrafiltration membranes, such as polyethersulfone (PES), are manufactured by different methods and may exhibit different properties with respect to mixed micelle exclusion. Examples of results from studies using PES ultrafilters of different MWC are shown in Fig.11 As shown, the 100 kDa and 50 kDa MWC PES membranes showed significantly greater mixed micelle retention than regenerated cellulose membranes of similar nominal MWC.

[0051] Fig.12 Shown is the effect of different ultrafiltration membrane compositions on the recovery of BoNT / A from media containing polysorbate 80, using BigCHAP as a secondary detergent. BoLISA TM The assay is an immunological assay that is not significantly affected by residual polysorbate 80 and was used to quantify BoNT / A in these studies. As shown, the recovery of BoNT / A was similar for both the regenerated cellulose and PES ultrafiltration membranes, with slightly lower recovery for PES.

[0052] In order to reduce nonspecific binding to ultrafiltration membranes and other surfaces, it is common to block available nonspecific binding sites on these surfaces using proteins such as casein, gelatin, ovalbumin, bovine serum albumin, human serum albumin, nonspecific immunoglobulins, etc. Such surfaces can be pre-blocked by protein treatment before exposure to the sample to be processed, or blocked during processing by adding proteins (usually small volumes of concentrated stock solutions) to the sample to be processed, or both. However, such blocking proteins may have adverse effects on downstream processes (e.g., using in vitro methods (e.g., BoLISA)). TM and BoTest TM ) or cell-based assays (such as BoCell TM )'s characterization) has a negative impact.

[0053] Fig.13 The effect of pre-blocking of the ultrafiltration membrane used in the treatment of mixed micelles with human serum albumin (2% w / v, overnight) and the addition of 0.1% w / v human serum albumin to the treated samples is shown. TM The first detergent content of the samples was characterized and the BoLISA TM and BoTest TMAs shown in the figure, the addition of human serum albumin to the sample followed by pre-blocking with ultrafiltration had little effect on detergent removal and when using BoLISA TM and BoTest TM The recovery of BoNT / A was slightly improved when the in vitro assay was characterized.

[0054] Fig.14 Shown using BioSentinel BoCell TM The effect of adding human serum albumin to samples characterized for BoNT / A content was determined. As shown in the figure, the addition of human serum albumin to samples processed using the mixed micelle method and pre-blocked ultrafiltration membrane had a significant effect on the BoCell TM The health of the cells used in the assay is negatively affected. Therefore, the impact of blocking proteins added to samples processed by mixed micelle separation methods on downstream processes is process dependent.

[0055] It should be clear to those skilled in the art that, without departing from the inventive concept of this article, in addition to the changes already described, more changes can be made. Therefore, the subject matter of the present invention is not limited in the spirit of the appended claims. In addition, when interpreting the specification and claims, all terms should be interpreted in the broadest possible manner consistent with the context. In particular, the terms "comprises" and "comprising" should be interpreted as referring to elements, components or steps in a non-exclusive manner, indicating that the referenced elements, components or steps can exist, be used or combined with other elements, components or steps that are not explicitly referenced. If the specification claims involve at least one selected from the group consisting of A, B, C... and N, it should be interpreted as requiring only one element in the group, rather than A and N or B and N, etc.

Claims

1. A method for removing a first surfactant or detergent from a solution containing a protein, the method comprising: adding a second surfactant or detergent to the solution to a concentration at least equal to the critical micelle concentration of the second surfactant or detergent to form a suspension comprising mixed micelles of the first surfactant or detergent and the second surfactant or detergent; as well as separating the protein from the first surfactant or detergent by removing the mixed micellar suspension from the solution using a size-based separation method; wherein the first surfactant or detergent is selected from polysorbate 20, polysorbate 80 and Triton X-100; The second surfactant or detergent is selected from octyl-β-galactoside, MEGA 8, MEGA 9 and MEGA 10, CHAPS, CHAPSO, BIGCHAPS and sulfobetaine 3-10; The size-based separation method is ultrafiltration, which is performed through an ultrafiltration membrane with a molecular weight cut-off of 10, 30, 50 or 100 kDa.

2. The method according to claim 1, comprising the step of sealing the ultrafiltration membrane used in the ultrafiltration process.

3. The method of claim 1, comprising collecting analytical fractions comprising protein from the size-based separation method.

4. The method of claim 3, comprising characterizing the analytical fractions using a cell-based assay.

5. The method of claim 1, wherein the protein is botulinum neurotoxin.

6. The method of claim 5, wherein the botulinum neurotoxin is botulinum neurotoxin type A.

Images