Coaxial Electrospray Ionization for the Study of Rapid In-source Chemistry
Coaxial electrospray has been used effectively for several dual-emitter applications, but has not been utilized for the study of rapid in-source chemistry. In this paper, we report the fabrication of a coaxial, micro-volume dual-emitter through the modification of a manufacturer’s standard electrospray probe. This modification creates rapid mixing inside the Taylor cone and the ability to manipulate fast reactions using a variety of solvents and analytes. We demonstrate its potential as a low-cost, dual-emitter assembly for diverse applications through three examples: relative ionization in a biphasic electrospray, hydrogen-deuterium exchange, and protein supercharging.
KeywordsCoaxial electrospray Hydrogen-deuterium exchange Protein folding Biphasic electrospray ionization Supercharging
Coaxial electrospray ionization (ESI) and its variations have been used in diverse dual-emitter applications. For example, coaxial ESI with a polymer sheath flow produces coated microdroplets and microemulsions for pharmaceutical, cosmetic, and food industrial purposes . Nanotechnology, biotechnology, optoelectronics, and polymer industries use coaxial ESI together with fiber spinning methods to produce nanofibers and nanotubes in a process known as electrospinning . In mass spectrometry, the use of coaxial ESI needles is common in capillary electrophoresis systems, in which a sheath buffer stabilizes Taylor cone formation and reduces ion suppression in the subsequent analysis [3, 4, 5]. Additionally, some ESI sources use a coaxial emitter for concomitant monitoring of calibrants or standards [6, 7, 8].
Recently, dual-emitter ESI sources such as capillary microfluidics [4, 9, 10], extractive electrospray ionization (ESS) , fused droplet ionization (FD-ESI) , solvent-assisted electrospray ionization (SAESI) , secondary electrospray ionization (SESI) , dual-sprayer microchips , and theta glass capillary nanospray  have allowed researchers to reimagine the scope of ESI. Dual-emitter sources are characterized by having two independent solvent flows or sprays that could contain a wide array of compounds besides an analyte, including reactants, standards, or ionizing agents, which can be used to manipulate analytes on-line for subsequent mass analysis. When preceded by separations methods, dual-emitters can adjust for pH, sensitivity, or ESI compatibility without interfering with separations chemistry. To date, the use of dual-spray ESI has contributed to the understanding of subjects such as protein structure and conformation [17, 18, 19], electrochemistry [20, 21], isotopic exchange , and peptide derivatization .
Many dual-spray ESI sources flow solvent in the microliter-to-milliliter range, incorporating a heated drying gas to provide desolvation during droplet formation. At lower flow rates, in the range of 1–1000 nL min−1, nanoelectrospray (nanospray) sources produce smaller initial droplet sizes that, because of their high charge state, promptly produce offspring droplets from which ions are released. When dual-emitters are used in nanospray, such as dual-sprayer microchips and theta glass capillary nanospray, the finer aerosolized spray affords greater surface interaction between droplets  to create and study rapid, in-source chemical reactions. However, in comparison to the aforementioned modified ESI methods, the implementation of nanospray can be cost-prohibitive when one considers the purchase of a stand-alone atmospheric source with microscope positioning optics, a separate heated MS orifice, and pumping systems in the nanoliter flow range.
In considering the multitude of modified ESI probes, subtle differences often produce profound outcomes in ionization due to the ability to manipulate ionization in the liquid phase or overlapping charged aerosol phases . In this work, we bring capabilities of both conventional and nano-ESI with the introduction of a dual-emitter with a single, coaxial, micro-volume spray adapted to a manufacturer’s standard electrospray emitter. In coaxial ESI probe configurations, the inner and outer capillaries are positioned so that they form a single Taylor cone upon the application of high voltage. This is distinct from the dual-spray charged aerosol overlap of EESI or the dual-Taylor cone overlap of theta glass nanospray or SAESI. In practice, the use of dual-spray, dual-emitters results in reaction chemistry in the spray overlap region where reactions occur at the surface of microdroplets . It is possible that with this coaxial, dual-spray, single-emitter design, in-source reactions could be created in the solvent system of the Taylor cone, in the liquid filament, or at the surface or in the bulk of the resultant microdroplets.
This study reports the straightforward, accessible construction of a coaxial ESI emitter through modification of a standard ESI probe, similar to earlier successes in coaxial ESI designated for capillary electrophoresis . Results are presented to demonstrate the validity of the method and its ability to study ionization from new and diverse areas of mass spectrometry: biphasic solvent systems, kinetics of rapid, on-line hydrogen-deuterium exchange (HDX), and on-line protein supercharging.
HPLC grade methanol (99.93%, Sigma-Aldrich, St. Louis, MO), HPLC grade chloroform (99%, J. T. Baker, Phillipsburg, NJ), D2O (99.9%, Sigma-Aldrich, St. Louis, MO), and distilled-deionized water (> 18 MΩ cm) from a Millipore Direct-Q water purification system (Billerica, MA) were used as solvents. Cytochrome C (equine) (> 95%, Sigma-Aldrich, St. Louis, MO), phenylalanine (≥ 98%, Sigma-Aldrich, St. Louis, MO), testosterone (≥ 98%, Sigma-Aldrich, St. Louis, MO), and testosterone-[13C3] (> 99%, IsoSciences, King of Prussia, PA) were used without further purification.
Coaxial ESI Fabrication
Mixing of Miscible and Immiscible Solvent Coaxial Sprays
Experiments were conducted using unlabeled and isotopically labeled compounds delivered independently in varying ratios of aqueous and non-aqueous solvent through the inner and outer capillaries. Two compounds were investigated; testosterone/testosterone-[13C3] (289/292 m/z), imidacloprid/imidacloprid-[13CD3] (256/260 m/z). In all experiments, the total combined flow rate of the inner and outer solutions was 20 μL min−1. In the miscible-spray experiment, a solution of the unlabeled compound in water was delivered through the inner capillary, and a solution of the isotopically-labeled compound in water was delivered through the outer capillary. In the two immiscible solvent experiments, (1) a solution of the unlabeled compound in chloroform was delivered through the inner capillary while a solution of the isotopically labeled compound in water was delivered through the outer capillary and (2) a solution of the isotopically labeled compound in water was delivered through the inner capillary while a solution of the unlabeled compound in chloroform was delivered through the outer capillary. Nebulizing gas flow was set at 40 psi (GS1 = 40).
HDX in Phenylalanine by Coaxial ESI
Using coaxial ESI, 6.0 mM phenylalanine (Phe) flowing at 10 μL min−1 was delivered through the inner capillary, and D2O was delivered at 10 μL min−1 through the outer capillary. The extent of HDX was observed as nebulizing gas was adjusted between 5 and 70 psi.
Protein Supercharging Experiments
Fifty-two micrometers of horse heart cytochrome C (Sigma, > 95%) in 3% acetic acid in water was delivered through the inner capillary at 10 μL min−1. Supercharging agents were infused at 3 μL min−1 from the outer capillary, 0.7% m-nitrobenzyl alcohol (m-NBA) in acetic acid/water/methanol (3%/47%/50%), and 50% glycerol in acetic acid/water (3%/97%). Nebulizing gas flow was set at 15 psi (GS1 = 15).
Results and Discussion
Taylor Cone Formation
Spray shape and stability in coaxial ESI are highly dependent on the incorporation of flow from the outer capillary into the Taylor cone. As previously observed in microencapsulation studies, the coaxial Taylor cone is reliant on electrospray parameters such as sheath gas flow, voltage, and flow rate, here investigated between 13 and 20 μL min−1, as well as mechanical considerations, such as protrusion distance of the inner capillary from the outer sheath capillary and the face cleave on the protruding edge of the fused silica capillary . Stable aqueous cone-jet images are captured in Fig. 2.
The experiments performed using coaxial ESI were selected to demonstrate the versatility of its design. In the solvent miscibility experiments, spray composition is easily altered by simply changing which syringe infuses into the outer or inner capillary. In the HDX experiments, residence time in the source, and subsequently, reaction time, is readily manipulated with adjustments of nebulizer gas flows. The relative percentages of solvents flowing through the inner/outer needle is controlled by adjusting flow rates, as in the supercharging experiments.
Mixing of Miscible and Immiscible Solvent Coaxial Sprays
When simultaneously spraying aqueous and chloroform solvents through independent capillaries, the compound in the non-aqueous chloroform was preferentially ionized over the compound in the aqueous solvent, regardless of whether delivered through the inner or outer capillaries (Fig. 3b, c). Unlabeled and labeled compounds approached equal ionization intensities when mixing approximately 5 μL/min of chloroform and 15 μL/min of water, demonstrating the feasibility of coaxial ESI experiments using non-aqueous solvents not typically used in ESI. It is worth noting that analyte detection optimized at a higher nebulizing gas flow could have assisted in this mixing phenomenon.
To rationalize this preferential ionization in chloroform, we hypothesize that proton transfer to the analyte could occur through a gas-phase protonated water molecule. In this case, a non-aqueous spray desolvates faster due to a higher vapor pressure, and so it may be that the gas phase analyte in chloroform is more available for ionization by protonated water clusters.
HDX in Phenylalanine by Coaxial ESI
Rapid on-line HDX mass spectrometry has been realized for the investigation and structural elucidation of small molecules, especially pharmaceuticals and large biomolecules . The methods that have been developed are diverse and range from exchange with deuterated gases or vapors [27, 28] to deuterated capillary electrophoresis sheath flows , post-column solvent tees , dual-sprayers , and other on-line systems [31, 32]. These techniques introduce deuterated molecules in the seconds or milliseconds before entering the mass spectrometer, providing structural insights with more accuracy and less effort than HDX in solution.
In a recent publication, Jansson et al. used theta capillaries to demonstrate that HDX of rapid-exchange sites on phenethylamine can be manipulated by increasing the distance between the spray tip and the MS orifice, increasing reaction time . Using coaxial ESI, HDX reaction time may be manipulated by adjusting the nebulizing gas pressure (Fig. 4). As nebulizing gas pressure decreases, the intensities of mono-, di-, and tri-deuterated Phe species increase, and the intensity of the [Phe + H]+ (at 166 m/z) decreases. In theory, this data suggests that higher nebulizing gas pressures either decrease in-source residence time for HDX or negatively affect inner and outer capillary mixing. From this, we anticipate that coaxial ESI with deuterated solvents may lend itself to pulsed labeling experiments , with a short period of deuterated solvent exposure before “quenching” upon entering the high vacuum region of the mass analyzer.
Protein Supercharging Experiments
Protein and peptide supercharging was first recognized by the Williams group in 2001 . Since then, research into the mechanism and potential uses for supercharging agents has increased dramatically, especially as applied to proteomics. Inclusion of supercharging agents create higher charge states beneficial for mass determination of large m/z proteins, improves mass resolution through the use of lower mass ranges, and increased fragmentation and sequence coverage by ETD and ECD [36, 37] without dissociating noncovalent complexes . Supercharging agents have also been shown to decrease ion suppression induced by ion-pairing agents .
Attempts to streamline the addition of supercharging agents for high-throughput workflows resulted in several experiments that included supercharging reagents into chromatographic mobile phases, but with mixed results . To circumvent these issues, Miladinović et al. introduced supercharging agents in-source using a dual-chip sprayer .
Though glycerol creates higher average charge states than m-NBA, it is less commonly used. This may be because of the higher concentrations required to achieve the effect and the potential for ionization suppression [35, 41]. Indeed, the high concentration of glycerol in these experiments induced ionization suppression at 1:1 inner-to-outer capillary flow rates. However, having independent solvent pumps connected to our coaxial probe gave us the flexibility to circumvent the ion suppression by lowering the flow rate of the external capillary (5:1 inner:outer flow rates), striking a balance between the amount of supercharging agent needed to induce higher charge states, without unnecessarily diluting the protein or suppressing ionization.
Applications such as capillary electrophoresis and coated microdroplets are only a fraction of the full potential of coaxial ESI. We have demonstrated that coaxial ESI provides an accessible means using conventional ESI probes for investigating and utilizing rapid in-source reactions that were previously possible only through nanospray/dual-spray assemblies with specialized solvent delivery instrumentation. Its simple assembly, unique mixing properties, and independent coaxial solvent delivery make coaxial ESI a viable alternative to more complicated systems as demonstrated by these biphasic, HDX, and supercharging examples.
- 11.Chen, H.W., Venter, A., Cooks, R.G.: Extractive electrospray ionization for direct analysis of undiluted urine, milk and other complex mixtures without sample preparation. Chem. Commun. 2042–2044 (2006)Google Scholar
- 21.Ling, X., Ding, C.F.: Online electrochemistry/electrospray mass spectrometry with a coaxial probe for investigation of electrochemical derivatization of anthracene with dodecylamine. Acta Phys. -Chim. Sin. 28, 2616–2624 (2012)Google Scholar
- 32.Jeong, E.S., Cha, E., Cha, S., Kim, S., Bin Oh, H., Kwon, O.S., Lee, J.: Online simultaneous hydrogen/deuterium exchange of multitarget gas-phase molecules by electrospray ionization mass spectrometry coupled with gas chromatography. Anal. Chem. 89, 12284–12292 (2017)Google Scholar