Subscriber access provided by NEW YORK MED COLL
Article
Reactive charged droplets for reduction of matrix effects in electrospray ionization mass spectrometry Dmytro S. Kulyk, Colbert F Miller, and Abraham K. Badu-Tawiah Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.5b02943 • Publication Date (Web): 05 Oct 2015 Downloaded from http://pubs.acs.org on October 10, 2015
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
Analytical Chemistry
In-situ modification of electrospray (ES) droplets (instead of the original analyte solution) with acid vapor (see picture) makes quantitative ESI-MS more sensitive, precise and accurate. Other features, made possible by inserting the ES emitter into a cavity, include increased reaction rates, improved dynamic range, and the generation of ion intensities that reflect actual analyte concentrations in mixtures (see picture for elimination of ion suppression effects in electrospray) ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 9
Reactive charged droplets for reduction of matrix effects in electrospray ionization mass spectrometry Dmytro S. Kulyk, Colbert F. Miller, and Abraham K. Badu-Tawiah* Department of Chemistry and Biochemistry, The Ohio State University, 100 West 18th Ave, Columbus, OH 43210, USA ABSTRACT: A new quantitative contained-electrospray (ES) process is described here that employs a movable ES emitter to control the reactivity of charged micro-droplets by varying their exposure time with acid vapor. The method allows elimination of ion suppression effects due to the presence of various surface active compounds that co-elute with the analyte. For mixtures, contained-ESI mass spectrometric analysis produces relative ion intensities that reflect the true concentrations of analytes in solution. The mechanism for this effect has been elucidated and ascribed to the generation of fine initial droplets in the presence of high abundance of protons; together these two factors eliminate competition for charge and space during ion formation. Examples of analytes tested include steroids, phospholipids, phosphopeptides and sialylated glycans. At least one order of magnitude improvement in detection limits, sensitivity, and accuracy of detection were observe when compared to conventional electrospray.
Today electrospray ionization (ESI) has become available as ion source of choice for a variety of different mass spectrometers,1-3 and is routinely employed for proteomics,4 drug discovery5 and bioanalysis.6,7 ESI mass spectrometry (MS) provides a sensitive, robust, and reliable tool to characterize non-volatile and thermally labile bio-molecules that are not amenable to analysis by other conventional techniques. In spite of these advances, concentration- and compound-dependent ion suppression caused by the presence of endogenous or exogenous matrices remain a constant battle in ESI-MS,6,7 which has beset quantitative analysis. Traditionally, matrix effects in ESI-MS are eliminated by physical separation of the analyte from the matrix or other interfering analyte(s) via liquid chromatography (LC);6 it is a common practice also to perform liquid-liquid extraction or enrichment step prior to LC.8 Due to the inherent mass selectivity of tandem mass spectrometry (MS/MS), many interfering sources can be removed by mass filtration.9,10 Although many substances available in the matrix will not be detected in the MS/MS mass spectrum, they can co-elute with the analyte6,7 causing ionization inefficiencies and signal variations that affect accuracy, precision, and ease of detection. In an attempt to improve ESI ionization efficiency in the presence of co-eluted analytes/matrices, we developed a new electrospray-based apparatus that is capable of ion generation and reaction in a single experimental step. Our motivation was to use the accelerated reaction rates typical under the charged micro-droplet environment11-15 for on-line analytes modification, and thus enhance their detection. We hypothesized that because all major mechanisms16-19 (e.g., competition for charge16 and
space,17 variation in solvent viscosity and surface tension due to the presence of matrix compounds,18,19 etc.) related to ion suppression occur in the charged droplet environment, it is necessary to develop methods that overcome ion suppression during the stages of the droplet or ion formation. We selected and tested chemical systems that are particularly difficult to ionize by ESI (e.g., steroids, sialylated glycan, phosphorylated lipids and peptides), and prepared separate mixtures containing (i) surface active matrices such as quaternary ammonium species that do not compete with the analyte for protons, and (ii) long chain alkyl amines to represent surface active species that compete for protons. The results after using the new ESI ion source indicate that exposure of electrospray droplets with headspace vapor of hydrochloric acid (HCl) represents an efficient means to reduce matrix effects during ESI-MS analysis; the relative ion intensities in the resultant mass spectra were found to reflect the actual concentrations of analytes present in the original solution. On-line modification of electrospray droplets represents a recent approach to enhancing analyte detection in ESI-MS. This approach has been utilized for the manipulation of bio-molecular charge state20-26 and analysis of salt-rich solutions.27-29 Both liquid reagents (via mixing of electrospray droplets generated from two different emitters/chambers)24-27,30,31 and gaseous acids and bases (doped in sheath gas and used to intercept droplet at atmosphere/vacuum (1 – 2 torr) interface)20-22 have been used. Some of the techniques28,30,31 for droplet modification fall under the rubrics of spray-based ambient ionization in which the analyte is removed from the ESI spray solvent and placed on a surface.32-34 In our experiments,
ACS Paragon Plus Environment
Page 3 of 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
the sample, nebulizer gas and modifying reagent are all housed (or contained) within a single apparatus, external to the mass spectrometer. This feature can allow implementation of our approach on variety of mass spectrometers, without the need for instrument modification. To achieve controllable and real-time electrospray droplet modification, we developed a new contained-ESI ion source (see experimental section for details). The apparatus embodies a single cross Swagelok unit and two concentric emitters – with a movable inner ES capillary – and can be operated in two different modes: (i) Type I Mode - the ES inner emitter is pushed slightly (~ 0.5 mm) outside of the outer capillary. This allows the doping of headspace vapor of acid into the charged droplet environment on microsecond time scale (distance between spray tip to MS inlet is 2-5 mm; droplet velocity is ~100 m/s),35 and (ii) Type II Mode - where the ES emitter is pulled inside (~5 mm) the outer capillary to create a cavity in which extended droplet modification occur. We discuss the performance of Type I operation mode first.
EXPERIMENTAL SECTION Contained-ESI apparatus. To construct the contained-ESI ion source (Scheme 1), the ES emitter (inner capillary) was inserted into a second capillary (outer capillary). This concentric ES emitter was created using a 1/16 cross Swagelok element and allowed three inlets in the ion source: analyte solution, nebulizer gas, and head space vapor of a reactive gas. There is only one outlet for continuous droplets modification and release after the application of DC voltage to the ES emitter. The main experimental parameters were as follows: solvent flow rate, 3.5 μL/min; spray voltage, 4 kV; nitrogen gas pressure, 140 psi; distance from ion source to MS analyzer inlet, 2-5 mm, 0.5 mL of HCl, and 5 mm cavity for Type II operational mode. (B) Different Operation Modes
(A) Contained-ES Apparatus N2
(i) TYPE I
kV Cavity Spray Solvent
MS
N2
N2
(ii) TYPE II ES Emitter
Outer Capillary
N2
Cavity (5 mm)
HCl (l) (headspace vapor)
Scheme 1. (A) Experimental apparatus for the contained-ESI process, in the presence of cavity. Electrospray emitter (100 µm ID fused silica (FS) capillary) and the outer capillary (250 µm ID FS capillary) are held in place by 0.2 and 0.4 mm graphite ferrules, respectively. (B) Different modes of operation for the contained-ESI apparatus: (i) Type I, and (ii) Type II operational modes are used with volatile reagents, for brief (microseconds) and extended droplet modification times scales, respectively.
Mass spectrometry and pH measurement. Samples were analyzed by a Thermo Fisher Scientific Velos Pro LTQ mass spectrometer (San Jose, CA, USA). MS parameters used were such as 150 °C capillary temperature, 4 kV spray voltage, 3 microscans, 100 ms ion injection
time, and 60% S-lens voltage. Spectra were obtained for at least 30 s, yielding an average of 300 individual scans. Thermo Fisher Scientific Xcalibur 2.2 SP1 software was applied for MS data collecting and processing. Tandem MS with collision-induced dissociation (CID) was utilized for analyte identification. 1.5 Th (mass/charge units) for isolation window and 30% (manufacturer’s unit) of normalized collision energy was chosen for the CID tests. Mettler Toledo pH/Ion meter S220 (Schwerzenbach, Switzerland) was applied for solution phase pH measurements. pH of droplets was measured by using Micro Essential Laboratory ultrasensitive pH paper (Hydrion UltraFine; Brooklyn, NY, USA). Chemicals and Reagents. Girard’s T reagent (99%), cortisone (98%), corticosterone (92%), 6′-sialyl-D-lactose (98.0%), 2-oleoyl-1-palmitoyl-sn-glycero-3phosphocholine (99.0%), methanol (99.9%, HPLC Grade), and acetonitrile (99.9%, HPLC Grade) were purchased from Sigma Aldrich (St. Louis, MO). 1.0 mg/mL standard solutions of benzoylecgonine, (-)-Δ9tetrahydrocannabinol, cocaine, and (±)methamphetamine were obtained from Cerilliant (Round Rock, TX). Dodecyl amine was purchased from Acros Organics (Geel, Belgium), hydrochloric acid (37.5%) was obtained from Fisher Scientific (Pittsburgh, PA), and ethanol 200 proof (99.5%) was acquired from Decon Laboratories (King of Prussia, PA). Milli-Q water was used for all solutions.
RESULTS AND DISCUSSION Fig. 1A shows a typical ESI-MS analysis of equimolar (100 µM) mixture of cortisone and Girard T reagent (GT, a quaternary ammonium compound) at neutral pH. Notice the significant suppression effect by the quaternary ammonium ion, GT.36,37 Optimization of flow rate, spray voltage and inlet capillary temperature did not prevent the suppression effect (Fig. 2A and Fig. S1, Supporting Information); this includes an attempt to increase the relative amount of cortisone in the mixture (see Fig. 1C for 1:3 molar ratio of GT/cortisone). The same equimolar mixture at neutral pH was analyzed by electrospray, but the derived charged droplets were briefly exposed to headspace vapor of HCl, using the contained-ESI apparatus in the Type I operation mode (Fig. 1B). At a reduced nebulizer gas pressure of 100 psi, the trend was reversed in which protonated cortisone ions became the most abundant species in the mass spectrum (compare Fig. 2A with 2B). Further optimization of nebulizer gas pressure and flow rate (Fig. S2, Supporting Information) revealed that the ion suppression effect was significantly removed at 30 psi (Fig. 2C). Fig. 2C also suggests that between 130 – 150 psi of nitrogen gas pressure, the contained-ESI ion source can yield mass spectra in which the relative ion intensities reflect the true concentration of analytes present in the mixture (Fig. 1B and D). For example, contained-ESI analysis of cortisone/GT mixtures prepared separately in 1:1, 1:2, 2:1, 1:3 and 3:1 mole ratios produced ion intensities that reflected the actual molar concentrations of each species
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 4 of 9
present in the respective solutions (Fig. S3, Supporting Information). In all cases, the intensity of all ion forms of cortisone including the protonated dimer (m/z 721) and sodiated (m/z 383) ions were included in the assessment. Improvements in quantitation accuracy of at least two orders of magnitude were observed for these mixtures when analyzed with contained-ESI ion source (Table S1, Supporting Information). In the presence of acid
vapor, the intensity of protonated cortisone also varied quantitatively and linearly (Fig. S4, Supporting Information) over a wide analyte concentration range (up to 75 µM) compared with the sub-micromolar concentration limits for conventional ESI ion source.38 Similar effects was observed for other steroids (Fig. 3). Aside from these binary analyte mixtures, a solution containing three structurally different illicit drugs Δ9-tetrahydrocannabinol (Δ9-THC), methamphetamine and benzoylecgonine, were prepared in equimolar quantities at 0.5 µM, and was analyzed both by the conventional and the contained ESI ion sources (Fig. 4). Signal intensity for both analysis techniques was similar, but in the case of contained-ESI, protonated cations of all three drugs had comparable intensities, indicating similar concentrations in solution.
Figure 1. GT (MW 132)/cortisone (MW 360) mixtures were prepared in 1:1 (A and B) and 1:3 (C and D) molar ratios and analyzed using conventional ESI and contained-ESI (Type I mode) ion sources, respectively. That is, A and C were analyzed using the conventional ESI, and B and D were recorded using the new contained-ESI at 140 psi nebulizer gas pressure. For the contained-ESI experiment, correct relative abundances that reflect analyte concentration in solution are obtained by taking into account all forms of cortisone ions including sodiated [M+Na]+ and protonated dimer [2M+H]+ ionic species.
Figure 3. (A) Conventional ESI and (B) contained-ESI (Type I mode) mass spectra of solution containing equimolar (100 µM) mixture of corticosterone and GT at 140 psi of nebulizer gas pressure.
Figure 2. Effect of spray voltage on ion intensity when equimolar (100 µM) mixture of cortisone and GT was analyzed with (A) conventional ESI ion source, in the absence of acid and (B) contained-ESI apparatus (Type I mode) in the presence of acid vapor; 100 psi of nebulizer gas pressure was used in both experiments. (C) The effect of nitrogen nebulizer gas pressure on ion yield when using the contained-ESI apparatus, in the presence of acid vapor.
Figure 4. Mass spectra obtained following the analysis of solution containing equimolar (0.5 µM) mixture of methamphetamine, benzoylecgonine, and Δ9-THC using (A) conventional ESI. and (B) contained-ESI (Type I mode), with HCl vapor heated to 50 oC.
The ability of the contained-ESI ion source to reduce ion suppression effects, and the marked pressure effect observed in Figure 2C are ascribed to differences in the size of the initial electrospray droplets at different nebulizer gas pressures (Scheme 2).
ACS Paragon Plus Environment
Page 5 of 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
HIGH PRESSURE
LOW PRESSURE (A) Conventional ESI
(B) Contained-ESI
GT+ GT+ H+ GT+ M GT+ H+ GT+ GT+
H+ H+ M M H+ GT+ M H+ M XM MH+ H+
Series of droplet evaporation and fission events
xGT+ + yMH+
xMH+ + yGT+
(C) Conventional ESI
M
H+M
H+
H+
GT+
H++
GT
(D) Contained-ESI + H+MH + H + + + H H+ H + GTH+ H+MH + H + H+ H+ H++ H GT H+ H+
Series of droplet evaporation and fission events
xGT+ + yMH+
xMH+ + xGT+
Scheme 2. Proposed mechanism of ion suppression effect, and the effect of nebulizer gas pressure on the final gas-phase ion intensity: (A) and (C) conventional ESI, in the absence of acid and at low and high pressures, respectively; (B) and (D) contained-ESI (Type I mode), in the presence of acid and at low high N2 gas pressures, respectively. x and y represent the number of the respective ions, where x >> y in A and B. M = cortisone; GT = Girard T reagent; X= counter anion.
Bigger initial charged droplets are generated at low pressures, and in the presence of acid vapor high concentration of protons occupy the droplet surface to form the source of excess charge on the droplet. GT ions form the excess charge on the droplet surface when using the conventional ESI ion source and in the absence of acid vapor (Schemes 2A versus B; x and y represent absolute ion intensities where x>>y). The bigger droplets presumably contain both GT and cortisone (M), but the one having high concentration of proton at its surface increases the surface activity of cortisone through polarization, while GT is neutralized (by counter ions) at the core of the droplet.37 We believe this process of selective movement of cortisone towards droplet surface (triggered by excess protons) results in the suppression of the quaternary ammonium species when using the contained-ESI apparatus at low pressures (Fig. 2C, and Fig. S5). At higher N2 gas pressures, however, there is high desolvation efficiency39 which results in the formation of much smaller initial droplets compared with lower pressure. Under such limited droplet volumes, competition for charge and space is eliminated to provide equal opportunity for ion formation (Scheme 2D). In the absence of acid vapor (Scheme 2C), ions intensities are dependent on the analyte’s own ionization efficiency. Further, an equimolar solution of dodecyl amine (DDA) and cortisone was selected to test if the hypothesis proposed in Scheme 2 will hold for surface active matrices that compete for protons during ESI. Again, this mixture was analyzed in the Type I mode, and in the presence of acid vapor. It was anticipated that since DDA has higher proton affinity (879.5 kJ/mol)40 than cortisone (
