Method of isolating and analyzing nanoparticles and proteins

Abstract

To provide a simple, high-resolution method of isolating and quantifying proteins, in particular antibody drugs and aggregates thereof, and suspended nanoparticles using the SEC (Size Exclusion Chromatography) method.SOLUTION: A size exclusion chromatography method is provided, involving analyzing nanoparticles with diameters in a range of 10 nm to 150 nm and proteins in an aqueous eluent containing one or more non-ionic surfactants.SELECTED DRAWING: Figure 11

Description

[Technical field]

[0001] The present invention relates to a method for separating and analyzing suspended nanoparticles and dissolved proteins under identical conditions in an aqueous eluent. [Background technology]

[0002] GPC (Gel Permeation Chromatography), also known as SEC (Size Exclusion Chromatography) or GFC (Gel Filtration Chromatography), is a mode of liquid chromatography and is the most widely used method for measuring the molecular weight of polymers. The SEC method is based on the mechanism of "size exclusion" between the pores on the surface of the porous packing material and the polymer in a column packed with the porous packing material. The principle of GPC (SEC) (size exclusion mechanism) is generally explained by the following model.

[0003] As shown in Figure 3, large polymers cannot reach the depths of the porous packing material, so they end up passing through a short flow path and reaching the outlet the fastest, while smaller polymers can reach deeper, so the flow path becomes longer and they are slower to reach the column outlet. Due to this principle, components with larger molecular size are eluted in order (see Figure 7). Within a certain molecular weight range, there is a linear relationship between the elution time and the logarithm of the molecular weight, and molecules larger than the pores of the packing material are uniformly passed through (exclusion limit), while molecules smaller than a certain size completely permeate the pores and are eluted at approximately the same position (permeation limit) (see Figure 7a).

[0004] Therefore, samples with molecular weights within the measurable range often have a peak shape with a gentle distribution. However, samples containing components with molecular weights above the exclusion limit often have a peak shape with a steep rise on the high molecular weight side, because the components in this range elute at roughly the same elution time (see Figure 7b).

[0005] In liquid chromatography, the sample must be dissolved in the eluent. The same is true for the SEC method, where the polymer is dissolved in a solvent such as tetrahydrofuran, and if there is any insoluble matter, it is filtered through a 0.1-0.2 μm membrane filter before analysis. The detector used in the SEC method is selected according to the application. The most commonly used detector is the refractometer. This is because many of the polymers being measured do not absorb in the ultraviolet / visible light region, and this detector is a concentration meter. In addition, if you want to calculate the absolute molecular weight, you may use a refractometer in combination with a light scattering detector or a viscometer. If the polymer being measured has an ultraviolet / visible light region, you may use a refractometer in combination with an ultraviolet / visible detector.

[0006] In recent years, the efficacy and usefulness of biopharmaceuticals (antibody drugs) has been recognized, but on the other hand, there has been insufficient research into their side effects, and aggregates in particular have attracted attention. These aggregates are often classified according to their size. They are classified into nanometer particles of 100 nm (0.1 μm) or less, submicron particles of 0.1 to 1 μm, and micron particles of 1 to 10 μm. Examples of analytical methods (particle size, concentration, number, etc.) recommended in each field are given. The SEC method is mainly used for the smallest nanometer particles of 100 nm (0.1 μm) or less. However, there are issues with the SEC method, such as insufficient separation, and the difficulty of passing large aggregates of several hundred nm or more through the column.

[0007] Figure 8 is a schematic diagram of a calibration curve when separating nanoparticles by the SEC method. As with general SEC methods, a linear relationship is obtained between the elution time and the logarithm of the particle size within a certain particle size range. However, nanometer particles larger than a certain size are not eluted because they cannot pass through the packing material, column end fittings, line filters, etc. Therefore, unlike general SEC methods, no peaks are observed in this region.

[0008] In the SEC method, a calibration curve of elution time vs. logarithm of molecular weight is prepared using multiple standard samples with known molecular weights and narrow distributions, and the molecular weight of an unknown sample is calculated.

[0009] Generally, when measuring the molecular weight of proteins, including antibody drugs, a calibration curve is created using protein molecular weight markers with known molecular weights.

[0010] In the case of preparing a calibration curve for nanoparticles, generally, multiple "polystyrene nanoparticles (hereinafter referred to as PS)" with a small distribution of particle shapes are often used. However, it was difficult to measure the dissolved protein and the suspended "polystyrene nanoparticles" simultaneously under the same conditions by molecular exclusion chromatography. Summary of the Invention [Problem to be solved by the invention]

[0011] The present invention provides a method for easily separating and quantifying proteins, particularly antibody drugs, their aggregates, and suspended nanoparticles with high resolution by SEC (size exclusion chromatography).

[0012] As a result of intensive research and development conducted by the present inventors in order to solve the above problems, the present invention has been developed. That is, the present invention includes the following aspects.

[0013] [1] Size exclusion chromatography, characterized by analyzing nanoparticles having a particle size of 10 to 150 nm and proteins using an aqueous eluent containing one or more nonionic surfactants.

[0014] [2] The size exclusion chromatography according to [1], wherein the concentration of the nonionic surfactant in the eluent is 0.02 to 2%.

[0015] [3] The size exclusion chromatography according to [1] or [2], wherein the nonionic surfactant has no absorption in the ultraviolet absorption region from 220 nm to 380 nm.

[0016] [4] The size exclusion chromatography according to [1] or [2], characterized in that the nonionic surfactant is one or more selected from the following types of nonionic surfactants:

[0017] 1) Higher alcohol ethylene oxide adducts 2) Polyoxyethylene fatty acid esters (fatty acid ethylene oxide adducts, polyethylene glycol fatty acid esters) 3) Higher alkylamine ethylene oxide adducts 4) Fatty acid amide ethylene oxide adducts (polyoxyethylene alkylamines, polyoxyethylene fatty acid amides) 5) Polypropylene glycol ethylene oxide adduct (Pluronic (registered trademark) type nonionic surfactant, polyoxyethylene polyoxypropylene glycol) 6) Glycerol fatty acid ester 7) Pentaerythritol fatty acid esters 8) Sorbitol fatty acid esters 9) Sorbitan fatty acid ester 10) Sucrose fatty acid esters 11) Alkyl polyglycosides 12) Fatty acid alkanolamides [5] The size exclusion chromatography according to [1] or [2], characterized in that the nonionic surfactant used is at least polyoxyethylene (23) lauryl ether (hereinafter also referred to as Brij 35).

[0018] [6] The size exclusion chromatography according to [1] or [2], characterized in that the gel base material for the size exclusion chromatography is silica-based or polyacrylate resin-based.

[0019] In the present invention, examples of nanoparticles include polystyrene nanoparticles, polystyrene-divinylbenzene copolymer nanoparticles, and gold nanoparticles.

[0020] In the present invention, examples of proteins include antibodies, enzymes, albumin, collagen, and the like.

[0021] In the present invention, examples of the eluent include pure water, phosphate buffer, citrate buffer, MES buffer, Tris buffer, PIPES buffer, ACES buffer, MOPSO buffer, BES buffer, MOPS buffer, TES buffer, HEPES buffer, TAPSO buffer, HEPSO buffer, and EPPS buffer.

[0022] In the present invention, the pH of the eluent is preferably in the range of 6 to 8 since the measurement targets include proteins, but can be appropriately set taking into consideration the acid denaturation or alkali denaturation of the measurement target proteins.

[0023] In the present invention, an inorganic salt may be appropriately added to the eluent, such as sodium chloride, potassium chloride, lithium chloride, sodium bromide, potassium bromide, lithium bromide, magnesium chloride, calcium chloride, magnesium bromide, calcium bromide, etc. Effect of the Invention

[0024] The present invention makes it possible to provide a method for easily separating and quantifying antibody pharmaceuticals and their aggregates with high resolution by SEC (size exclusion chromatography). [Brief description of the drawings]

[0025] [Figure 1] This is an example of a synthetic polymer. [Diagram 2] FIG. 1 is a diagram illustrating an example of a system configuration according to the present invention. [Diagram 3] FIG. 1 is a schematic diagram illustrating the principle of separation of a dissolved polymer sample by the SEC method. [Figure 4] FIG. 1 is a schematic diagram illustrating the principle of separation of undissolved nanoparticles by the SEC method. [Diagram 5] FIG. 1 shows the UV absorption spectra of dissolved and undissolved polystyrene nanoparticles. [Figure 6] FIG. 2 is a schematic diagram showing the behavior of undissolved polystyrene nanoparticles when irradiated with ultraviolet light. [Figure 7] FIG. 1 is a schematic diagram showing a calibration curve and a separation pattern when a dissolved polymer sample is separated by the SEC method. [Figure 8] FIG. 1 is a schematic diagram showing a calibration curve and a separation pattern when dissolved nanoparticles are separated by the SEC method. [Figure 9] 1 shows the effect of surfactant in Example 1. FIG. 1 a shows the dissolution behavior of an antibody drug, and FIG. 1 b shows the dissolution behavior of undissolved polystyrene nanoparticles. [Figure 10] FIG. 1 shows the effect of surfactant concentration on the separation of an antibody drug and polystyrene nanoparticles in Example 1. FIG. a shows the elution time, FIG. b shows the area, and FIG. c shows the pressure at that time. [Figure 11] FIG. 1 shows the reproducibility of separation of a mixture of an antibody drug and polystyrene nanoparticles in Example 1, and is a diagram showing chromatograms obtained by eight measurements overlaid on each other. [Figure 12]FIG. 1 shows the reproducibility in the separation of a mixture of an antibody drug and polystyrene nanoparticles in Example 1. [Figure 13] FIG. 2 is a diagram showing the elution behavior (chromatograms) of polystyrene nanoparticles (20 to 200 nm) and standard proteins (a mixture of five types) in Example 1. [Figure 14] FIG. 2 is a diagram showing overlaid chromatograms of polystyrene nanoparticles (20 to 200 nm) and standard proteins (a mixture of five types) in Example 1. [Figure 15] 1 shows calibration curves for standard proteins (a mixture of five types) in Example 1. In FIG. 1A, the vertical axis represents the logarithm of molecular weight, and in FIG. 1B, the vertical axis represents the logarithm of particle size converted thereto. [Figure 16] This is a calibration curve in which standard proteins (a mixture of five types) and polystyrene nanoparticles are plotted with the horizontal axis representing elution time and the vertical axis representing logarithm of particle size in Example 1. Figure a shows the case where the proteins are approximated individually, and Figure b shows the case where the proteins are approximated by one function. [Figure 17] FIG. 1 shows a chromatogram and calibration curve of protein molecular weight markers (mixture of five types). [Figure 18] FIG. 1 is a schematic diagram showing the process of forming aggregates of an antibody drug in Example 1. [Figure 19] These are chromatograms overlaid with the results of separating the antibody drug that generated aggregates in Example 1. Figure a shows the whole image, and Figure b shows an enlarged view of the elution region of the aggregate peak. The right axis of Figure a shows the logarithm of the particle size (broken line in the figure). [Figure 20] FIG. 2 is a diagram showing the change in the area of ​​each peak depending on the degree of stress treatment in Example 1. [Figure 21] FIG. 2 is a diagram showing the change in the ratio of the area of ​​each peak depending on the degree of stress treatment in Example 1. [Figure 22] FIG. 1 shows the effect of surfactant on the separation of an antibody drug and polystyrene nanoparticles in Example 2. FIG. a shows the elution behavior (chromatogram) in the case where surfactant is included, and FIG. b shows the elution behavior (chromatogram) in the case where surfactant is not included. [Diagram 23]FIG. 1 shows the results of separating polystyrene nanoparticles (20, 100 nm) and protein molecular weight markers (mixture of five types) in Example 2. [Figure 24] 1 shows a calibration curve in Example 2, in which the horizontal axis represents elution time and the vertical axis represents logarithm of particle size, and protein molecular weight markers (mixture of five types) and polystyrene nanoparticles (20, 100 nm) are plotted. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0026] In order to verify the effects of the present invention, examples carried out using actual liquid chromatography will be described below, but the present invention is not limited to these examples in any way.

[0027] The liquid chromatography system was constructed using a Shimadzu Corporation liquid delivery pump LC-20AD, a sample injection device SIL-20ACXR, a column oven CTO-20AC, and an ultraviolet-visible detector SPD-20AV. The system configuration is shown in Figure 2. The common analytical conditions are as follows: Flow rate: 0.5mL / min Column temperature: 30℃ Sample injection volume: 5 or 10 μL Detection wavelength: 220nm Example 1 In this example, molecular exclusion chromatography was performed using a silica-based column. The analytical columns were two TSKgel G3000SWXL (inner diameter 7.8 mm × length 30 cm) manufactured by Tosoh Corporation, packed with a highly porous silica gel base material, and arranged in series.

[0028] The eluent used was a phosphate buffer solution with a small amount of Brij35 (polyoxyethylene (23) lauryl ether) made by Kishida Chemical, a non-ionic surfactant, added. Details of the phosphate buffer solution are as follows: Sodium dihydrogen phosphate dihydrate: 61.314 mmol / L Disodium hydrogen phosphate 12-water: 38.686 mmol / L Sodium chloride: 300 mmol / L pH:7.0 The samples used were polystyrene nanoparticles, protein molecular weight markers, and the antibody drug "Rituxan." The polystyrene nanoparticles used were Thermo Scientific's Nanosphere size standard 3000 series (20-200 nm). The detailed specifications are shown in Table 1.

[0029] [Table 1]

[0030] In addition, a mixed lyophilized product manufactured by Oriental Yeast Co., Ltd. was used as a molecular weight marker for proteins. This marker is a mixed lyophilized product of the following five types of marker proteins with molecular weights ranging from 12,400 to 290,000. Horse heart cytochrome c (molecular weight 12,400) Yeast myokinase (molecular weight 32,000) Yeast enolase (molecular weight 67,000) Porcine heart muscle lactate dehydrogenase (molecular weight 142,000) Yeast glutamate dehydrogenase (molecular weight 290,000) The antibody drug used was "Rituxan." The drug's ingredients and physicochemical findings are shown in Table 2.

[0031] [Table 2]

[0032] First, the effect of adding a non-ionic surfactant to the eluent was examined. Brij35 was added to the phosphate buffer eluent at 0.05%, 0.1%, 0.2%, 0.4%, 0.625%, 1.25%, and 2.5%, and the elution behavior of polystyrene nanoparticles and Rituxan (an antibody drug) was confirmed. To compare the effect of surfactants, measurements were also performed using phosphate buffer containing no surfactant. The samples used were polystyrene nanoparticles (a mixture of 20 and 100 nm) and the antibody drug Rituxan as the protein.

[0033] Figure 9 shows a part of the obtained chromatogram. Figure 9a shows the chromatogram of Rituxan, and Figure 9b shows the chromatogram of the polystyrene nanoparticle mixture (mixture of 20 and 100 nm). Table 3 shows the elution time, Table 4 shows the peak area, and Table 5 shows the peak height. The short dashed line in the figure shows the case where Brij35 is not contained, the solid line shows the case where Brij35 is contained at 0.1%, and the long dashed line shows the case where Brij35 is contained at 1.25%. As can be seen from this, in the eluent that does not contain Brij35, Rituxan is eluted, but polystyrene nanoparticles are not eluted. It can also be seen that polystyrene nanoparticles are eluted if Brij35 is contained at 0.05% or more. It can also be seen that there is no significant change in the elution time, area, and height of polystyrene nanoparticles (20 and 100 nm) even when the concentration of Brij35 is increased (Figure 10b). On the other hand, with increasing Brij35 concentration, there is a slight tendency for the elution time of Rituxan to become slower (FIG. 10a).

[0034] Brij35 is a surfactant and therefore tends to foam easily. If the eluent foams, this can cause problems such as poor elution, so it is preferable to minimize the Brij35 concentration. FIG. 10c is a plot of Brij35 concentration and column pressure at that time. As the Brij35 concentration increases, the viscosity of the eluent increases, and the column pressure also increases. For this reason, it is also preferable to minimize the Brij35 concentration.

[0035] [Table 3]

[0036] [Table 4]

[0037] [Table 5]

[0038] Next, the reproducibility of this separation system was verified. Based on the above results, the concentration of Brij35 was set to 0.10%. As a sample, 100 nm and 20 nm polystyrene nanoparticles and the antibody drug "Rituxan" were mixed and used as one specimen. Figure 11 shows the chromatograms obtained from eight measurements overlaid on each other. This also shows that by adding a small amount of the non-surfactant Brij35, suspended nanoparticles and proteins can be analyzed simultaneously and accurately. The peak at 20.8 minutes is 100 nm polystyrene nanoparticles, the peak at 22.1 minutes is 20 nm polystyrene nanoparticles, the peak at 32.6 minutes is the main component of the antibody drug, and the peak at 45.6 minutes is the additive contained in the antibody drug.

[0039] Figure 12a and Table 6 show the measurement results and reproducibility of the elution time of each peak. The Cv% of the elution time was less than 0.1%, which was an extremely good value for all peaks. Figure 12b and Tables 7 and 8 show the measurement results and reproducibility of the area and height of each peak. For area and height, the CV% was approximately 2%, which was an excellent value for all peaks. This suggests that qualitative and quantitative analysis is fully possible using this method.

[0040] [Table 6]

[0041] [Table 7]

[0042] [Table 8]

[0043] Next, polystyrene nanoparticles and protein molecular weight markers were used to confirm the elution behavior according to molecular size. Based on the above results, the concentration of Brij35 was set to 0.10%. Protein molecular weight markers were dissolved in 2 mL of eluent per vial and injected into the column.

[0044] Polystyrene nanoparticles of 20nm, 50nm, 100nm, 125nm, and 200nm were each diluted 100-fold with eluent and injected into the column to confirm their elution behavior. Figure 13 shows the chromatograms of each. From the top, they are 20nm, 50nm, 100nm, 125nm, 200nm, and a protein molecular weight marker. Figure 14 shows all the obtained chromatograms superimposed on the same axis.

[0045] As can be seen from this, the substances eluted in the following order: PS_125nm, PS_100nm, PS_50nm, PS_20nm, yeast glutamate dehydrogenase (molecular weight 290,000), porcine myocardial lactate dehydrogenase (molecular weight 142,000), yeast enolase (molecular weight 67,000), yeast myokinase (molecular weight 32,000), and myocardial cytochrome c (molecular weight 12,400), indicating that the substances eluted were determined by their "bulk."

[0046] However, no elution of polystyrene nanoparticles with a nominal diameter of 200 nm was confirmed. This is presumed to be because they are trapped in the sintered filter (nominal diameter 2 μm) installed at the inlet of the analytical column or cannot pass through the packed gel. In this measurement system, the particle size that can be eluted is presumed to be around 150 nm (the "non-eluted region" in Figure 8a).

[0047] Moreover, the reason why the peak intensity differs greatly depending on the polystyrene nanoparticle size is because scattered light is detected at low wavelengths, not the absorption of polystyrene (see Figure 6). Polystyrene itself is a styrene polymer that has ultraviolet absorption, as shown in Figure 1, and in organic size exclusion chromatography, it is a polymer that is often dissolved in tetrahydrofuran (THF) and used as a molecular weight marker. In the dissolved state, there is strong absorption in the ultraviolet light region around 254 nm, but since polystyrene nanoparticles are suspended in an aqueous solution, they do not exhibit strong absorption (see Figure 5).

[0048] First, the relationship between the elution time and the logarithm of the molecular weight of the five proteins included in the protein molecular weight markers is shown in Figure 15a. In general size exclusion chromatography, components with larger molecular weights elute earlier. Furthermore, the elution time and the logarithm of the molecular weight can be approximated by a cubic function within a certain range. Components with molecular weights above a certain level almost always pass through the column (exclusion limit). In the case of this separation system, the elution time is 10 5.5 ~10 6 can be approximated by the following cubic equation with an exclusion limit (Equation 1).

[0049]

number

[0050] Proteins have different shapes, but if we assume that they are "globular," we can calculate their approximate size from their molecular weight, etc. (Assumed) Shape: Perfect sphere Specific gravity: 1.2g / cm 3 For example, Cytochrom C, which has a molecular weight of 12.4 kD, The weight per molecule is

[0051]

number

[0052] (Equation 2) The volume per molecule is

[0053]

number

[0054] (Equation 3) The volume of the sphere is

[0055]

number

[0056] 4, the approximate radius of the protein is

[0057]

number

[0058] (Equation 5) Similarly, calculations for other proteins result in Table 9.

[0059] [Table 9]

[0060] The relationship between the logarithm of the protein size (radius) and the elution time obtained here can also be approximated by a cubic function, similar to the molecular weight mentioned above (Figure 15b, Equation 6).

[0061]

number

[0062] Next, the relationship between the elution time and the logarithm of particle size (radius) for polystyrene nanoparticles of 20 nm, 50 nm, 100 nm, and 125 nm is shown in Figure 16 (with ● as legend). As with the protein mentioned above, the larger the particle, the earlier it is eluted, and this can be approximated by a cubic function (Equation 7).

[0063]

number

[0064] Figure 16 shows the relationship between the logarithm of the size (radius) of the protein obtained above and the elution time (legend: ◯), and the relationship between the logarithm of the size (radius) of the polystyrene nanoparticles and the elution time (legend: ●) on the same axis.

[0065] A typical SEC calibration curve shows a linear relationship within a certain molecular weight range, rising sharply near the exclusion limit and then dropping sharply near the permeation limit. It is difficult to express this with a simple, single polynomial function, so the calibration curve is created by approximating the interval with a linear equation, or by taking a slightly longer interval and approximating with a cubic equation.

[0066] Since nanoparticles are eluted near the exclusion limit, the calculation range becomes even wider, and cubic approximation does not provide an accurate fit. The same problem exists with the general SEC method, and other methods have been used, such as combining hyperbolic functions and polynomial functions to fit the steep rise in the area near the exclusion limit.

[0067] Figure 17 shows the chromatogram and calibration curve of the molecular weight standard sample obtained under the conditions in Table 10. Note that this data was obtained using a high-speed GPC device "HLC-8420GPC" manufactured by Tosoh Corporation, and details are omitted here. Figure 17a shows the chromatogram of two types of standard sample mixture, Figure 17b shows the calibration curve approximated by a simple cubic equation (horizontal axis: elution time, vertical axis: logarithm of molecular weight), and Figure 17c shows the calibration curve approximated by a combination of a cubic equation and a hyperbolic function. The right axis shows the error of the approximation result.

[0068] The error of the approximation is expressed as the square of (logarithm of the molecular weight)-(logarithm of the molecular weight value obtained by approximation)) for each standard sample.

[0069] Cubic formula: y=Ax 3 +Bx 2 +Cx+D Cubic + Hyperbolic Function: y=Ax 3 +Bx 2 +Cx+D+(E / Lx) 2 x: Time, y: Log (molecular weight) A, B, C, D, E: Coefficients L: Exclusion limit (constant, user input) As can be seen from FIG. 17b, in the cubic approximation, the approximation error becomes larger in the region of early elution time, i.e., in the region of high molecular weight. This is because, under these conditions, the molecular weight is 10 7 The exclusion limit is located near this region, and according to the principles of SEC, the calibration curve rises significantly in this region, but this is because the approximation is not accurate.

[0070] When a cubic expression + hyperbolic function is combined as in Figure 17c, the rise of the calibration curve in this region can be made larger and the approximation error reduced. At first glance, using a cubic expression + hyperbolic function seems to be a good idea, but the constant L in the hyperbolic function part represents the elution time of the exclusion limit and must be specified by the user. Even a slight change in the L value can cause a large change in the calibration curve profile.

[0071] [Table 10]

[0072] In this verification, by using the Biphasic Dose Response Function (double Boltzmann function) as an approximation, an approximation was made that could express the exclusion limit with a single function. This function is as shown in equation 8 and is composed of seven constants. Table 11 shows the calculated constants. This approximation calculation was performed using "Origin 2021b," a graph creation and data analysis software made by Lightstone.

[0073]

number

[0074] [Table 11]

[0075] Next, the antibody drug "Rituxan" used above was subjected to heating / cooling stress to generate aggregates and measure their size. As shown in Figure 18, the stress was applied by heating at 60°C for 30 minutes, cooling at 5°C for 15 minutes, and then subjecting to size exclusion chromatography. After separation by size exclusion chromatography (SEC method) was completed, the remaining sample was heated at 60°C for 30 minutes, cooled at 5°C for 15 minutes, and measured by SEC method. This operation was repeated to increase the stress. Measurements were repeated until the total heating time reached 120 minutes. The changes in the chromatogram are shown in Figure 19. Note that the time notations in Figure 19b indicate the total heating time.

[0076] In addition to the main component, multiple peaks were confirmed for the "Rituxan" used in this verification. As can be seen from Table 2, the large peak observed at around 32.6 minutes is presumed to be the active ingredient "rituximab," and the sharp peak observed at around 45.7 minutes is presumed to be the additive polysorbate. In addition, the three small peaks observed from 26 to 29 minutes are presumed to be aggregates of "rituximab" that are originally present in this drug.

[0077] By applying heat / cool stress, a new peak appears around 22 minutes and gradually becomes stronger in intensity. This is presumably a larger aggregate component of the drug (Figure 19b). Figure 20 shows the change in area of ​​the main peak of the drug (left axis), the peak of the originally existing aggregate (29.3 minutes) (right axis), and the peak of the larger aggregate (22 minutes) generated by stress (right axis) versus the total heating time.

[0078] In addition, Figure 21 shows the area ratio of the peak area eluted up to about 37 minutes. Here, the three aggregates originally present at 26 to 29 minutes are shown as the sum total. As can be seen from this, the number of large molecular-sized aggregates that appeared around 22 minutes increases with each application of heating / cooling stress.

[0079] Next, the molecular size of the representative peak detected was calculated from the calibration curve obtained above. The elution time of the main component peak (rituximab) eluted at approximately 32.7 minutes remains almost unchanged even when heating / cooling stress is applied. In other words, the molecular size does not change. The molecular weight was calculated to be 2.05 × 10 5 The formula for Rituxan is 1.45 x 10 5 The values ​​are relatively consistent (see Table 2). The molecular size is calculated to be about 7.6 nm in diameter (3.8 nm in radius) (see Table 11).

[0080] It can also be seen that the elution time of the peak generated by applying heat / cooling stress at approximately 22 minutes changes slightly depending on the degree of heat / cooling stress. When heat / cooling stress is applied, the elution time tends to become earlier, in other words, the molecular size becomes slightly larger. From the above calibration curve, the molecular size is calculated to be approximately 19-20 nm in diameter (radius 11-15 nm) (see Table 12). In other words, it is strongly suggested that the new component derived from Rituxan that appeared due to heat / cooling stress is a component of the same size as the standard 20 nm polystyrene nanoparticles.

[0081] [Table 12]

[0082] [Table 13]

[0083] Example 2 In this example, molecular exclusion chromatography was performed using a resin-based column. The analytical columns were two TSKgel G3000PWXL (inner diameter 7.8 mm × length 30 cm) manufactured by Tosoh Corporation, packed with a methacrylate polymer-based gel, arranged in series.

[0084] The eluent was based on 2-[4-(2-hydroxyethyl)-1-piperazinyl]ethanesulfonic acid (hereinafter referred to as HEPES) buffer, to which a small amount of Brij35 (polyoxyethylene (23) lauryl ether), a non-ionic surfactant made by Kishida Chemical, was added in order to elute the suspended nanoparticles. Details of the HEPES buffer are as follows.

[0085] HEPES: 20 mmol / L Sodium chloride: 300 mmol / L pH:7.0 As samples, polystyrene nanoparticles, protein molecular weight markers, and Rituxan, an antibody drug, which was used as a protein, were used as in Example 1 (see Example 1 for details).

[0086] First, the effect of adding a surfactant was verified in the same way. Brij35 was added to the HEPES-based eluent at 0.5%, and the change in elution was confirmed. Figure 22 shows the chromatogram obtained by injecting 20 nm polystyrene nanoparticles and the antibody drug "Rituxan" into a polyacrylate resin-based gel SEC column.

[0087] FIG. 22a shows the case where Brij35 is contained at 0.5%, and FIG. 22b shows the case where Brij35 is not contained. As can be seen from this, like the silica-based gel of Example 1, the polyacrylate resin-based gel also shows that if Brij35 is not contained, both the polystyrene nanoparticles and the antibody drug "Rituxan" cannot be eluted. FIG. 23 shows a chromatogram obtained by preparing 20 nm and 100 nm polystyrene nanoparticles and a protein molecular weight marker (a mixture of five types) as one sample and injecting it into a SEC column of a polyacrylate resin-based gel. The peaks eluting from 20 to 24 minutes are the 100 nm and 20 nm polystyrene nanoparticles, and the five peaks eluting from 25 to 32 minutes are the protein molecular weight markers (see Table 14 and FIG. 23).

[0088] As in Example 1, the method of the present invention makes it possible to elute suspended polystyrene nanoparticles and proteins under the same analytical conditions, demonstrating that they are separated based on the principle of "bulkiness," i.e., molecular exclusion chromatography.

[0089] 24 is a calibration curve in which the horizontal axis indicates elution time and the vertical axis indicates the logarithm of the size of the protein and polystyrene nanoparticles. The approximation was performed using the Biphasic Dose Response Function (double Boltzmann function) of Equation 1, as in Example 1. Table 15 shows the approximation results.

[0090] [Table 14]

[0091] [Table 15]

[0092] As described above, according to the present invention, even when molecular exclusion chromatography is performed using a silica-based gel or a polyacrylate resin-based gel, it is possible to separate proteins soluble in an aqueous solution and nanoparticles suspended in the aqueous solution using an eluent of the same composition, without any ultraviolet absorption in the measurement range, and by adding a small amount of nonionic surfactant.

[0093] In addition, with the usual SEC method, because standard samples cannot be obtained near the exclusion limit, the calibration curve is estimated using a cubic equation + hyperbola. However, because suspended nanoparticles that dissolve near the exclusion limit can also be measured, a more accurate calibration curve can be obtained.

[0094] Furthermore, because it is possible to measure proteins and suspended nanoparticles simultaneously, while conventional SEC methods were used to calculate "molecular weight," this method makes it possible to also calculate molecular bulk (particle size). [Explanation of symbols]

[0095] 1. Eluent 2. Deaeration device 3. Liquid delivery pump 4. Sample injection mechanism 5.Analytical column 6. Constant temperature bath 7. UV-Visible Detector

Claims

1. Nanoparticles having a particle size of 10 nm to 150 nm and proteins, The method is characterized in that the analysis is performed using an aqueous eluent containing one or more nonionic surfactants. Size exclusion chromatography.

2. The composition of the eluent is as follows: The concentration of the nonionic surfactant is 0.02 to 2%.

2. The method of claim 1, wherein the size exclusion chromatography is performed.

3. The nonionic surfactant has no absorption in the ultraviolet absorption region of 220 nm to 380 nm.

3. Size exclusion chromatography according to claim 1 or 2.

4. The nonionic surfactant is Characterized in that the surfactant is selected from one or more of the following types of nonionic surfactants:

3. Size exclusion chromatography according to claim 1 or 2. 1) Higher alcohol ethylene oxide adduct 2) Polyoxyethylene fatty acid esters (fatty acid ethylene oxide adducts, polyethylene glycol fatty acid esters) 3) Higher alkylamine ethylene oxide adducts 4) Fatty acid amide ethylene oxide adducts (polyoxyethylene alkylamines, polyoxyethylene fatty acid amides) 5) Polypropylene glycol ethylene oxide adduct (Pluronic (registered trademark) type nonionic surfactant, polyoxyethylene polyoxypropylene glycol) 6) Glycerol fatty acid ester 7) Pentaerythritol fatty acid esters 8) Sorbitol fatty acid esters 9) Sorbitan fatty acid esters 10) Sucrose fatty acid ester 11) Alkyl polyglycosides 12) Fatty acid alkanolamides

5. The nonionic surfactant contains at least polyoxyethylene (23) lauryl ether.

3. Size exclusion chromatography according to claim 1 or 2.

6. The gel matrix of the size exclusion chromatography is silica-based or polyacrylate resin-based.

3. Size exclusion chromatography according to claim 1 or 2.

Images