Method for non-covalent in-migratio fluorophore labeling of SDS-proteins using counter-current electromigration in SDS-cge

Abstract

The present invention relates to a method for countercurrent electromigration-mediated non- covalent fluorophore labeling of SDS-proteins in sodium dodecyl sulfate capillary gel electrophoresis (SDS-CGE), wherein the gel-buffer system contains fluorophore labeling agent, preferably propidium iodide.

Description

[0001] METHOD FOR NON-COVALENT IN-MIGRATIO FLUOROPHORE LABELING OF SDS- PROTEINS USING COUNTER-CURRENT ELECTROMIGRATION IN SDS-CGE

[0002] FIELD OF THE INVENTION

[0003] The present invention relates to a method for counter-current electromigration-mediated non- covalent fluorophore labeling of SDS-proteins in sodium dodecyl sulfate capillary gel electrophoresis (SDS-CGE), wherein the gel-buffer system contains fluorophore labeling agent, preferably propidium iodide.

[0004] DESCRIPTION OF THE RELATED ART

[0005] Over the past decade, the field of therapeutic protein development has experienced remarkable progress, with a wide range of biopharmaceuticals having a high impact by entering the pharmaceutical market {Wang, 2022 #26}. Cytokines, enzymes, monoclonal antibodies, and growth factors among others are becoming more and more available and significantly transformed the way diseases are treated and controlled. With the increasing demand for therapeutic proteins, there is a growing requirement for accurate and reliable analytical methods to assess the quality and purity of biopharmaceuticals.

[0006] In recent years, several bioanalytical techniques such as high-performance liquid chromatography and capillary electrophoretic have been routinely used to characterize therapeutic proteins. Sodium dodecyl sulfate capillary gel electrophoresis (SDS-CGE) in combination with laser-induced fluorescence (LIF) detection is one of the most frequently applied methods for rapid protein characterization with high sensitivity {Lacroix, 2005 #4}. While the innate fluorescence of several amino acid building blocks like tyrosine and tryptophan can be utilized in fluorescent detection, the achievable detection limit is not always satisfactory {Koutny, 1993 #22}. Therefore, techniques involving covalent labeling of proteins using fluorescent dyes that react with the amine or thiol groups and methods based on non-covalent interactions between the fluorophore and the protein molecule have recently gained increased attention {Colyer, 2000 #29}. In addition, the process of fluorescent labeling by covalent or non-covalent interaction can be further categorized into three different subtypes of i) pre-column, ii) on-column and iii) post-column labeling. Pre-column labeling involves tagging the analyte molecules with the fluorophore before they enter the separation capillary. On-column labeling is characterized by utilizing a direct interaction of the labeling reagents with the analyte molecules within the separation capillary. On the post-column labeling scenario, labeling of the solute molecules occurs after their separation is completed, i.e., at the outlet end of the capillary {Lee, 1998 #10}.

[0007] Some of the most frequently used covalent labeling dyes are naphthalene-2,3-dicarboxaldehyde (NDA) {Kaneta, 2009 #5 }, 3-(2-(2-furoyl)quinoline-2-carboxaldehyde (FQ) {Michels, 2007 #1 }, fluorescein isothiocyanate (FITC) {Nguyen, 2016 #24} {Emonts, 2021 #25} and the Chromeo dyes {Duhamel, 1983 #23} {Ramsay, 2009 #27}. A study by Michels et al. introduced an efficient sample preparation protocol for therapeutic protein analysis using sodium dodecyl sulfate capillary gel electrophoresis with LIF detection, especially focusing on the evaluation of the purity of recombinant monoclonal antibodies. This method involved labeling therapeutic proteins with the fluorophore dye FQ in the presence of a nucleophile, eliminating the need for purification to remove the excess dye, and achieving low detection limits {Michels, 2007 #1 }. Dovichi and coworkers have developed an on -column electrophoretically mediated derivatization method for protein labeling by FQ for native gel electrophoresis of proteins {Lee, 1998 #6}. The challenge of covalent labeling is that a single protein molecule may contain numerous reactive sites, all of which have the potential to interact with the fluorescent labeling agent. In other words, it is difficult to uniformly label all of the available sites because of the differences in reactivity among them. Consequently, the covalent labeling approach can result in a mixture containing unlabeled, fully labeled, and partially labeled proteins. Separation of such a mixture can lead to peak broadening or even the appearance of multiple peaks of the same sample component {E. Oldenburg, 1997 #28}. To overcome this problem, Kaneta and co-workers introduced a post-column derivatization technique for the analysis of proteins by capillary gel electrophoresis (CGE) using NDA in the presence of the 2 -mercaptoethanol reducing substance in the fluorophore derivatization reaction {Kaneta, 2009 #9}.

[0008] On the non-covalent labeling side, the negatively charged Sypro Red and Sypro Orange offer distinct advantages over the covalent amine- or thiol-reactive counterparts by simply complexing with SDS-proteins, thereby not requiring a chemical reaction {Steinberg, 1996 #10} {Harvey, 1998 #2} {Moody, 1999 #3}. This approach was even utilized in two-dimensional electrophoresis-based protein analysis using ultra-thin layer gels combined with imaging of the separated analytes in real time taking advantage of on-column fluorescence staining with Sypro Red that rapidly co-migrated, overpassing the sample molecules {Guttman, 2002 #9}. A microchip electrophoresis platform utilizing non-covalent fluorophore labeling was also developed for protein analysis, able to separating proteins with molecular weights ranging from 5 kDa to 250 kDa and allow a wide range of protein loading capacities {Bousse, 2001 #7}. Non-covalent fluorophore tagging of biopolymers with the positively charged ethidium or propidium ions has long been reported in DNA analysis via intercalation mechanism, with only a few examples to visualize proteins by ethidium bromide staining after SDS polyacrylamide slab gel electrophoresis separations {Vincent, 1979 #8}.

[0009] THE DISCOVERY ACCORDING TO THE PRESENT INVENTION

[0010] To overcome the problem caused by the covalent labeling of proteins, the inventors of the present invention have performed significant experimental work, which has resulted in the present invention. The present invention is based on the unexpected finding that propidium ions are suitable as a fluorophore labeling agent in capillary gel electrophoresis (CGE) which can be used in the separation matrix. In the present invention, when the proteins are separated by SDS-CGE technique, whereas the proteins are subjected to electromigration in a separation matrix containing propidium iodide, the propidium ions bind to the negatively charged SDS-protein complexes in an equilibration reaction during the separation process in a non-covalent manner as a general biomass stain, thereby obtaining SDS-protein-propidium adduct having strong fluorescence.

[0011] BRIEF DESCRIPTION OF THE INVENTION

[0012] The invention relates to technical solutions and preferred embodiments defined under the following numbered paragraphs.

[0013] 1. A method for separation of proteins carried out by sodium dodecyl sulfate capillary gel electrophoresis (SDS-CGE) using fluorescence detection, wherein the separation matrix comprises fluorophore labeling agent having positive charge, which forms non-covalently fluorophore labeled SDS- protein adducts.

[0014] In an embodiment of the method according to the present invention the fluorophore labeling agent has double positive charge.

[0015] In a preferred embodiment of the method according to the present invention the fluorophore labeling agent is propidium iodide.

[0016] In an embodiment of the method according to the present invention the fluorescence detection is laser-induced fluorescence (LIF) detection.

[0017] 2. The use of a fluorophore protein labeling agent having positive charge which binds with non- covalent binding to the proteins in the capillary in sodium dodecyl sulfate capillary gel electrophoresis (SDS-CGE).

[0018] In an embodiment of the use of the fluorophore protein labeling agent according to the present invention the fluorophore labeling agent has double positive charge.

[0019] In a preferred embodiment of the use of the fluorophore protein labeling agent according to the present invention the fluorophore labeling agent is propidium iodide.

[0020] In an embodiment of the use of the fluorophore protein labeling agent according to the present invention the fluorescence detection is laser-induced fluorescence (LIF) detection.

[0021] 3. Sodium dodecyl sulfate (SDS) capillary gel separation matrix (for use in SDS capillary gel electrophoresis (SDS-CGE)) comprising a fluorophore labeling agent having positive charge wherein said labeling agent binds with non-covalent binding to the proteins in the capillary.

[0022] In an embodiment of the separation matrix according to the present invention the fluorophore labeling agent has double positive charge.

[0023] In a preferred embodiment of the separation matrix according to the present invention the fluorophore labeling agent is propidium iodide.

[0024] In an embodiment of the separation matrix according to the present invention the fluorescence detection is laser-induced fluorescence (LIF) detection.

[0025] 4. A kit for use in separation of proteins carried out by capillary gel electrophoresis (CGE), said kit comprising a separation matrix, preferably sodium dodecyl sulfate (SDS) capillary gel separation matrix and a fluorophore labeling agent, wherein the separation matrix comprises the fluorophore labeling agent, and wherein the fluorophore labeling agent has positive charge which binds with non-covalent binding to the proteins in the capillary.

[0026] In an embodiment of the kit according to the present invention the fluorophore labeling agent has double positive charge.

[0027] In a preferred embodiment of the kit according to the present invention the fluorophore labeling agent is propidium iodide.

[0028] In an embodiment of the kit according to the present invention the fluorescence detection is laser- induced fluorescence (LIF) detection.

[0029] BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1: Schematic illustration of an electropherogram of the intact (mAh) and subunit fragments (LC, ngHC and HC) of daratumumab obtained by an embodiment of the present invention. The formula represents the structure of propidium iodide used as a non-covalently bound fluorophore labeling agent. IS is an internal standard.

[0031] Figure 2: Sodium dodecyl sulfate capillary gel electrophoresis of the intact (mAh) and subunit fragments (LC, ngHC and HC) of daratumumab. Lower trace: UV detection (220 nm) with no propidium iodide in the separation matrix. Upper trace: laser induced fluorescent (LIF) detection with the use of 100 pg / mL propidium iodide containing gel-buffer system. Peaks: 1 - lysozyme (internal standard), 2 - light chain (LC), 3 - non-glycosylated heavy chain (ngHC), 4 - heavy chain (HC) and 5 - intact daratumumab (mAb); * heavy-heavy-light chain fragment impurity of the intact mAh. Conditions: 10% dextran (2M) / 4% boric acid-based sieving matrix, 20 cm effective length (30 cm total, 50 pm ID) BFS capillary, Applied electric potential: 15 kV in reversed polarity mode (cathode at the injection side); separation temperature: 25°C; Sample tray temperature: 20°C, Electrokinetic injection: 5 kV / 20 s for the denatured intact mAh, followed by 5 kV / 20 s for the reduced and denatured mixture of lysozyme, LC, ngHC and HC fragments of daratumumab.

[0032] Figure 3: Effect of the increasing propidium iodide concentration in the gel-buffer system on the migration time and peak area of the sample components. The numbers on the traces represent the propidium iodide concentration (pg / mL) in the gel-buffer system. (A) UV detection at 220 nm, (B) Laser induced fluorescent detection (excitation 488 nm / emission 600 nm). Conditions: same as in Figure 2 with the propidium iodide concentrations specified on the individual traces.

[0033] Figure 4: Relationship between the peak areas of Figure 3 and the propidium iodide concentration in the gel-buffer system. Dotted lines: UV 220 nm detection (starting at zero propidium iodide concentration); Solid lines: LIF detection (starting at 10 pg / mL propidium concentration). Figure 5: Effect of the capillary temperature on the separation of the sample components. Peaks and conditions were the same as in Figure 2 with 100 pg / mL PI in the gel -buffer system and the temperatures are given on the traces.

[0034] Figure 6: Relationship between the separation temperature and resolution between the consecutively migrating sample components using 100 pg / mL propidium iodide containing gel-buffer system.

[0035] Figure 7: Arrhenius (panel A) and activation energy (panel B) plots of the SDS-protein-propidium complexes of the lysozyme internal standard (*) as well as the light chain (A), non-glycosylated heavy chain (□), heavy chain (0) subunits, and the intact mAh (o), based on the data derived from Figure 5.

[0036] ABBREVIATIONS

[0037] SDS = Sodium Dodecyl Sulfate,

[0038] CGE = Capillary Gel Electrophoresis,

[0039] EOF = Electroosmotic Flow,

[0040] LIF = Laser-Induced Fluorescence,

[0041] PI = Propidium Iodide,

[0042] BFS = Bare Fused Silica,

[0043] BGE = Background Electrolyte, mAh = Monoclonal Antibody,

[0044] LC = Light Chain, ngHC = Non-Glycosylated Heavy Chain,

[0045] HC = Heavy Chain,

[0046] DEFINITIONS

[0047] The term “sample” is meant herein to refer to a substance comprising a mixture of compounds prepared separation, e.g. for analysis, e.g. by capillary electrophoresis. The sample may be derived from, for example, a substance obtained from an environmental source, e.g. bodily fluid or tissue from a subject, i.e. taken from said bodily fluid or tissue of the subject and optionally processed to prepare for analysis. The sample as used herein preferably comprises proteins, preferably proteins of biological or biotechnological interest, as defined, described or exemplified herein.

[0048] A “fluorescent” or a “fluorescently labeled” compound as used herein is a compound which can be detected by irradiating with an UV or VIS electromagnetic radiation and the compound absorbs the irradiating light and emits light (emitted light) at another, preferably longer wavelength than that of the irradiating light. Preferably the fluorescent labeling agents have a positive charge, in a particularly preferred embodiment double positive charge. The fluorescent labeling agents, which are present in the gel-buffer system, are non-covalently bound to SDS-protein complexes during the separation process. “Detecting” as used herein is understood broadly as obtaining an observation regarding a substance or compound of interest (preferably an analyte), as a result of the separation method on a sample.

[0049] A “detector” is a device for detecting and which is located typically in connection with a specific site of the CE capillary to detect compounds of interest migrating therein. Preferably the operating principle of the detector is based on fluorescence detection, more preferably laser-induced fluorescence (LIF) detection.

[0050] The term “comprises” or “comprising” or “including” are to be construed here as having a non- exhaustive meaning and allow the addition or involvement of further features or method steps or components to anything which comprises the listed features or method steps or components. “Comprising” can be substituted by “including” if the practice of a given language variant so requires or can be limited to “consisting essentially of’ if other members or components are not essential to reduce the invention to practice, or “consisting of’ in case no other elements or components are present.

[0051] The singular forms “a”, “an” and “the”, or at least “a”, “an”, include plural reference unless the context clearly dictates otherwise.

[0052] DETAILED DESCRIPTION OF THE INVENTION

[0053] The present invention relates to a novel in-migratio staining concept by adding propidium iodide to the separation matrix in SDS-CGE of proteins. The positively charged propidium ion, as a component in the gel-buffer system, migrates against the negatively charged SDS -proteins in the capillary and forms non-covalent fluorescent adducts in situ during the separation process.

[0054] The present inventors have introduced a method for counter-current electromigration-mediated non-covalent fluorophore staining in migration staining), where electromigrating proteins are analysed in a separation matrix containing propidium iodide. The propidium ions having two positive charges bind to the negatively charged SDS-protein complexes in an equilibration reaction during the separation process in a non-covalent manner as a general biomass stain, providing strong fluorescence to the resulting SDS- protein-propidium adducts. Due to the double positive charge of the propidium ion, the surface charge density of the resulting SDS-protein-propidium adducts is decreased, concomitantly, reducing the electrophoretic mobility of the complex. In addition, the propidium ions neutralize the negatively charged silanol groups at the inner surface of the bare fused silica capillary, thereby reducing the counter-current electroosmotic flow (EOF) to almost zero, which is beneficial for reducing the separation time. Based on the results detailed below, in the case of the gel-buffer systems containing propidium iodide, the electrophoretic mobility of the intact biotherapeutic protein used as an example, and all its subunits, as well as the non-glycosylated lysozyme used as an internal standard, was somewhat reduced compared to the electrophoretic mobility of the sample components separated without fluorophore staining, but was detected with high sensitivity. To achieve the highest possible separation capability, the concentration of propidium iodide and the separation temperature can be optimized for a given sample components.

[0055] Propidium iodide is one of the frequently used non-covalent fluorophore labels for DNA analysis. Propidium iodide, once predominantly recognized for its DNA and RNA staining capabilities, has ventured beyond its original scope to offer profound insights into protein research.

[0056] In support of the present invention, the inventors have described a novel in migratio based SDS- protein staining approach by successfully utilizing propidium iodide as a non-covalent fluorophore agent in SDS-CGE separation of the intact and subunit forms of a therapeutic anti-cancer monoclonal antibody daratumumab. The two positive-charge holding propidium ion (P2+) apparently binds to the negatively charged SDS-protein complexes (Sn) in an equilibration reaction during the separation process in a non- covalent manner as a general biomass stain {Rocha-Santos, 2014 #20}, providing strong fluorescence to the resulting SDS-protein-propidium adduct. Due to the double positive charge of the propidium ion, the surface charge density of the resulting SDS-protein-propidium adduct is decreased to SP(n 2m), concomitantly, reducing the electrophoretic mobility of the complex.

[0057] S” + mP2+sp(n 2m) (Equation 1)

[0058] K = [SP(n 2m)] / [Sn] [P2+]m(Equation 2) where K is the complex formation constant, n is the number of the negatively charged SDS molecules bound to the protein (approximately one SDS per 2 amino acids {Shirahama, 1974 #19}) and m is the number of the complexing propidium ions in the SDS-protein-propidium adduct.

[0059] Considering the very low counter-current electroosmotic flow (- 1.8 / 101(1m2 / V s) in the high viscosity borate cross-linked 2 M molecular weight dextran polymer based no propidium iodide - containing sieving matrix filled capillary {Guttman, 2021 #21 }, the effective mobilities of the SDS- protein molecules (ps) can be calculated as the algebraic sum of their apparent (papp) mobility and the electroosmotic flow ( EOF): ps = Papp + PEOF (Equation 3)

[0060] At the pH of the SDS-CGE buffer, the electroosmotic flow was cathodic; therefore, the EOF was counter-current to the electromigration of the negatively charged SDS-protein complexes (anodic migration). Thus, the value of the electroosmotic flow mobility ( EOF) was added to the apparent mobility (papp) to determine the effective SDS-protein mobility, since the two mobilities are of opposite signs. Without EOF, all solute molecules would have migrated faster. The migration of the charged gel matrix under the applied electric field was insignificant, and so, was not considered in Equation 1.

[0061] However, with the presence of the positively charged propidium ions in the gel-buffer system, the EOF was further decreased via neutralization of a portion of the negatively charged silanol groups at the inner surface of the bare fused silica capillary by the positively charged fluorophore. In addition, the actual electrophoretic mobility (p) of the solute molecules is also less than that of their no-propidium- containing counterparts due to the complexed positively charged propidium ions: i = is Rs (Equation 4) where Rs is the molar ratio of the propidium-free SDS-proteins. By introducing cs as the total SDS- protein concentration, Rs can be expressed as:

[0062] Rs = [Sn] / cs= 1 / (l+K[P2+]m) (Equation 5)

[0063] Thus, the resulting mobility of the negatively charged SDS -protein -propidium complex (with the double positive charge holding propidium molecule) is as follows: p = ps / (l+K[P2+]m) (Equation 6)

[0064] Equation 6 suggests that increasing the complex formation constant and / or the propidium concentration would result in reduced electrophoretic mobility of the SDS-protein-propidium complex. In case when K[P2+]mis significantly greater than unity, Equation 6 can be simplified as: p = ps / K[P2+]m(Equation 7)

[0065] An even more interesting feature in the LIF profile is the biased peak area increase in some of the sample components, especially visible for the lysozyme (peak 1) and the intact monoclonal antibody (peak 5) components in comparison to the UV profile (see Example 2). Lysozyme is a very basic protein with a pl of 11.35, therefore probably not only binds the SDS molecules via hydrophobic interaction but maybe by ionic bonds as well. In this latter instance the hydrophobic tail of the SDS molecules could attract the large planar propidium molecule, forming close van der Waals contacts, i.e., increasing the number of bound fluorophores, and concomitantly resulting in greater fluorescent signal. Regarding the intact non-reduced mAh, while denatured, i.e., lost its secondary structure, thus might have extra hydrophobic patches in comparison to the reduced subunits and binds more fluorophore.

[0066] The electromigration of poly-ionic biopolymers in a sieving medium is an activated process {Guttman, 2015 #12; Cottet, 2001 #13; Lu, 1994 #14} in which each solute molecule requires a certain activation energy (Ea) to traverse through the reticulations. Based on the modified Arrhenius equation {Arrhenius, 1889 #15; Eyring, 1935 #16}, the present inventors has earlier developed a mobility equation for SDS-protein separations in borate cross-linked dextran gels within narrow bore capillaries {Filep, 2020 #17} that can be used to assess the required activation energies for the sample components. p = Q Mw1 / 6eEa / RT(Equation 8) where Q and Mw are the overall charge and molecular weight of the solute, respectively, Eais the activation energy, R is the universal gas constant and T is the absolute temperature. Combining Equations 7 and 8 can shed light on the function of the ligand concentration and the molecular characteristics of the analyte molecules including Mw, charge and complexation constant in view of the activation energy concept.

[0067] Q Mw1 / 6eEa / RT= ps / K[P2+]m(Equation 9) ln(Q Mw1 / 6) - Ea / RT = ln(ps / K[P2+]m) (Equation 10)

[0068] Ea= RT ln(K[P2+]m Q / psMwl / 6) (Equation 11) Equation 11 suggests that under isotherm separation conditions, the activation energy is a direct logarithmic function of the complex formation constant, the number of bond propidium molecules and the overall charge of the analyte but a reciprocal logarithmic function of the electrophoretic mobility of the SDS-protein complex and the molecular weight of the solute on the l / 6thpower.

[0069] Based on the forgoing, the aspect of the invention is a method for separation of proteins carried out by sodium dodecyl sulfate capillary gel electrophoresis (SDS-CGE) using fluorescence detection, wherein the separation matrix comprises fluorophore labeling agent having positive charge, which forms non- covalently fluorophore labeled SDS-protein adducts.

[0070] In an embodiment of the method the fluorescence detection is laser-induced fluorescence (LIF) detection.

[0071] In an embodiment of the method the fluorophore labeling agent has double positive charge.

[0072] In a preferred embodiment of the method the fluorophore labeling agent is propidium iodide.

[0073] The invention is further illustrated below by non-limiting examples. The skilled person is aware that based on the examples alternative embodiments or variants thereof may also fulfill the object of the invention.

[0074] EXAMPLES

[0075] Materials used in Examples:

[0076] Boric acid, Tris, EDTA.Na2, glycerol, sodium dodecyl sulfate (SDS), 2-mercaptoethanol, hydrochloric acid and sodium hydroxide were from VWR (Radnor, PA, USA). Dextran (2 MDa), iodoacetamide, and lysozyme (non-glycosylated) were from Sigma-Aldrich (St. Louis, MO, USA). Propidium iodide was from Thermo Fisher Scientific (Waltham, MA, USA). The therapeutic monoclonal antibody daratumumab (Darzalex®) was obtained from Janssen Biotech (Horsham, PA, USA) and the PNGase F enzyme was from the University of Pannonia (Veszprem, Hungary)

[0077] Example 1 - General methods

[0078] Gel-buffer and sample preparation

[0079] For all capillary electrophoresis separations, the 10% Dextran (2 MDa) / 4.0% (w / v) boric acid gel buffer system was used as described in {Guttman, 2021 #21 }. For fluorescent detection, 10 - 200 pg / mL propidium iodide (PI) was also added to the separation matrix in a way that the dilution series were adjusted by the no Pl-containing gel-buffer and stirred overnight in the dark at room temperature. The sample buffer contained 100 mM Tris-HCl, 1% SDS (pH 9). The intact mAh sample was prepared by mixing 5 pL of 20 mg / mL therapeutic monoclonal antibody with 5 pL of 250 mM iodoacetamide and 85 pL of sample buffer. For reduced protein analysis, 5 pL of 20 mg / mL therapeutic monoclonal antibody and 5 pL of 10 mg / mL lysozyme were respectively denatured by the addition of 2 pL of denaturation mixture (Bio-Science Kft, Budapest, Hungary) followed by incubation at 70°C for 15 min. The N-glycans were released from 5 pL of denatured mAb solution by the addition of 2 pL of 200 mU / mL PNGase F enzyme, and incubated for one hour at 37°C. The final reduced sample mixture contained 5 pL of the reduced mAb, 5 pL of the deglycosylated mAb, and 5 pL of 10 mg / mL reduced lysozyme internal protein standard. Both the non-reduced and reduced samples were denatured at 70°C for 15 min.

[0080] Capillary SDS-Gel Electrophoresis

[0081] All separations were performed on a P / ACE MDQ Capillary Electrophoresis System (Beckman Coulter, Brea, CA, USA) controlled by the 32 Karat software (ver 10.1) and equipped with a UV (220 nm) detector for the analysis with non-fluorophore containing and an LIF detector (488 nm excitation with 600 nm emission) for the 10 - 200 pg / mL propidium iodide containing gel-buffer systems. The effective length of the bare fused silica (BFS) separation capillary was 20 cm (30 cm total, 50 pm ID) / 375 pm OD). The capillary was rinsed with 0.5 M NaOH, 0.5 M HC1, and HPLC grade water for 2 min each prior to the separations. Injection parameters were as follows: 5 kV / 20 s intact denatured mAb followed by 5 kV / 20 s for the reduced and denatured sample mixture containing the lysozyme, the light chain, the non-glycosylated heavy chain, and the heavy chain fragments of daratumumab. All measurements were made in triplicates with the average migration time and peak area RSD of 0.18% and 0.78%, respectively.

[0082] Example 2 - Comparing SDS-CGE analysis of intact (mAb) and subunit fragments (LC, ngHC and HC) of daratumumab with UV detection (without propidium iodide) and with LIF detection (with propidium iodide)

[0083] The results of the present example are shown in Figure 2 and Table 1.

[0084] Figure 2 compares the separation of the intact (mAb) and the subunit forms of light chain (LC), non-glycosylated heavy chain (ngHC), and heavy chain (HC) fragments of the therapeutic anti-cancer monoclonal antibody using UV detection with no propidium iodide containing gel-buffer (lower trace) and laser-induced fluorescent detection with 100 pg / mL propidium iodide containing background electrolyte (upper trace). Non-glycosylated lysozyme (14 kDa) was used as an internal standard. As one can observe, the migration time of all solute molecules with the propidium iodide -containing gel-buffer system was slower (upper trace) than their no fluorophore complexed counterparts (lower trace). As a first approximation, we consider this phenomenon caused by the decreased surface charge density with the non-covalently bound positively charged labeling agent (Equation 7, decreasing the electrophoretic mobility) and the reduced counter-current electroosmotic flow (Equation 3, increasing the electrophoretic mobility). The interplay of these two opposite effects resulted in a surprisingly little increase in the migration time in the presence of propidium iodide in the separation medium. Please also note the differences in the heights / areas of the corresponding peaks between the two traces, which were probably due to the different number of fluorophores bound to (LIF detection) and the number of UV active groups (UV detection) of the solute molecules, i.e., one cannot expect to obtain the same profile with these two fundamentally different detection systems. Table 1 depicts the resolution (Rs) and selectivity (a) values of the zero (UV detection) and 100 pg / mL (LIF detection) propidium iodide -containing gel-buffer system-mediated separations, respectively. The resolutions between the consecutively migrating peak pairs were higher with LIF detection, apparently due to the higher selectivity values. Please note that even minor differences in a close to 1.00 have an exaggerated effect on the resolution.

[0085] Table 1: Resolution and selectivity values without (UV detection) and with 100 pg / mL propidium iodide (LIF detection) in the gel-buffer system

[0086] Example 3 - Study of effect of the propidium iodide concentration

[0087] The effect of propidium iodide concentration in the gel -buffer system ranging from 0 (for UV detection) and 10 pg / mL (for LIF detection) to 200 pg / mL was investigated on the migration time and peak area changes of the solute molecules. Figure 3 compares the resulting electropherograms using UV (panel A) and LIF (panel B) detection. Both panels show slightly but continuously increasing migration time (i.e., lower electrophoretic mobility) with higher propidium concentrations, as was predicted by Equation 7 and the decreasing EOF in Equation 3. On the other hand, while with increasing propidium iodide concentration in the background electrolyte the peak areas with UV detection only marginally increased (panel A), significantly greater peak areas were observed with LIF detection (panel B). Please note that above 200 pg / mL propidium iodide in the gel -buffer system, the fluorescent background was high and noisy.

[0088] The plots in Figure 4 compare the peak areas of Figure 3 as the function of propidium iodide concentration with UV (dotted lines) and LIF (solid lines) detection. As suggested above, it only shows a marginal increase in UV detection, e.g., 1.23x between zero and 200 pg / mL PI for the intact mAb ingredient, while displays a considerable increase for the same molecule in LIF detection, e.g., 17x between 10 pg / mL and 200 pg / mL PI concentration in the gel-buffer systems. The steepest slope with LIF detection was obtained for the intact mAb, whereas the plot corresponding to the heavy chain fragment had the shallowest elevation. The other three sample components had approximately similar slopes between the above-mentioned two sample components. Please note that the heavy chain fragment was glycosylated and since the sugar moiety did not bind SDS it led to decreased propidium binding. The LC, ngHC, and lysozyme were all non-glycosylated, so bound the fluorophore according to their very similar SDS-mediated surface charge densities. The intact mAh, on the other hand, albeit glycosylated but the sugar structures faced inward, therefore, did not affect SDS binding {Scanlan, 2008 #11 }. However, due to the loss of secondary structure in the denatured but not reduced mAh molecule, the extra hydrophobic patches were prone to bind additional propidium ions, increasing the fluorescent signal. The limit of detection (LOD) for the lysozyme internal standard was 6 ng / pL, similar to that of the microchip electrophoresis system offers with on-column labeling {Bousse, 2001 #7}.

[0089] Example 4 - Study of effect of the separation temperature

[0090] The effect of capillary temperature on the separation and the peak heights / areas of the sample components were evaluated at every 5°C between 20°C and 40°C as shown in Figure 5. The mobility of all sample components increased with elevated temperature as expected, due to the decreasing viscosity of the sieving matrix. To evaluate the possible temperature dependent PI complexation changes with the SDS-proteins, the resulting relative peak area percent values are listed in Table 2. The trends indicated variable decreasing relative peak areas for the lysozyme (3.5%) as well as the LC (1.9%), ngHC (1.2%) and HC (1.2%) fragments, but increasing tendency for the intact mAh (7.7%), probably as a result of the altered complex formation constant (K in Equations 2, 5 and 6) due to the temperature changes.

[0091] Table 2: Effect of the separation temperature on the peak area % distribution of the sample mixture components in CGE-LIF with 100 pg / mL PI containing gel-buffer system

[0092] The resolution between the sample components was also evaluated as the function of the separation temperature using the 100 pg / mL propidium iodide containing sieving matrix. As one can observe in Figure 6, the resolution between the HC and ngHC slightly increased (3.81%) with elevating temperature. On the other hand, the resolution values between all other consecutively migrating sample components decreased with increasing temperature. The greatest decline was obtained between the LC and the lysozyme (17.86%). This suggested that in addition to considering the propidium iodide concentration dependence, the separation temperature should be also optimized to achieve the best separation performance of the solute components of interest.

[0093] Example 5 - Study of the activation energy requirement

[0094] To obtain the activation energy requirement of the analyte molecules, the Arrhenius plots were generated. The data derived from Figure 5 were plotted in Figure 7 A revealing the relationship between the logarithmic electrophoretic mobility (corrected by 4% for every 5°C temperature increase {Demorest, 1991 #18}) and the reciprocal absolute temperature, as Equation 8 suggested.

[0095] The slopes of the plots in Figure 7A were used to calculate the activation energy values of sample components including the lysozyme internal standard as well as the intact and subunit forms of the mAb. The resulting Eavalues were plotted against the molecular weights of the sample components as shown in Figure 7B. The dotted line is to guide the eye to emphasize the good correlation between the Eaand the Mw of the mAb related, i.e., similar amino acid sequence holding sample components, also emphasizing the outlier position of the lysozyme as an non-relating protein. The distribution of the data points clearly shows the lower activation energy requirement for the high Mw intact mAb and the highest electrophoretic mobility lysozyme molecules as both increase the denominator in Equation 10. Please note that the lysozyme vs LC pair showed the greatest decline in resolution with increasing temperature, again underscoring the significance of capillary temperature optimization during the separation of SDS- protein-propidium complexes.

[0096] INDUSTRIAL APPLICABILITY

[0097] Based on the foregoing findings above, a method for counter-current electromigration-mediated non-covalent fluorophore labeling of SDS-proteins in SDS-CGE, wherein the gel-buffer system contains fluorophore labeling agent, preferably propidium iodide.

[0098] The method according to the present invention allows to eliminate the problem caused by the covalent labeling of proteins due to the fact that proteins may have numerous reactive sites, so that their labeling can result in a mixture containing unlabeled, fully labeled, and partially labeled proteins. Separation of such a mixture can lead to peak broadening or even the appearance of multiple peaks of the same sample component.

[0099] Furthermore, the method according to the present invention provides a better resolution and allows to reduce the counter-current electroosmotic flow (EOF), which is beneficial for reducing the separation time.

[0100] Consequently, the method according to the present invention is suitable for high-efficiency separation of any protein mixture, thus providing a promising method for high-sensitivity detection of biotherapeutic preparations.

[0101] REFERENCES

[0102] (1) Michels D. A., Brady L. J., Guo A., Balland A., “Fluorescent derivatization method of proteins for characterization by capillary electrophoresis-sodium dodecyl sulfate with laser-induced fluorescence detection” Anal. Chem. (2007) 79(15):5963-5971, doi:10.1021 / ac0705521.

[0103] (2) Harvey M. D., Bandilla D., Banks P. R., “Subnanomolar detection limit for sodium dodecyl sulfate- capillary gel electrophoresis using a Anorogenic, noncovalent dye” Electrophoresis. (1998) 19(12):2169-74, doi:10.1002 / elps.H50191221. (3) Moody E. D., Viskari P. J., Colyer C. L., “Non-covalent labeling of human serum albumin with indocyanine green: a study by capillary electrophoresis with diode laser-induced fluorescence detection” J Chromatogr B Biomed Sci Appl. (1999) 729(l-2):55-64, doi:10.1016 / s0378- 4347(99)00121-8.

[0104] (4) Lacroix M., Poinsot V., Fournier C., Couderc F., “Laser-induced fluorescence detection schemes for the analysis of proteins and peptides using capillary electrophoresis” Electrophoresis (2005) 26(13):2608-2621, doi:10.1002 / elps.200410414.

[0105] (5) Kaneta T., Yamamoto D., Imasaka T., (2009). “Postcolumn derivatization of proteins in capillary sieving electrophoresis / laser-induced fluorescence detection” Electrophoresis (2009) 30(21):3780- 3785.

[0106] (6) Lee I. H., Pinto D., Arriaga E. A., Zhang Z., Dovichi N. J., “Picomolar analysis of proteins using electrophoretically mediated microanalysis and capillary electrophoresis with laser-induced fluorescence detection” Anal Chem. (1998) 70(21):4546-8, doi: 10.1021 / ac980360t.

[0107] (7) Bousse L., Mouradian S., Minalla A., Yee H., Williams K., Dubrow R., “Protein sizing on a microchip” Anal Chem. (2001) 73(6): 1207-1212. doi:10.1021 / ac0012492.

[0108] (8) Vincent A., Scherrer K., “A rapid and sensitive method for detection of proteins in polyacrylamide SDS gels: staining with ethidium bromide” Mol Biol Rep. (1979) 5(4):209-214. doi:10.1007 / BF00782890.

[0109] (9) Guttman A, Csapo Z, Robbins D. Rapid two-dimensional analysis of proteins by ultra-thin layer gel electrophoresis. Proteomics. 2002 Apr;2(4):469-74. doi: 10.1002 / 1615-9861(200204)2

[0110] (10) Steinberg T. H., Jones L. J., Haugland R. P., Singer V. L., “SYPRO orange and SYPRO red protein gel stains: one-step fluorescent staining of denaturing gels for detection of nanogram levels of protein” Anal Biochem. (1996) 239(2):223-37, doi:10.1006 / abio,1996.0319.

[0111] (11)

[0112] (12) Guttman A., Kerekgyarto M., Jarvas G., “Effect of Separation Temperature on Structure Specific Glycan Migration in Capillary Electrophoresis” Anal. Chem. (2015) 87(23): 11630-11634.

[0113] (13) Cottet H., Gareil P., “On the use of the activation energy concept to investigate analyte and network deformations in entangled polymer solution capillary electrophoresis of synthetic poly electrolytes. Electrophoresis” (2001) 22(4):684-691. doi:10.1002 / 1522-2683(200102)22:4<684::AID-

[0114] ELPS684>3.0.CO;2-O.

[0115] (14)

[0116] (15) Arrhenius S., “Uber die Dissociationswarme und den Einfluss der Temperatur auf den Dissociationsgrad der Elektrolyte.” Zeitschrift fur Physikalische Chemie 4U (1889): 116-196.

[0117] (16) Eyring H., “The Activated Complex in Chemical Reactions” J. Chem. Phys. (1935) 3: 107-115. (17) Filep Cs., Gutman A., “The Effect of Temperature in Sodium Dodecyl Sulfate Capillary Gel Electrophoresis of Protein Therapeutics” Anal. Chem. (2020) 92:4023-4028

[0118] (18) Demorest D., Dubrow R., “Factors influencing the resolution and quantitation of oligonucleotides separated by capillary electrophoresis on a gel-filled capillary” Journal of Chromatography A (1991) Journal of Chromatography A, 559(1 -2):43-56.

[0119] (19) Shirahama K., Tsujii K., Takagi T., “Free-boundary electrophoresis of sodium dodecyl sulfate - protein polypeptide complexes with special reference to SDS-polyacrylamide gel electrophoresis” J Biochem. (1974) 75(2):309-319. doi:10.1093 / oxfordjournals.jbchem.al30398.

[0120] (20)

[0121] (21) Gutman A., Filep C., Karger B.L., “Fundamentals of Capillary Electrophoretic Migration and Separation of SDS Proteins in Borate Cross-Linked Dextran Gels” Anal Chem. (2021) 93(26):9267- 9276. doi: 10.1021 / acs.analchem. lc01636.

[0122] (22) Lance B. Koutny and Edward S. Yeung, “On-line detection of proteins in gel electrophoresis by ultraviolet absorption and by native fluorescence utilizing a charge -coupled device imaging system” Anal. Chem. (1993) 65(2): 183-187.

[0123] (23) Duhamel R. C., “Differential staining of collagens and non -collagens with Coomassie Brilliant Blue G and R” Coll Relat Res. (1983) 3(3): 195-204. doi:10.1016 / s0174-173x(83)80003-x.

[0124] (24) Nguyen D. B., Ly T. B., Wesseling M. C., Hittinger M.; Torge A.; Devitt A.; Perrie Y.; Bernhardt I.; “Characterization of Microvesicles Released from Human Red Blood Cells” Cell Physiol. Biochem. (2016) 38(3):1085-1099, doi:10.1159 / 000443059.

[0125] (25) Emonts P., Avohou H. T., Hubert P., Ziemons E., Fillet M., Dispas A., “Optimization of a robust and reliable FITC labeling process for CE-LIF analysis of pharmaceutical compounds using design of experiments strategy” J Pharm Biomed Anal. (2021) 205: 114304, doi: 10.1016 / j.jpba.2021.114304.

[0126] (26) Wang et al. “Applications of capillary electrophoresis in the fields of environmental, pharmaceutical, clinical and food analysis (2019-2021)” J Sep Sci (2022) 45(11): 1918-1941, doi:10.1002 / jssc.202100727.

[0127] (27) Lauren M. Ramsay, Jane A. Dickerson, Norman J. Dovichi “Atomole protein analysis by CIEF with LIF detection” Electrophoresis (2009) 30(2):297-302.

[0128] (28) Kurt E. Oldenburg and Jonathan V. Sweedler “Simple Sheath Flow Reactor for Post-column Fluorescence Derivatization in Capillary Electrophoresis” Analyst (1997) 122: 1581-1585.

[0129] (29) Colyer C.; “Noncovalent labeling of proteins in capillary electrophoresis with laser-induced fluorescence detection” Cell Biochem Biophys (2000) 33(3):323-337, doi:10.1385 / CBB:33:3:323.

Claims

CLAIMS1. A method for separation of proteins carried out by sodium dodecyl sulfate capillary gel electrophoresis (SDS-CGE) using fluorescence detection, wherein the separation matrix comprises fluorophore labeling agent having positive charge which binds with non-covalent binding to the proteins in the capillary.

2. The method according to claim 1 wherein the fluorophore labeling agent has double positive charge, preferably the fluorophore labeling agent being propidium iodide.

3. The method according to any of claims 1 or 2 wherein the fluorescence detection is laser-induced fluorescence (LIF) detection.

4. The use of a fluorophore protein labeling agent having positive charge which binds with non- covalent binding to the proteins in the capillary in sodium dodecyl sulfate capillary gel electrophoresis (SDS-CGE).

5. The use according to claim 4 wherein the fluorophore labeling agent has double positive charge, preferably the fluorophore labeling agent being propidium iodide.

6. A sodium dodecyl sulfate (SDS) capillary gel separation matrix (for use in SDS capillary gel electrophoresis (SDS-CGE)) comprising a fluorophore labeling agent having positive charge wherein said labeling agent binds with non-covalent binding to the proteins in the capillary.

7. The separation matrix according to claim 6 wherein the fluorophore labeling agent has double positive charge, preferably the fluorophore labeling agent being propidium iodide.

8. A kit for use in separation of proteins carried out by capillary gel electrophoresis (CGE), said kit comprising a separation matrix, preferably sodium dodecyl sulfate (SDS) capillary gel separation matrix; and a fluorophore labeling agent, wherein the separation matrix comprises the fluorophore labeling agent, and wherein the fluorophore labeling agent has positive charge which binds with non-covalent binding to the proteins in the capillary.

9. The kit according to claim 8 wherein the fluorophore labeling agent has double positive charge, preferably the fluorophore labeling agent being propidium iodide.

10. The use according to claim 4 or 5, the separation matrix according to claim 6 or 7 or the kit according to claim 8 or 9, wherein the fluorescence detection is laser-induced fluorescence (LIF) detection.

Images