Assessment of Reproducibility of Laser Electrospray Mass Spectrometry using Electrospray Deposition of Analyte
- 384 Downloads
A nonresonant, femtosecond (fs) laser is employed to desorb samples of Victoria blue deposited on stainless steel or indium tin oxide (ITO) slides using either electrospray deposition (ESD) or dried droplet deposition. The use of ESD resulted in uniform films of Victoria blue whereas the dried droplet method resulted in the formation of a ring pattern of the dye. Laser electrospray mass spectrometry (LEMS) measurements of the ESD-prepared films on either substrate were similar and revealed lower average relative standard deviations for measurements within-film (20.9%) and between-films (8.7%) in comparison to dried droplet (75.5% and 40.2%, respectively). The mass spectral response for ESD samples on both substrates was linear (R2 > 0.99), enabling quantitative measurements over the selected range of 7.0 × 10−11 to 2.8 × 10−9 mol, as opposed to the dried droplet samples where quantitation was not possible (R2 = 0.56). The limit of detection was measured to be 210 fmol.
KeywordsFemtosecond laser vaporization Electrospray deposition Quantitative analysis
LEMS  is a surface sampling technique that employs an intense fs duration laser pulse to vaporize analyte followed by electrospray postionization at atmospheric pressure. The nonresonant fs laser pulse vaporizes all molecules, and thus sample preparation and matrix application is not necessary. Direct analysis of explosives [2, 3, 4], smokeless powders , proteins [6, 7], pharmaceuticals , plant tissue [9, 10], and biological tissue imaging with high spatial and spectral resolution has been performed . In principle, the method eliminates two sample biases because all molecules are vaporized into the gas phase, and all molecules, whether hydrophilic or hydrophobic, are transferred to the electrospray droplets. The preferential ionization of more hydrophobic species in the electrospray droplet is not observed for small molecules and proteins [12, 13].
The dried droplet method is a common method of sample preparation where liquid aliquots of sample are deposited onto a substrate followed by drying at room temperature [7, 14, 15, 16, 17, 18]. While the dried droplet technique is convenient, a spatially inhomogeneous distribution of analytes is deposited, typically in a ring-like pattern [19, 20]. This phenomenon is responsible for the high signal variance measured in previous LEMS experiments and is due to vaporization of an inconsistent amount of sample from the surface resulting in a high relative standard deviation (RSD) of 40% [3, 12]. A similar situation has been observed in MALDI analyses  where dried droplet samples of desamino [8-D-arginine] vasopressin (DDAVP) were detected with an RSD of ~50%. To increase the homogeneity of the sample film, many techniques have been developed, including fast evaporation , aerospray [23, 24], and electrospray deposition (ESD) [21, 25, 26, 27, 28, 29]. The ESD method prepares a homogeneous film of sample, resulting in a decrease in the measured variance in the DDAVP MALDI experiment where the RSD improved to ~15% . Similarly, in a desorption/ionization on silicon mass spectrometry (DIOS-MS) experiment, analysis of the peptide thymopentin (m/z = 679.2 Da) and the amino acids phenylalanine and tyrosine resulted in an RSD of 7% or better for the ESD samples compared with 6%–20% for dried droplet samples .
Previous experiments [3, 16] have shown that ESI-MS measurements had a lower signal variance in comparison with LEMS measurements. The higher RSD for the LEMS measurements was attributed to inhomogeneity of the films prepared by dried droplet technique. To test the hypothesis that the inhomogeneity of the sample film resulted in the LEMS signal variability, electrospray deposition is used here to prepare uniform films for comparison with dried droplet deposition. In this investigation, the ESD and dried droplet method were employed to prepare samples of Victoria blue on stainless steel and indium tin oxide (ITO) coated glass substrates. The films were analyzed with the LEMS technique. The mass spectral responses of Victoria blue solutions were measured as a function of concentration. The reproducibility was evaluated by calculating the within-film and between-films relative standard deviations. The reproducibility for the ESD samples was compared with the dried droplet samples.
Solid samples of Victoria blue, methylene blue, and sample slides of ITO coated glass were purchased from Sigma Aldrich (St. Louis, MO, USA). Acetic acid was purchased from J. T. Baker (Phillipsburg, NJ, USA). HPLC grade water and methanol were purchased from Fisher Scientific (Pittsburgh, PA, USA). A stock solution of Victoria blue was prepared in a plastic container at a concentration of 10–2 M using 1:1 (v/v) water/methanol. An aliquot of the stock solution was then diluted into 1:1 (v/v) water/methanol and 1% acetic acid to yield a final concentration of 50, 125, 250, 500, 1000, and 2000 μM respectively.
Factors contributing to a stable spray include the flow rate of the solution, the voltage applied between the capillary tip and the substrate, and the distance between the tip of the capillary and the substrate. Previous studies have reported that a circular conductive plate (guard plate) that is physically attached to the needle can be used to establish a more steady Taylor cone jet [31, 32]. The addition of the guard plate controls the electric field near the tip of the capillary and allows the system to be operated in a wider range of applied voltages. The guard plate can be used to control the angle of divergence in the spray plume by adjusting the distance between the guard plate and the tip of the capillary . As the distance is reduced, the spray plume angle becomes narrower. Using these parameters, the solution film of Victoria blue had a diameter of 3 mm. Five films of sample were prepared for each concentration.
To prepare samples by dried droplet method, 1.4 μL aliquots of solution were pipetted onto the substrate (stainless steel or ITO coated glass) and allowed to dry at room temperature. Dried droplet deposition resulted in a film of sample with diameter of approximately 5 mm. Five films of sample were prepared for each concentration.
Laser Vaporization and Electrospray Ionization
The instrumentation for laser vaporization, electrospray ionization, and mass spectral detection has been described in detail previously . A Ti:sapphire laser oscillator (KM Laboratories, Inc., Boulder, CO, USA) seeded a regenerative amplifier (Coherent, Inc., Santa Clara, CA, USA) to create a 2 mJ pulse centered at 800 nm with a duration of 70 fs, operating at 10 Hz to couple with the electrospray ion source (ESI). The laser was focused to a spot size of ~450 μm in diameter using a 16.9 cm focal length lens, with an incident angle of 45 with respect to the sample. The intensity of the laser at the substrate was approximately 1.5 × 1013 W/cm2. The mass spectrometer capillary inlet was biased to −4.5 kV, while the electrospray needle was grounded. A copper sample stage was placed approximately 6 mm below the electrospray needle, and was biased to −1.8 kV to compensate for the distortion of the electric field between the mass spectrometer capillary inlet and the needle. Sample slides were placed on the sample stage attached to a mini Z-axis linear stage (UMR5.16; Newport, Irvine, CA, USA), which was mounted on a motorized XY microscopy stage (MLS203-1; Thorlabs, Newton, NJ, USA), to enable automatic XY translation of the sample slides under computer control.
Each film was laser vaporized at nine different locations in a 3 × 3 grid pattern. The sample was ejected in a direction perpendicular to the electrospray plume, where capture and post-ionization of the analytes occurred. The electrospray solution for laser vaporization analyses was 1:1 (v/v) water/methanol with 1% acetic acid, and 10 nM methylene blue was used as an internal standard. The flow rate of the ES solvent was set at 3 μL/min using a syringe pump (Harvard Apparatus, Holliston, MA, USA). A countercurrent of hot nitrogen gas (200 °C) was flowed at 4 L/min to assist the desolvation process. The vaporized and post-ionized analytes were analyzed by a high resolution mass spectrometer (Bruker MicroTOF-Q II; Bruker Daltonics, Bremen, Germany) which was tuned for m/z below 600. Note that prior to laser vaporization electrospray ionization, the mass spectrometer was calibrated with a solution of 99:1 (v/v) acetonitrile/ESI calibrant (#63606-10 ML; Fluka Analytical/Sigma Aldrich, Buchs, Switzerland) over a mass range of m/z 50–3000. Raw spectra were recorded at a rate of 5 Hz.
The mass spectra obtained in the first 10 s of each run were used for background subtraction. This was done by running the ESI without firing laser. The laser was fired at a spot within the film for 2 s at a repetition rate of 10 Hz and then the sample stage was moved to a new position. Nine spots were analyzed for each film. The ion current for m/z 478.3 ± 0.1 was monitored as a function of time and the area of each peak was integrated using Compass Data Analysis 4.0 software by Bruker Daltonics. The peak intensity ratio of the analyte, Victoria blue (m/z 478.3), to that of the internal standard, methylene blue (m/z 284.1), was extracted from the mass spectrum. The average peak intensity was plotted as a function of concentration. Measurements were acquired for sample amounts ranging from 7.0 × 10−11 to 2.8 × 10−9 mol. The relative standard deviation of the peak intensity was calculated for each film and averaged. To quantify the between-films reproducibility, the RSD of five averaged peak intensities was calculated.
Appropriate laser eye protection was worn by all lab personnel during the experiments and the high voltage area was enclosed in plexiglass to prevent accidental contact with the biased electrodes.
Results and Discussion
Reproducibilty of LEMS for the Quantitative Analysis of Victoria Blue Deposited on Stainless Steel
RSD Values for Victoria Blue Deposited onto Stainless Steel by ESD and Dried Droplet Method
Average within-film reproducibility (%RSD)
Between-films reproducibility (%RSD)
Moles of Victoria blue
7.0 × 10−11
18.3 ± 8.2
82.1 ± 35.8
1.8 × 10−10
22.5 ± 3.5
53.8 ± 19.1
3.5 × 10−10
23.3 ± 6.3
46.7 ± 24.3
7.0 × 10−10
27.2 ± 10.7
41.2 ± 22.7
2.8 × 10−9
30.9 ± 5.2
67.5 ± 24.1
To determine the limit of detection (LOD), a 2.1 μL aliquot of 5 μM Victoria blue was electrospray deposited onto the metal substrate yielding a total sample amount of 10.5 × 10−12 moles (10.5 pmol). The LEMS measurement was acquired by focusing the laser to a spot size of ~450 μm in diameter which resulted in a measurement with a signal-to-noise ratio (S/N) of 3. Therefore, approximately 2% of the Victoria blue sample was vaporized from the metal substrate during the LEMS analysis of each spot. Thus, the quantity of Victoria blue consumed from the sample per laser spot was 210 fmol.
LEMS of Victoria Blue Deposited onto ITO Coated Glass
Victoria blue was deposited onto transparent ITO coated glass to determine whether absorption of the laser by the substrate was required for vaporization of the sample and whether transparent substrates affect quantification. Supplementary Figure 1S a and b show microscopic images of pre- and post-analyzed Victoria blue samples on ITO coated glass prepared by ESD and dried droplet, respectively. The transparent surface would eliminate the absorption of the laser that might induce thermal vaporization, and we conclude that the dye molecule is absorbing the vaporization laser pulse.
Previous studies have shown that femtosecond laser vaporization occurs from a variety of surfaces due to nonresonant excitation of the sample . Trace amounts of pharmaceuticals, including loratadine, oxycodone, and atenolol have been deposited and detected on wood, steel, glass, and fabric . The signal to noise ratio (S/N) was approximately one order of magnitude higher for the steel substrate because of the ability of steel to compensate for the distortion of the electric field between the electrospray needle and the mass spectrometer capillary inlet. Dielectric materials like glass (1.0 mm in thickness) cause the buildup of charge in localized areas from both the electrospray plume and laser vaporization process. This results in the distortion of the electrospray potentials and a decrease in intensity for the dielectric substrates. This problem has been overcome by using ITO coated glass slide, which is covered with a thin layer of ITO providing a conductive transparent surface. Within-film RSD values for all ESD samples on ITO coated glass were within the range of 15.2% ± 3.3% to 22.1% ± 5.6%; the majority of the films analyzed displayed relative standard deviations of < 20%. In comparison, dried droplet measurements had RSDs of 65.4% ± 18.9% to 113.2% ± 40.0%. Between-films RSD values for ESD samples were within the range of 3.3%–9.9%, compared with 24.7%–55.6% for the dried droplet samples. The calibration curve for ESD samples had a linear response (R2 = 0.9996) in the range investigated (1.75 × 10−10 to 2.8 × 10−9 mol), whereas the dried droplet samples had poor linearity (R2 = 0.5699). The results are shown in Supplementary Table S1, Supplementary Figures 1S and 2S. The reproducibility, within-film and between-films RSDs, and linearity of the LEMS signal on ITO coated glass are comparable to LEMS measurement from the steel substrate since both substrates have conductive surfaces. Identical LEMS responses from both opaque and transparent surfaces suggest that femtosecond laser pulses couple directly with the Victoria blue molecules presumably via a multiphoton mechanism with laser intensity of approximately 1013 W/cm2.
Effects of the Number of Laser Shots on Mass Spectral Response
Electrospray sample deposition is used to prepare thin, homogeneous films of Victoria blue on stainless steel and ITO coated glass substrates for LEMS analysis. The mass spectral responses are similar for the films on both substrates and exhibit a significant improvement in LEMS signal stability in comparison with the conventional dried droplet method. Larger RSD values were observed for within-film measurements in comparison to between-films measurements. We attribute this to the fact that our ESD system did not homogeneously deposit material. Visual analysis of ESD samples revealed a disk-like pattern suggesting that the concentration is not consistent across the sample. A lower RSD for between-films measurements was observed because each film was deposited in the same manner and nine measurements were averaged together for a given film. We have demonstrated that spatially resolved quantitative analysis of ESD films is possible. The RSD values obtained using LEMS of ESD films suggest the ability to provide two-dimensional, spatially resolved chemical analysis, and verifies that LEMS has the ability to provide accurate mass spectrometric images.
The limit of detection was measured to be ~210 fmol per laser spot for Victoria blue using a S/N threshold of 3. The LOD of LEMS technique is higher than conventional ESI-MS presumably due to the low capture and ionization efficiency (~3%) of laser vaporized analytes in the ESI plume. The LOD of LEMS could be improved by using technologies such as the design of novel neutral material transfer devices , neutral/liquid interfaces , or continuous flow solvent probes , which are able to increase the neutral capture and ionization efficiency for mass spectrometric techniques with use of ESI as a postionization source.
The authors gratefully acknowledge the funding support from the National Science Foundation (CHE-1362890, CHE-0957694).
- 4.Brady, J.J., Flanigan, P.M., Perez, J.J., Judge, E.J., Levis, R.J.: Multidimensional detection of explosives and explosive signatures via laser electrospray mass spectrometry. In: Proceedings of SPIE (presented in part): Chemical, Biological, Radiological, Nuclear, and Explosives (CBRNE) Sensing XIII, Baltimore, MD, 8358, 83580X, 1 May 2012Google Scholar
- 6.Karki, S., Sistani, H., Archer, J.J., Shi, F., Levis, R.J.: Isolating protein charge state reduction in electrospray droplets using femtosecond laser vaporization. J. Am. Soc. Mass Spectrom. (2017). doi: 10.1007/s13361-016-1576-9
- 8.Judge, E.J., Brady, J.J., Dalton, D., Levis, R.J.: Analysis of pharmaceutical compounds from glass, fabric, steel, and wood surfaces at atmospheric pressure using spatially resolved, nonresonant femtosecond laser vaporization electrospray mass spectrometry. Anal. Chem. 82, 3231–3238 (2010)CrossRefGoogle Scholar
- 10.Flanigan IV, P.M., Radell, L.L., Brady, J.J., Levis, R.J.: Differentiation of eight phenotypes and discovery of potential biomarkers for a single plant organ class using laser electrospray mass spectrometry and multivariat e statistical analysis. Anal. Chem. 84, 6225–6232 (2012)CrossRefGoogle Scholar
- 11.Shi, F., Archer, J.J., Levis, R.J.: Nonresonant, femtosecond laser vaporization and electrospray post-ionization mass spectrometry as a tool for biological tissue imaging. Methods. 144, 79–85 (2016)Google Scholar
- 14.Brady, J.J., Judge, E.J., Simon, K., Levis, R.J.: Laser electrospray mass spectrometry of adsorbed molecules at atmospheric pressure. In: Proceedings of SPIE: Imaging, Manipulation, and Analysis of Biomolecules, Cells, and Tissues VIII, San Francisco, CA, 7568, 75680R, 24 Feb 2010Google Scholar
- 18.Flanigan IV, P.M., Shi, F., Perez, J.J., Karki, S., Pfeiffer, C., Schafmeister, C., Levis, R.J.: Determination of internal energy distributions of laser electrospray mass spectrometry using thermometer ions and other biomolecules. J. Am. Soc. Mass Spectrom. 25, 1572–1582 (2014)CrossRefGoogle Scholar
- 28.Wei, H., Nolkrantz, K., Powell, D.H., Woods, J.H., Ko, M.C., Kennedy, R.T.: Electrospray sample deposition for matrix‐assisted laser desorption/ionization (MALDI) and atmospheric pressure mass spectrometry with attomole detection limits. Rapid Commun. Mass Spectrom. 18, 1193–1200 (2004)CrossRefGoogle Scholar
- 29.Axelsson, J., Hoberg, A.M., Waterson, C., Myatt, P., Shield, G.L., Varney, J., Haddleton, D.M., Derrick, P.J.: Improved reproducibility and increased signal intensity in matrix‐assisted laser desorption/ionization as a result of electrospray sample preparation. Rapid Commun. Mass Spectrom. 11, 209–213 (1997)CrossRefGoogle Scholar