Field of the invention

The present invention falls into the field of novel analytical techniques. In particular, the present invention provides a connection of liquid-phase microseparation methods to mass spectrometric detection of general nature and/or design, especially for capillary gel electrophoresis analysis of proteins and peptides including SDS-protein complexes.

Background of the invention

Capillary electrophoresis (CE) coupled with mass spectrometry (MS) is a unique combination of a high performance liquid phase separation technique and a special detection method, providing excellent selectivity and high sensitivity.

CE is a separation technique that utilizes the charge-to-hydrodynamic volume ratio of the analyte molecules for separation under the influence of an external electrical field. CE is a highly selective separation method, which requires small sample volumes and offers short analysis times with high efficiency and relatively low running cost. In the CE-MS coupled technique, the mass spectrometer provides a second separation dimension that further fractionates the incoming samples by their mass-to-charge ratio in the gas phase and finally acts as detector. From the viewpoint of fluid (sample) flow, the output of the CE separation is the input of the MS detection. In bioanalytical applications, where the sample usually contains sensitive molecules with large molecular masses, the most widely used soft ionization method of LC- and CE-MS coupling is electrospray ionization (ESI). However, online coupling of CE with ESI-MS is not a straightforward task due to their different electric current levels.

Capillary electrophoresis (CE) is one of the frequently used liquid phase separation methods for the analysis of peptides and proteins, mostly utilizing UV or fluorescent detection both in zone and gel electrophoresis modes [1-3], Hyphenation of CE with electrospray ionization mass spectrometry (ESI-MS), however, provides additional structural information about the separated sample components [4-6], In the past decades, various CE-MS interfaces have been developed including sheath flow [7], sheathless [8] and liquid junction [9] based approaches. Sheath flow CE nano ESI-MS interfaces, while supporting stable electrospray, somewhat dilute the analytes prior to entrance into the MS. Sheathless interfaces do not dilute the samples [10], but their fabrication requires delicate manipulation of miniaturized components [11] and may have intermittent spray stability issues. In the liquid junction interface, a liquid reservoir is surrounding the separation and transfer capillaries providing electric contact for the CE and MS circles [12], typically driven by an applied positive [13] or negative [14] pressure.

Over the past decade, there have been many attempts to improve CE-MS coupling with highly competitive features [15, 16], however, most of them still have challenging issues, such as accommodating the option to use gel filled capillaries and non-MS friendly buffer components, such as sodium dodecyl sulfate. SDS is a strong detergent that is widely used in proteomic studies for solubilization or extraction of hydrophobic or aggregated proteins and size-based separation of proteins by SDS-polyacrylamide gel electrophoresis. However, The SDS concentration has to be reduced prior to ESI-MS. The presence of SDS can cause ion suppression in ESI-MS and the appearance of strong SDS related signals. To achieve better spectra in ESI-MS analysis, SDS removal is required. The development of SDS removal methods prior to ESI-MS analysis has become an important and active research area in proteomics. These methods include solid phase extraction, filter aided sample preparation, electrofiltration, etc. The article [31] disclose a method where gamma cyclodextrin (y-CD) was used for the removal of SDS before ESI-MS. However, the prior art does not disclose any method for the removal of SDS in a direct onedimensional online coupled CE-MS system.

Other problems are the maintenance of the stable electrospray process [17], formation of bubbles during CE separation [18] and preserving the electrical contact [19], all seem to be recurring complications in CE-MS hyphenation. In addition, in most interfaces, specific modifications are required to the separation capillary such as drilling [20], sharpening and etching [21], or other labor-intensive and not easily reproducible alterations. Accurate positioning of the fused silica capillary into an outer sprayer tip is of also high importance [22] and a two- dimensional setting was recently introduced to accommodate the connection of different separation principles before MS detection [23],

All the above mentioned connection types can be used for CE-MS, but the electrospray interface with a coaxial liquid sheath flow arrangement is very often selected because of its robustness. In this setup, there is a makeup flow (sheath liquid) at the outlet end of the separation capillary providing electrical contact closure with the background electrolyte inside the separation capillary and simultaneously aiding the electrospray ionization process [24-26], Using a nanosource with liquid sheath flow systems usually does not utilize nebulizer gas, making the desolvation and ionization processes less efficient. To address this issue, users typically fine tune the following parameters to achieve a stable electrospray: 1) adjust the distance between the sprayer tip and the MS orifice, 2) optimize the electrospray voltage and 3) vary the flow rate of the sheath liquid. Finding the right distance between the sprayer tip and the orifice is quite challenging, because reducing the gap can cause corona discharges that might damage the MS instrument [27], This may also represent a problem when too high electrospray voltage is applied, in addition to possible in-source fragmentation issues [28], Regarding the sheath liquid stream, extremely low flow rates may result in faltering spray formation, while too high flow rates cause droplet formation. Therefore, operating commercial (i.e., factory developed and optimized) ESI sources for CE-MS with the option to use heated nebulizer gas would make optimization significantly easier, also assuring stable electrospray even at lower (-1500 V) spray voltages.

Thus, in light of the above, an online CE-MS coupling, preferably SDS-CGE-MS coupling would be of great interest. A general object of the present invention is to provide such a device.

Summary of the invention

One aspect of the present invention, thus, resides in a novel instrument based on the online coupling of CE with MS through coaxial sheath flow reactor interfacing.

To address the above-mentioned issues and further improve the versatility and robustness of CE-MS couplings, our group has developed a new approach by implementing a Coaxial Flow Reactor Interface (CFRI). In this setting, the sample transport from the separation capillary to the actual factory designed ESI source is based on a different principle. While the electric contact closure still takes place inside the coaxial sheath liquid mixing section of the interface part, the separated sample components are carried towards the ESI source by the sheath flow through a closed-circuit narrow bore flow reactor tube. This approach enables to use the original factory designed ESI sources for any MS system connected to the CE unit. In addition, this novel design opens the possibility for using gel filled capillaries, even with non-MS friendly buffer components such as SDS, as the addition of the flow reactor tube readily accommodates (online) post column reactions prior to MS detection, such as inclusion complexation, adsorption, chemical reaction, etc. Accordingly, a coupled CE-MS apparatus is disclosed herein, wherein the apparatus comprises a device for performing capillary electrophoresis of a sample, said device having a separation capillary of a given length and with an end portion distant to the device, forming an outlet of the device a mass spectrometer equipped with an electrospray ionization source forming an inlet of the mass spectrometer wherein the separation capillary and the electrospray ionization source define a contiguous flow path for the sample to be separated and analyzed by the mass spectrometer; the apparatus further comprising a sheath flow device arranged in said flow path and extending along at least a portion of the length of the separation capillary, said sheath flow device being configured to introduce a sheath fluid into the flow path at the end portion of said separation capillary; a flow reactor arranged in said flow path downstream of the sheath flow device and upstream of the electrospray ionization source, said flow reactor being configured to accommodate a reaction between at least one component of said sheath fluid and at least one component exiting the outlet end portion of the separation capillary.

In an embodiment of the invention the capillary electrophoresis is capillary gel electrophoresis, capillary zone electrophoresis, capillary isoelectric focusing, capillary isotachophoresis and capillary electrochromatography, preferably capillary gel electrophoresis. In case of capillary gel electrophoresis the components exit the outlet end portion of the separation capillary are in the form of ions. In the other cases, the components exit the capillary into the sheath fluid.

There is no particular restriction regarding the separation capillary, it depends on the molecules to be separated. In an embodiment of the invention, the separation capillary is commercially available.

In another embodiment of the invention the sheath fluid is a multicomponent fluid. In another embodiment of the invention the sheath flow device is a coaxial sheath flow device.

Coaxial sheath flow devices are commercially available.

In another embodiment of the invention the flow reactor is a narrow bore flow reactor tube, preferably a bare fused silica capillary. The capillary volume of the flow reactor is at least 100 nl, preferably at least 500 nl, more preferably at least 1000 nl. Said capillary volume of the flow reactor depends on the amount of said component to be reacted and the reaction speed, i.e. the capillary volume must be high enough to complete the reaction before entering the MS. However, increasing the volume leads to peak broadening, therefore the skilled person has to find an optimum volume. The flow reactor is 1-30 cm, preferably 2-20 cm, more preferably 5-10 cm long capillary with 30-250 pm, preferably 100-200 pm internal diameter. In an embodiment the flow reactor is an 8 cm long, 200 pm i.d. / 365 pm o.d. bare fused silica capillary.

The flow reactor is also referred to as flow reactor section or flow reactor zone.

The flow reactor is connected to the separation capillary through a non-conductive, preferably flexible tubing which is preferably PEEK tubing. The separation capillary and the flow reactor are arranged in the connection tubing to provide a closed-circuit connection. Said arrangement also defines a contact volume in the connection where the components exiting the outlet end portion of the separation capillary meet and first contact with the sheath fluid before they enter the flow reactor. The distance between the separation capillary and the flow reactor (i.e. the length of the contact volume) must be appropriate to contribute the decoupling of the electric circuit from the MS, usually about 0.05 mm and smaller than 0.1 mm to decrease the deadvolume induced band broadening. The flow reactor contacts only with the non-conductive tubing, not with the separation capillary.

The outlet of the flow reactor is connected to the electrospray ionization source of the mass spectrometer. The contiguous flow path therefore includes the separation capillary, the contact volume, the flow reactor and the electrospray ionization source, in this given order, which corresponds to the direction of flow in the system.

The part of the apparatus comprising the sheath flow device extending along at least a portion of the length of the separation capillary and said portion of the length of the separation capillary, the flow reactor and the connection between them (and also the contact volume) is referred to as Coaxial Flow Reactor Interface (CFRI) in the present disclosure. The arrangement above allows efficient decoupling of the electric circuit from the mass spectrometer, i.e., no current flow from the CE into the MS and vice-versa, enabling stable electrospray formation apparently independent of the capillary electrophoretic circuit of the system. The CFRI connection did not require any microfabrication and specially modified (i.e., etched, sharpened, etc.) capillaries, as a matter of fact, a conventional blunt edge, rugged fused silica capillary can be simply attached and detached to and from any commercial ESI sources through the flow reactor within a matter of minutes. This easy CE-MS coupling option allows the use of any commercial ESI sources originally developed and optimized for the actual mass spectrometer used with no modification requirements.

Another aspect of the present invention resides in a process for separating and analyzing components of a sample, preferably peptides and/or proteins in CE-MS, more preferably CGE-MS, the process comprising separating said components in a separation capillary of a capillary electrophoresis device, contacting the components exiting the separation capillary with a sheath fluid in a contact volume, then introducing said components together with the sheath fluid into the MS via a flow reactor, wherein the flow reactor is connected to the separation capillary through a non-conductive tubing, and the electric circuit of the CE is appropriately decoupled from the electric circuit of the MS.

The terms “components” and “chemical compounds” are used interchangeably in the present description.

In a preferred embodiment the process is performed in the apparatus of the invention.

In an embodiment of the invention the sheath fluid is a multicomponent fluid.

In an embodiment of the process, in addition to the target components, other components of the sample may also exit the end portion of the separation capillary. Some of these components might be harmful or not desired for MS analysis. In that embodiment, the process comprises contacting the component exiting the outlet end portion of the separation capillary with at least one other component of said sheath fluid thereby performing a sample preparation step for the MS analysis. In that case said contacting is followed by a reaction between the component eluting from the separation capillary and the component of the sheath fluid, such as inclusion com- plexation, adsorption, chemical reaction, etc. The component exiting the outlet end portion of the separation capillary meets and contacts with the sheath fluid in the contact volume. The components (from the sheath fluid and from the separation capillary) together pass through the flow reactor in a continuously flowing stream, thereby a reaction takes place. During said reaction or sample preparation step, the component which is harmful or not desired for MS analysis, is deactivated/eliminated/transformed, and thus it does not disturb the MS analysis.

In certain cases, a not desired component from a gel or a buffer used in the CE may exit the separation capillary.

In an embodiment of the invention the contacting of the components includes inclusion com- plexation. In a preferred embodiment of the invention, the capillary electrophoresis is capillary gel electrophoresis, the at least one component exiting the outlet end portion of the separation capillary is SDS and the at least one component of said sheath fluid is y-CD, and the reaction between these components forms an SDS— y-CD inclusion complex. In sodium dodecyl sulfate capillary agarose gel electrophoresis mode (where both the sample and the gel may contain SDS), addition of gamma cyclodextrin to the sheath liquid in the Coaxial Flow Reactor Interface efficiently removes the SDS content of the sample and the background electrolyte by inclusion complexation, while maintaining good separation efficiency and decreased ion suppression detection with high sensitivity. This was previously not possible with a direct online connection. The concentration of the y-CD in the sheath liquid is 0,01-20 %, preferably 0,1-5 %, more preferably 0,8-2 % (weight/volume %). The composition of the sheath fluid is otherwise not restricted, the skilled person can choose the appropriate buffer components.

In another embodiment of the invention the contacting of the component includes a chemical reaction. An alternative embodiment of the invention is permethylation of sugars to enhance ionization efficiency in the MS

Brief description of the drawings

Figure 1 : Schematic drawing of the apparatus according to the invention.

Figure 2: Capillary zone electrophoresis analysis of a peptide and a protein mixture with UV (214 nm, Panels A and C) and CZE-CFRI-ESI-MS (Total Ion Electropherogram, TIE, Panels B and D) detection. Peaks: Peptides - Panels A and B: (1) bradykinin; (2) angiotensin II and (3) neurotensin. Proteins - Panels C and D: (1) lysozyme, (2) ribonuclease A and (3) human insulin. Conditions: Capillary: 70 cm effective length, separation buffers: Inlet: 4.0% formic acid; Outlet/Sheath liquid: 4.0% formic acid with 20% methanol. Applied electric field strength: 285 V/cm (normal polarity); Injection: 5.0 psi / 20 sec. Detection: UV 214 nm (Panel A and C) and MS (Panel B and D). MS parameters: spray voltage: +2300 V; cone voltage: +30 V; desolvation gas flowrate: 600 L/h; desolvation temperature: 250°C; acquisition mode: MS (full scan) between 400 to 1000 (for peptides) and 700 to 2000 (for proteins) m/z.

Figure 3: Mass spectra of the peptides and proteins separated in Figure 2. The 2 ⁺ charge state for bradykinin and angiotensin as well as the 3 ⁺ for neurotensin were clearly identifiable in the peptide MS spectra (panel A). In case of protein analysis, 8 ⁺ and 9 ⁺ charge states were detected for lysozyme, 8 ⁺ to 12 ⁺ for RNase A, while 3 ⁺ 4 ⁺ and 5 ⁺ for human insulin (panel B).

Figure 4: Detection linearity as well as LOD and LOQ determination for RNase A using CZE- UV (panel A) and CZE-CFRI-ESI-MS (panel B) detection modes.

Figure 5: Native capillary agarose gel electrophoresis analysis of a peptide (panel A) and protein (panel B) mixture with MS detection using the CFRI-ESI setting. Peaks: Panel A: (1) bradykinin, (2) angiotensin II, (3) neurotensin; Panel B: (1) lysozyme (2) RNase A and (3) insulin. Conditions: Effective capillary length 70 cm, gel-buffer: Inlet: 4.0% formic acid, 0.6% agarose and 20% glycerol; Outlet/Sheath liquid: 4.0% formic acid with 20% methanol. Applied electric field strength: 285 V/cm (normal polarity). Injection: 5.0 psi / 20 sec. MS parameters: spray voltage: +2300 V; cone voltage: +30 V; desolvation gas flowrate: 600 L/h; desolvation temperature: 250°C; acquisition mode: MS ² (full scan) between 400 to 1000 m/z (for peptides) and 700 to 2000 m/z (for protein analysis).

Figure 6: SDS-capillary agarose gel electrophoresis of a 10 kDa size protein standard with UV (A) and MS (B) detection as well as the resulting MS spectra with (C) and without (D) y-cy- clodextrin in the sheath liquid. Conditions: Effective capillary length 70 cm; Gel-buffer: Inlet: 4.0% formic acid, 0.6% agarose and 20% glycerol; Outlet/Sheath liquid: 4.0% formic acid with 20% methanol and 0.8% y- cyclodextrin. Separation: 215 V/cm (reversed polarity, positive MS detection mode). Injection: 5.0 psi / 20 sec. MS parameters: spray voltage: +2300 V; cone voltage: +30 V; desolvation gas flowrate: 600 L/h; desolvation temperature: 250°C; acquisition mode: MS ² (full scan) between 400 to 1000 m/z (for peptides) and 700 to 2000 m/z (for protein analysis). Figure 7: Schematic drawing of the connection between the separation capillary and the flow reactor

Detailed description of preferred embodiments

In what follows, the invention is described in more detail with reference to Figure 1. Coupling capillary zone electrophoresis (CZE) and capillary gel electrophoresis (CGE) to a mass spectrometer via the Coaxial Flow Reactor Interface only required 1) a rugged bare fused silica separation capillary cut to size for analysis and filled with the corresponding buffer or gel for CZE and native- as well as SDS-CGE modes, 2) quickly insert the separation capillary into the coaxial interface part, and 3) connect to the commercial ESI source of the MS via the closed- circuit flow reactor tube in just a few minutes.

The bare fused silica (BFS) separation capillary 1 connected the inlet vial 2 of the electrically isolated capillary electrophoresis unit to the CFRI 3. This closed-circuit setup connected the outlet end of the coaxial sheath liquid tube of interface 3 and the ~2 mm protruding separation capillary 1 to the flow reactor 8 through a flexible PEEK tubing 11. Contact volume 12 is the space in the flexible PEEK tubing 11 between the separation capillary 1 and the flow reactor 8. The outlet reservoir 4 of the CE instrument (in our case a large-volume buffer reservoir) was filled with the sheath liquid and a capillary 5 was used for delivery a sufficient flow rate to maintain a stable spray at the other end of the setup, i.e. the electrospray capillary inside the commercial ESI source 10. Since the sheath liquid delivery tubing 5 was connected to the outlet reservoir 4 electrode of the CE instrument, it closed the CE electric circuit, and the ground was isolated by an isolation transformer 6 to avoid any excessive current flow to the MS power supply. The analyte molecules exiting the separation capillary 1 were carried downstream to the flow reactor 8 by the sheath liquid flow via the closed-circuit connection part 7 of the CFRI 3 and further into the ionization source 10 of the mass spectrometer. The nebulizer gas flow 9 of the ESI source 10 was carefully optimized not to cause any turbulences during the process.

Figure 1 shows the operation principle of CFRI, utilizing a continuous sheath liquid flow carrying the separated sample components from the outlet end of the CE separation capillary via the closed-circuit connection (also on Figure 7) of the flow reactor towards the commercial ESI sprayer. The aspiration rate of the MS nebulizer gas stream is optimized not to disrupt the laminar flow in the CFRI during the movement of the sample components to the ESI source via a short BFS capillary, enabling in this way a stable electrospray. Most importantly, the flow reactor allows special treatments e.g., inclusion complexation, adsorption, chemical reaction, etc., to take place after the fluid exits the separation capillary and before entering the MS unit.

We utilized this feature in our work to remove the surfactant during SDS-CGE-CFRI- ESI-MS analysis of a protein analyte by inclusion complexation of the SDS ingredient with y- cyclodextrin.

Examples

Materials and Methods

Chemicals. Formic acid, water (HPLC grade), methanol, glycerol, Tris, ribonuclease A, lysozyme, angiotensin II, bradykinin, neurotensin and agarose (ultra-low gelling temperature), were purchased from Sigma Aldrich (St. Louis, MO, USA). Human Insulin (Humulin) was a kind gift from the Medical School of University of Debrecen (Debrecen, Hungary). The 10 kDa size protein standard and the SDS-MW sample buffer were from Bio-Science Kft (Budapest, Hungary). y-cyclodextrin was kindly provided by the CycloLab Cyclodextrin Research and Development Laboratory Ltd. (Budapest, Hungary).

Sample and gel preparation. Peptides: angiotensin II, bradykinin and neurotensin were dissolved in HPLC grade water to a stock concentration of 10 mM each, then mixed and diluted with 4.0% formic acid to reach the final concentration of 100 pM. Proteins: 1.0 mg of RNase A and lysozyme were dissolved in HPLC grade water (stock solution of 10 mg/mL). For insulin, the content of the U-100 Humulin ampoule served as stock solution, with approximately 3.5 mg/mL concentration. The 10 kDa protein size standard was prepared by mixing 2 pL of standard solution with 80 pL of SDS-MW sample buffer and denatured by using the gradient temperature protocol developed earlier in our lab [29],

The low pH agarose-based native gel buffer system (pH 2.3) for CGE-UV and CGE-CFRI-ESI- MS consisted of 4% formic acid, 0.6% agarose and 20% glycerol. For the SDS-capillary agarose gel electrophoresis measurements, a Tris-acetate-EDTA (TAE) buffer was prepared containing 428 mM Tris base with 2 mM of EDTA Na2, followed by the addition of 10% (v/v) glycerol. The pH of the gel buffer system was adjusted to 7.0 with glacial acetic acid. Then, the ultra-low gelling temperature agarose was added to obtain 0.6 % (w/v) final concentration, stirred overnight at 75°C (250 RPM using a magnetic hotplate), followed by the addition of 0.1% (w/v) SDS and slowly mixed at room temperature (~100 RPM) for an additional hour to prevent foaming.

CZE-, native CGE- and SDS-CGE with UV detection. All separations were performed by using a CESI 8000 System (Beckman Coulter, Fullerton, CA) with a bare fused silica capillary (70 cm effective length, 80 cm total length, 50 pm ID/365 pm OD) and 214 nm UV detection.

- CZE separation buffer: Inlet: 4.0% formic acid, Outlet: 4.0% formic acid and 20% methanol. Applied electric field strength: 285 V/cm (normal polarity, anode at the injection side); Sample injection: 5.0 psi / 20 sec.

- Native CGE gel buffer system: Inlet and outlet reservoirs: 4% formic acid, 0.6% agarose and 20% (v/v) glycerol (pH 2.3). Applied electric field strength: 285 V/cm (normal polarity); Sample injection: 5.0 psi / 20 sec.

- SDS-CGE gel buffer system: Inlet and outlet reservoirs: 428 mM TAE, 10% (v/v) glycerol, 0.1% SDS and 0.6% agarose (pH 7.0). Applied electric field strength: 215 V/cm (reversed polarity, cathode at the injection side). Please note that in this instance, lower field strength was applied due to the higher conductivity of the acetate-based gel buffer system. Sample injection: 5.0 psi / 20 sec.

CZE-, native CGE- and SDS-CGE via CFRI coupling with MS detection. CZE-, native CGE- and SDS-CGE-CFRI-ESI-MS experiments were performed on the above-described capillary electrophoresis system connected to a Waters Xevo TQ-S triple quadrupole mass spectrometer (Milford, MA) via the Coaxial Liquid Sheath Flow Interface. The 70 cm long separation capillary was inserted into a modified OptiMS Capillary Cartridge (Beckman Coulter) with shortened tube length and cutoff protective housing, connected through the closed-circuit flow reactor extension to the commercial MS ESI sprayer of the unit (ESI/APCI/ESCI® probe of a Waters Z- spray dual orthogonal API source) as shown in Figure 1.

- CZE/native CGE-CFRI-ESI-MS: Separation buffer: Inlet: 4.0 % formic acid for CZE- MS and 4.0 % formic acid, 0.6% agarose and 20% glycerol CGE-MS. Outlet: 4.0% formic acid and 20% methanol. Applied electric field strength: 285 V/cm (normal polarity); Sample injection: 5.0 psi / 20 sec. - SDS-CGE-CFRI-MS: Separation buffers: Inlet: 428 TAE with 10% (v/v) glycerol, 0.1% SDS and 0.6% agarose (pH 7.0). Outlet: 4% formic acid, 20% methanol and 0.8% y-cyclodextrin. Applied electric field strength: 215 V/cm (reversed polarity with positive MS detection mode). Sample injection: 5.0 psi / 20 sec.

- MS parameters: spray voltage: +2300 V; cone voltage: +30 V; desolvation gas flowrate: 600 L/h; desolvation temperature: 250°C; acquisition mode: MS2 (full scan) between 500 to 2000 m/z.

The CFRI device used in the examples is as follows: the BFS separation capillary (70 cm total length, 50 pm ID/365 pm OD) connected the inlet vial of the electrically isolated capillary electrophoresis unit to the CFRI. This closed-circuit setup connected the outlet end of the coaxial sheath liquid tube of interface and the ~2 mm protruding separation capillary to an 8 cm long 200 pm i.d./365 pm o.d. BFS capillary through a flexible peek tubing. The outlet reservoir of the CE instrument (in our case a large-volume buffer reservoir) was filled with the sheath liquid and a 70 cm long, 200 pm i.d./365 pm o.d. capillary was used for delivery at a minimum flow rate of 75 pL/min to maintain a stable spray at the other end of the setup, i.e., at the 150 pm i.d. stainless-steel electrospray capillary inside the commercial ESI source. Since the sheath liquid delivery tubing was connected to the outlet reservoir electrode of the CE instrument, it closed the CE electric circuit, and the ground was isolated by an isolation transformer to avoid any excessive current flow to the MS power supply.

A simple and generally applicable approach is disclosed for connecting capillary electrophoresis to mass spectrometric detection by introducing a Coaxial Flow Reactor Interface (CFRI). UV and MS detections were used under identical separation conditions in a comparative manner for peptide and protein analysis, both in CZE and CGE modes, to show the excellent potential of this approach. In SDS-CGE mode, inclusion complexation was utilized with gamma cyclodextrin in the sheath liquid to remove the SDS content from the sample and the background electrolyte in the flow reactor section before entering the MS unit, i.e., minimizing ion suppression. The dynamic range as well as limits of detection and quantification were also determined.

Capillary zone electrophoresis of peptides and proteins with UV and CFRI-ESI-MS detection Capillary electrophoresis coupled to electrospray ionization mass spectrometry offers a sensitive detection option with the additional structural information about the exact mass over charge values of the separated analyte molecules. To evaluate any possible separation efficiency loss due to the use of the coaxial flow reactor section of the interface, the CE-CFRI-ESI-MS approach was compared with conventional CE-UV detection using identical separation conditions, i.e., column length, running buffer composition, separation temperature as well as injection mode and applied electric field strengths. First, a mixture of three peptides was analyzed by CZE-UV and CZE- CFRI-ESI-MS (Figure 2, panels A and B, respectively) using a simple 4% formic acid BGE.

As Figure 2 exhibits, no significant differences were observed in the resulting electropherograms between the use of the two detection systems. All three peptides were well separated with both methods by using the same simple background electrolyte of 4.0% formic acid in the capillary and 4.0% formic acid with 20% methanol in the sheath liquid (Panels A and B). According to the numeric evaluation of the separation performance (resolution, peak efficiency, etc., detailed in Table SI), there were only minor differences between the two detection methods. The slightly longer migration time with the CFRI setup was due to the extra distance of the flow reactor between the capillary end and the MS orifice.

CZE analysis of the protein mixture of lysozyme, RNase A and insulin revealed similar results to the peptide separations. The test proteins were baseline resolved with good mass spectromet- ric identification. Here again, all separation performance values were very similar between the UV and CFRI-ESI-MS detection modes (Table S2).

The mass spectra extracted from the Total Ion Electropherograms (TLEs) in Figure 2 provided extra information in addition to the migration times of the separated analytes, as shown in Figure 3. Based on the collected MS spectra, the molecular masses and the migration order of the sample components were easily determined as 1060 Da (bradykinin, peak 1), 1046 Da (angiotensin II, peak 2) and 1673 Da (neurotensin, peak 3) for the peptides and -14300 Da (lysozyme, peak 1), -13700 Da (RNase A, peak 2) and -5800 Da (insulin, peak 3) for the proteins. As one can observe, the 2 ⁺ charge state of bradykinin and angiotensin as well as the 3 ⁺ for neurotensin were clearly detectable in the peptide MS spectra (Figure 3 / panel A). Regarding the protein analysis part, 8 ⁺ and 9 ⁺ charge states were identified for lysozyme, 8 ⁺ to 12 ⁺ for RNase A, while 3 ⁺ 4 ⁺ and 5 ⁺ for human insulin (Figure 3 / panel B).

Quantification with CFRI-MS

Ribonuclease A was used to compare the limit of detection (LOD) and limit of quantification (LOQ) with the UV and CFRI-ESI-MS settings, as well the detection linearity in the concentration range of 3.0 pg/mL to 10 mg/mL for CZE-UV and 0.6 pg/mL to 2.0 mg/mL for CZE-CFRI- ESI- MS. Five-fold dilutions were carried out starting from the 10 mg/mL concentration of the RNase A stock solution. Figure 4 compares the detection linearity plots of the two detection methods after injecting the serially diluted samples. The limit of detection (LOD) and limit of quantification (LOQ) were 15 and 50 pg/mL in case of CZE-UV setup, respectively, with the calibration plot covering ~3 orders of magnitude with good linearity (r2=0.993). The linear dynamic range was apparently similar for the CZE-CFRI-ESI-MS setting in a somewhat narrower concentration range but with slightly better linearity (r2=0.997). Here the LOD and LOQ values were 3 and 10 pg/mL (5 times better than with UV detection), respectively, corresponding well with the literature data of mass spectrometric detection of small intact proteins using a triple quadrupole instrument [30],

It is important to note that detection and quantification limits can be further optimized depending on the application and equipment used, since CFRI-ESI parameters can be easily adjusted (nitrogen flow rate, ESI voltage, flow reactor capillary diameters and lengths, etc.). In addition, the type and versatility of the MS instruments and acquisition methods applied are important too as all systems can use their own commercially available optimized electrospray setting.

Native capillary agarose gel electrophoresis with MS detection using the coaxial flow reactor interface

The CFRI setting was next evaluated to accommodate native capillary gel electrophoresis separation of protein analytes. Figure 5 shows the TIEs of the separated peptides (panel A) and proteins (panel B) in 0.6% agarose gel filled capillaries.

As one can observe in Figure 5, the use of gel filled capillaries resulted in very steady baseline proving the high stability of the electrospray process with the CFRI setup. Most importantly, no gel components were detected in the MS spectra. Due to the sieving matrix in the separation capillary, sample components showed increased migration times (1.7x and 1.4x longer for peptides and proteins, respectively) and significantly enhanced resolution (from an average of 1.52 and 2.22 to 1.55 and 4.81 for peptides, and 1.45 and 1.67 to 2.48 and 3.24 for proteins as delineated in Tables S1-S4). Three consecutive separation runs were performed using the peptide and protein mixtures to compare the two detection methods, which demonstrated excellent migration time, peak area, resolution and theoretical plate number repeatability (see in Tables S1-S4, all less than 2.35% RSD) proving the robustness of the CGE-CFRI-ESI-MS approach. The marginal (up to -12%) loss in resolution with the CFRI-ESI-MS detection system compared to UV detection was probably due to the laminar hydrodynamic flow mediated band broadening in the 8 cm long flow reactor part. The longer migration times (Tables S1-S4) obtained with the CFRI setting was also due to the additional flow reactor connection part. Please note that the effective capillary lengths (70 cm) and the applied electric field strengths (285 V/cm) were identical in both separation / detection methods.

Sodium dodecyl sulfate capillary agarose gel electrophoresis with CFRI-ESI-MS detection

To take further advantage of this novel interface design, we evaluated the sodium dodecyl sulfate (SDS) - cyclodextrin (CD) inclusion complexation reaction in the flow reactor section during SDS- capillary agarose gel electrophoresis coupled to MS detection using a well characterized commercially available protein size standard (10 kDa). Following previous attempts to reduce SDS-mediated ionization suppression 31, we added y-CD to the sheath liquid to utilize its hydrophobic cavity for inclusion complexation based capture of the SDS molecules by their hydrophobic tail. Since the resulting cyclodextrin-SDS inclusion complexes were not observed in the MS spectra, this approach readily enabled the long sought SDS-CGE of proteins with MS detection.

As Panels A and B in Figure 6 show, the separation performance of SDS capillary agarose gel electrophoresis was very similar between the use of UV and MS detection modes, with peak efficiencies of N=5140 and N=6737, respectively. More importantly, the capture of SDS molecules in the flow reactor section by the y-CD in the sheath liquid during SDS-CGE-CFRI- ESIMS significantly reduced SDS-induced ion suppression and the injected protein remained clearly detectable in positive ionization mode (Panel B). Intensities of the different protein charge states (7 ⁺ to 16 ⁺ ) were slightly lower (<15 %) as a result of the SDS removal process. Due to the addition of the cyclic oligosaccharide to the sheath liquid, y-CD ⁺ and y-CD + Na ⁺ adducts appeared in the spectrum at 1298 1320 m/z (Panel C). However, the characteristic charge states of the 10 kDa protein standard were barely identifiable without the addition of this macrocyclic additive. Please note that the characteristic charge states of the 10 kDa protein standard were barely identifiable without the addition of y-CD to the sheath liquid, and the ion suppression caused by the SDS containing BGE was significant (Figure 6D). Most importantly, the SDS - y-CD inclusion complexation in the closed-circuit flow reactor part of the interface adequately removed the SDS content from the system, allowing efficient electrospray and resulting in strong MS signal of the protein sample in SDS-CGE analysis.

Table SI: Statistical evaluation and separation performance data of the UV and CFRI-ESI-MS detection signals of the test peptide mixture separated by capillary zone (CZE) electrophoresis

Table S2: Statistical evaluation and separation performance data of the UV and CFRI-ESI-MS detection signals of the test protein mixture separated by capillary zone (CZE) electrophoresis

Table S3: Statistical evaluation and separation performance data of the UV and CFRI-ESI-MS detection signals of the test peptide mixture separated by capillary gel (CGE) electrophoresis

Table S4: Statistical evaluation and separation performance data of the UV and CFRI-ESI-MS detection signals of the test protein mixture separated by capillary gel (CGE) electrophoresis

References

(1) Dawod, M.; Arvin, N. E.; Kennedy, R. T. Recent advances in protein analysis by capillary and microchip electrophoresis. Analyst 2017, 142 (11), 1847-1866. DOI: 10. 1039/c7an00198c.

(2) Kasicka, V. Recent developments in capillary and microchip electroseparations of peptides (2015- mid 2017). Electrophoresis 2018, 39 (1), 209-234. DOI: 10.1002/elps.201700295.

(3) Sarkozy, D.; Guttman, A. Capillary Sodium Dodecyl Sulfate Agarose Gel Electrophoresis of Proteins.

Gels 2022, 8 (2). DOI: 10.3390/gels8020067.

(4) Maxwell, E. J.; Chen, D. D. Twenty years of interface development for capillary electrophoresiselectrospray ionization-mass spectrometry. Analytica chimica acta 2008, 627 (1), 25-33. DOI:

10. 1016/j.aca.2008.06.034.

(5) Stolz, A.; Jooss, K.; Hocker, O.; Romer, J.; Schlecht, J.; Neususs, C. Recent advances in capillary electrophoresis-mass spectrometry: Instrumentation, methodology and applications. Electrophoresis 2019, 40 (1), 79-112. DOI: 10.1002/elps.201800331.

(6) Romer, J.; Montealegre, C.; Schlecht, J.; Kiessig, S.; Moritz, B.; Neususs, C. Online mass spectrometry of CE (SDS)-separated proteins by two-dimensional capillary electrophoresis. Analytical and bioanalytical chemistry 2019, 411 (27), 7197-7206. DOI: 10.1007/s00216-019-02102-8.

(7) Smith, R. D.; Udseth, H. R. Capillary zone electrophoresis-MS. Nature 1988, 331 (6157), 639-640. DOI: 10. 1038/331639a0.

(8) Moini, M. Capillary electrophoresis-electrospray ionization mass spectrometry of amino acids, peptides, and proteins. Methods Mol Biol 2004, 276, 253-290. DOI: 10.1385/l-59259-798-X:253.

(9) Lee, E. D.; Muck, W.; Henion, J. D.; Covey, T. R. Liquid Junction Coupling for Capillary Zone Electrophoresis Ion Spray Mass-Spectrometry. Biomedical and Environmental Mass Spectrometry 1989, 18 (9), 844-850. DOI: DOI 10. 1002/bms. 1200180932.

(10) Yin, Y.; Li, G.; Guan, Y.; Huang, G. Sheathless interface to match flow rate of capillary electrophoresis with electrospray mass spectrometry using regular-sized capillary. Rapid CommunMass Spectrom 2016, 30 Suppl 1, 68-72. DOI: 10. 1002/rcm.762L

(11) Issaq, H. J.; Janini, G. M.; Chan, K. C.; Veenstra, T. D. Sheathless electrospray ionization interfaces for capillary electrophoresis-mass spectrometric detection - Advantages and limitations. Journal of Chromatography A 2004, 1053 (1-2), 37-42. DOI: 10.1016/j.chroma.2004.08.116.

(12) Reinhoud, N. J.; Niessen, W. M. A.; Tjaden, U. R.; Gramberg, L. G.; Verheij, E. R.; van der Greef, J. Performance of a liquid-junction interface for capillary electrophoresis mass spectrometry using continuous-flow fast-atom bombardment. Rapid Communications in Mass Spectrometry 1989, 3 (10), 348- 351. DOI: https://doi.org/10.1002/rcm.1290031009. (13) Wachs, T.; Sheppard, R. L.; Henion, J. Design and applications of a self-aligning liquid junctionelectrospray interface for capillary electrophoresis-mass spectrometry. J Chromatogr B Biomed Appl 1996, 685 (2), 335-342. DOI: 10.1016/s0378-4347(96)00176-4.

(14) Foret, F.; Zhou, H.; Gangl, E.; Karger, B. L. Subatmospheric electrospray interface for coupling of microcolumn separations with mass spectrometry. Electrophoresis 2000, 21 (7), 1363-1371. DOI: 10.1002/(SICI)1522-2683(20000401)21:7<1363::AID-ELPS1363& gt;3.0.CO;2-U.

(15) Ramautar, R.; Somsen, G. W.; de Jong, G. J. CE-MS for metabolomics: Developments and applications in the period 2016-2018. Electrophoresis 2019, 40 (1), 165-179. DOI:

10. 1002/elps.201800323.

(16) Zhang, W.; Ramautar, R. CE-MS for metabolomics: Developments and applications in the period 2018- 2020. Electrophoresis 2021, 42 (4), 381-401. DOI: 10.1002/elps.202000203.

(17) Smith, R. D.; Barinaga, C. J.; Udseth, H. R. Improved Electrospray Ionization Interface for Capillary Zone Electrophoresis - Mass-Spectrometry. Analytical Chemistry 1988, 60 (18), 1948-1952. DOI: Doi 10. 1021/Ac00169a022.

(18) Smith, A. D.; Moini, M. Control of electrochemical reactions at the capillary electrophoresis out- let/electrospray emitter electrode under CE/ESI-MS through the application of redox buffers. Anal Chem 2001, 73 (2), 240-246. DOI: 10.1021/ac0007940.

(19) Chao, B. F.; Chen, C. J.; Li, F. A.; Her, G. R. Sheathless capillary electrophoresis-mass spectrometry using a pulsed electrospray ionization source. Electrophoresis 2006, 27 (11), 2083-2090. DOI: 10. 1002/elps.200500811.

(20) Fanali, S.; D'Orazio, G.; Foret, F.; Klepamik, K.; Aturki, Z. On-line CE-MS using pressurized liquid junction nanoflow electrospray interface and surface-coated capillaries. Electrophoresis 2006, 27 (23), 4666-4673. DOI: 10.1002/elps.200600322.

(21) Moini, M. Simplifying CE-MS operation. 2. Interfacing low-flow separation techniques to mass spectrometry using a porous tip. Anal Chem 2007, 79 (11), 4241-4246. DOI: 10. 1021/ac0704560.

(22) Liu, C. C.; Alary, J. F.; Vollmerhaus, P.; Kadkhodayan, M. Design, optimisation, and evaluation of a sheath flow interface for automated capillary electrophoresis-electrospray-mass spectrometry. Electrophoresis 2005, 26 (7-8), 1366-1375. DOI: 10.1002/elps.200410133.

(23) Kohl, F. J.; Montealegre, C.; Neususs, C. On-line two-dimensional capillary electrophoresis with mass spectrometric detection using a fully electric isolated mechanical valve. Electrophoresis 2016, 37 (7-8), 954-958. DOI: 10.1002/elps.201500579.

(24) Muck, W. M.; Henion, J. D. Determination of Leucine Enkephalin and Methionine Enkephalin in Equine Cerebrospinal-Fluid by Microbore High-Performance Liquid-Chromatography and Capillary Zone Electrophoresis Coupled to Tandem Mass-Spectrometry. Journal of Chromatography-Biomedical Applications 1989, 495, 41-59. DOI: Doi 10.1016/S0378-4347(00)82608-0.

(25) Bezy, V.; Chaimbault, P.; Morin, P.; Unger, S. E.; Bernard, M. C.; Agrofoglio, L. A. Analysis and validation of the phosphorylated metabolites of two anti-human immunodeficiency virus nucleotides (stavudine and didanosine) by pressure-assisted CE-ESI-MS/MS in cell extracts: sensitivity enhancement by the use of perfluorinated acids and alcohols as coaxial sheath-liquid make-up constituents.

Electrophoresis 2006, 27 (12), 2464-2476. DOI: 10.1002/elps.200500850.

(26) Foret, F.; Thompson, T. J.; Vouros, P.; Karger, B. L.; Gebauer, P.; Bocek, P. Liquid Sheath Effects on the Separation of Proteins in Capillary Electrophoresis Electrospray Mass-Spectrometry.

Analytical Chemistry 1994, 66 (24), 4450-4458. DOI: DOI 10.1021/ac00096a010.

(27) Gonzalez-Ruiz, V.; Codesido, S.; Far, J.; Rudaz, S.; Schappler, J. Evaluation of a new low sheathflow interface for CE-MS. Electrophoresis 2016, 37 (7-8), 936-946. DOI: 10.1002/elps.201500523.

(28) Friese, O. V.; Smith, J. N.; Brown, P. W .; Rouse, J. C. Practical approaches for overcoming challenges in heightened characterization of antibody-drug conjugates with new methodologies and ult- rahigh- resolution mass spectrometry. mAhs 2018, 10 (3), 335-345. DOI:

10. 1080/19420862.2018. 1433973.

(29) Szigeti, M.; Guttman, A. Sample Preparation Scale-Up for Deep N-glycomic Analysis of Human Serum by Capillary Electrophoresis and CE-ESLMS. Molecular & cellular proteomics : MCP 2019, 18 (12), 2524- 2531. DOI: 10. 1074/mcp.TIRl 19.001669.

(30) Wang, E. H.; Combe, P. C.; Schug, K. A. Multiple Reaction Monitoring for Direct Quantitation of Intact Proteins Using a Triple Quadrupole Mass Spectrometer. J Am Soc Mass Spectrom 2016, 27 (5), 886-896. DOI: 10. 1007/s 13361-016- 1368-2.

(31) Quirino, J. P. Sodium dodecyl sulfate removal during electrospray ionization using cyclodextrins as simple sample solution additive for improved mass spectrometric detection of peptides. Analytica chimica acta 2018, 1005, 54-60. DOI: 10. 1016/j.aca.2017. 12.012.