Electrospray Ionization

---

ac research

Anal. Chem. 2009, 81, 868–874

Articles Using Laser-Induced Acoustic Desorption/ Electrospray Ionization Mass Spectrometry To Characterize Small Organic and Large Biological Compounds in the Solid State and in Solution Under Ambient Conditions Sy-Chyi Cheng,† Tain-Lu Cheng,‡,§ Hui-Chiu Chang,§,| and Jentaie Shiea*,†,§ Department of Chemistry, National Sun Yat-Sen University, Kaohsiung 804 Taiwan, Institute of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan, National Sun Yat-Sen University-Kaohsiung Medical University Joint Research Center, and Faculty of Biomedical Science and Environmental Biology, Kaohsiung Medical University, Kaohsiung, Taiwan We have coupled laser-induced acoustic desorption (LIAD) with electrospray ionization (ESI) mass spectrometry (LIAD/ESI/MS) to characterize molecules in the solid state and in solution under ambient conditions. To perform an LIAD/ESI analysis, the sample droplet is deposited on the surface of a thin aluminum foil by a micropipette; the rear side of the foil with the sample spot is then irradiated with a pulse from a Nd:YAG IR laser. The resulting shockwave and heat cause the sample on the rear side to change from the condensed phase to the gas phase. The desorbed species then move upward to enter an ESI plume to react with charged solvent species (methanol- and water-related ions and droplets), forming singly or multiply charged analyte ions. A quadrupole/ time-of-flight (Q-TOF) mass analyzer attached to the LIAD/ESI source detects the analyte ions to obtain an ESIlike mass spectrum. Both small organic and large biological compounds (including amino acids, peptides, and proteins) were successfully ionized and detected by the LIAD/ESI/MS system. Although native and denatured myoglobin ions were both detected from a liquid sample solution, only the denatured myoglobin ions were detected from a dried sample. Laser-induced acoustic desorption (LIAD) is a technique developed for desorbing and ionizing organic and biological * To whom correspondence should be addressed. E-mail: jetea@ mail.nsysu.edu.tw. † National Sun Yat-Sen University. ‡ Institute of Medicine, Kaohsiung Medical University. § National Sun Yat-Sen University-Kaohsiung Medical University Joint Research Center. | Faculty of Biomedical Science and Environmental Biology, Kaohsiung Medical University.

868

Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

molecules from surfaces of solid substrates in the vacuum.1-6 Because this molecular desorption/ionization method does not require direct exposure of the sample to intense laser light, it can be more useful than other laser desorption methods [e.g., matrix-assisted laser desorption/ionization (MALDI)7 or laser desorption (LD)8] in cases where the sample is sensitive to the laser wavelength used. In LIAD, the sample solution is first deposited directly on either a thin metal surface or a thermally insulating layer.1-6,9-17 After drying, the sample spot is (1) Lindner, B.; Seydel, U. Anal. Chem. 1985, 57, 895–899. (2) Golovlev, V. V.; Allman, S. L.; Garrett, W. R.; Chen, C. H. Appl. Phys. Lett. 1997, 71, 852–854. (3) Golovlev, V. V.; Allman, S. L.; Garrett, W. R.; Taranenko, N. I.; Chen, C. H. Int. J. Mass Spectrom. Ion Processes 1997, 169, 69–78. (4) Peng, W. P.; Yang, Y. C.; Kang, M. W.; Tzeng, Y. K.; Nie, Z. X.; Chang, H. C.; Chang, W.; Chen, C. H. Angew. Chem., Int. Ed. 2006, 45, 1423– 1426. (5) Peng, W. P.; Lin, H. C.; Lin, H. H.; Chu, M.; Yu, A. L.; Chang, H. C.; Chen, C. H. Angew. Chem., Int. Ed. 2007, 46, 1–7. (6) Peng, W. P.; Yang, Y. C.; Lin, C. W.; Chang, H. C. Anal. Chem. 2005, 77, 7084–7089. (7) Tanaka, K.; Waki, H.; Ido, Y.; Akita, S.; Yoshida, Y.; Yoshida, T. Rapid Commun. Mass Spectrom. 1988, 2, 151–153. (8) Posthumus, M. A.; Kistemaker, P. G.; Meuzelaar, H. L. C.; Brauw, M. C. Anal. Chem. 1978, 50, 985–991. (9) Shea, R. C.; Habicht, S. C.; Vaughn, W. E.; Kentta¨maa, H. I. Anal. Chem. 2007, 79, 2688–2694. (10) Shea, R. C.; Petzold, C. J.; Campbell, J. L.; Li, S.; Aaserud, D. J.; Kentta¨maa, H. I. Anal. Chem. 2006, 78, 6133–6139. (11) Pe´rez, J.; Ramı´rez-Arizmendi, L. E.; Petzold, C. J.; Guler, L. P.; Nelson, E. D.; Kentta¨maa, H. I. Int. J. Mass Spectrom. 2000, 198, 173–188. (12) Campbell, J. L.; Crawford, K. E.; Kentta¨maa, H. I. Anal. Chem. 2004, 76, 959–963. (13) Campbell, J. L.; Fiddler, M. N.; Crawford, K. E.; Gqamana, P. P.; Kentta¨maa, H. I. Anal. Chem. 2005, 77, 4020–4026. 10.1021/ac800896y CCC: $40.75  2009 American Chemical Society Published on Web 12/30/2008

irradiated from the rear side with a pulsed laser in the vacuum. The resulting ablation of the metal by the laser pulse creates a large-amplitude acoustic wave (shockwave) that propagates through the metal foil. It was suggested that the molecules predeposited on the other side of the foil would be desorbed through the mechanical actions of the acoustic wave, and the fast expansion of the acoustic wave ensured that the molecules did not break apart.1-3 Previous studies indicated that the wavelength of the laser beam did not have a critical effect on ion intensity but the power density of the laser on the focused spot was an important factor.9 In comparison to that used in LD and MALDI, the laser power used in LIAD is usually much higher, so strong acoustic waves can be generated for desorption.3 The material of the foil is another important factor. It was found that the material with large thermal expansion and small thermal conductivity coefficients (e.g., mercury and titanium) best allows for propagation of the acoustic wave.2,3,10 A unique feature of the acoustic desorption process in LIAD is that the laser energy transfers specifically to the surfaceadsorbed bonds with little or no excitation of intramolecular vibrational states.3 With the use of a simple LIAD probe, a wide variety of molecules have been desorbed as intact neutral species; however, ionization of large analytes such as proteins with a molecular weight greater than 20 000 is generally difficult.3,4 This behavior arises presumably because most of the mechanical energy provided by the acoustic wave is coupled into the desorption stage so that insufficient energy is available for ionization; therefore, the ejection of neutral species will predominate in LIAD unless the analyte molecules are precharged when deposited.1,3 The predominant desorption of neutral molecules, proposed by Lindner and Seydel, prompted many scientists to explore the combination of LIAD with postdesorption ionization to increase the detection limit of large analytes.1 The development of an efficient postionization process for neutral molecules desorbed through LIAD may provide new possibilities for improving the sensitivity of mass spectrometry (MS)-based analyses of large biomolecules. Previously, electron impact ionization (EI), chemical ionization (CI), and photoionizaton have been successfully used to postionize small molecules desorbed from petroleum distillates, base oil, synthetic polymers, saturated hydrocarbons, and peptides through LIAD in a vacuum.11-17 To obtain the ion signal of a large molecule like protein, a MALDI matrix was mixed with the sample solution to increase the desorption and ionization efficiencies of the protein molecule in LIAD prcesses.3 Unfortunately, the ionization efficiency and mass range of the detected protein remained low. Electrospray ionization (ESI) is a feasible postionization method for large molecules (such as peptides and proteins) present in the droplets produced by an ultrasonic nebulizer; this approach has been named “fused-droplet electrospray ionization” (FD-ESI).18-20 In FD-ESI, the analytes in a gaseous state or in (14) Crawford, K. E.; Campbell, J. L.; Fiddler, M. N.; Duan, P.; Qian, K.; Gorbaty, M. L.; Kentta¨maa, H. I. Anal. Chem. 2005, 77, 7916–7923. (15) Duan, P.; Qian, K.; Habicht, S. C.; Pinkston, D. S.; Fu, M.; Kentta¨maa, H. I. Anal. Chem. 2008, 80, 1847–1853. (16) Shea, R. C.; Petzold, C. J.; Liu, J.-a.; Kentta¨maa, H. I. Anal. Chem. 2007, 79, 1825–1832. (17) Zinovev, A. V.; Veryovkin, I. V.; Moore, J. F.; Pellin, M. J. Anal. Chem. 2007, 79, 8232–8241. (18) Shiea, J.; Wang, C. H. J. Mass Spectrom. 1997, 32, 247–250.

the neutral droplets are carried into an electrospray plume by nitrogen gas. An acidic methanol/water solution is usually used to generate the ESI plume. The sample subsequently reacts (via ion-molecule reactions) with the charged solvent species (e.g., protonated methanol and water ions) or fuses with the charged solvent droplets in the ESI plume to create ESI-like mass spectra. Because the ionization and nebulization processes are separate events in FD-ESI, independent control is available over the states of the sample solution and the composition of the ESI solvent. Moreover, by varying the methods used to introduce the sample into the ion source, unique applications for the analysis of gas and liquid samples have been developed. For example, ion signals have been detected from analytes introduced through outlets of gas chromatographs, Curie-point pyrolyzers, and ultrasonic and concentric nebulizers.21-25 Recently, we developed an ambient ionization technique called electrospray-assisted laser desorption ionization (ELDI) in which a pulsed laser beam is used to irradiate the surface of the solid sample; the desorbed analyte is then postionized in an ESI plume.26 The combination of LD and ESI in ELDI allows direct, sensitive, and rapid characterization of small organic and large biological compounds (peptides and proteins) in the solid state under ambient conditions.26,27 In ELDI, the laser beam directly irradiates the solid sample surface; this approach is the opposite of LIAD, where a metal foil shields the sample from exposure to the laser light. Obviously, the process of sample desorption in ELDI (direct laser desorption) is different from that in LIAD (shockwave desorption). However, like ELDI and FD-ESI, postionization of the desorbed neutrals in LIAD can be performed in an ESI plume. In this study, we developed a technique that combines ESI with LIAD under ambient conditions. The analytes in the neutral particles or droplets desorbed through LIAD are postionized upon entering an ESI plume. ESI’s characteristic generation of multiply charged ions allows us to overcome the innate shortcomings of LIAD (e.g., low mass limits) when detecting proteins in solid state and in a solution. EXPERIMENTAL SECTION All amino acid, peptide, and protein standards were purchased from Sigma (St. Louis, MO) and used without further purification. Yogurt and cow milk were purchased from a local supermarket. The human serum and whole blood samples were obtained from a healthy volunteer. Methanol was purchased from Merck (HPLCgrade, Darmstadt, Germany); acetic acid and glycerol were purchased from J. T. Baker (reagent-grade, Phillipsburg, NJ). Distilled deionized water (produced by Milli-Q plus, Millipore; Molsheim, France) was used to prepare the sample solutions. (19) Shiea, J.; Chang, D. Y.; Lin, C. H.; Jiang, S. J. Anal. Chem. 2001, 73, 4983– 4987. (20) Shiea, J.; Lee, C. C.; Chang, D. Y.; Jeng, J. J. Mass Spectrom. 2002, 37, 115–117. (21) Chang, D. Y.; Lee, C. C.; Shiea, J. Anal. Chem. 2002, 74, 2465–2469. (22) Shieh, I. F.; Lee, C. Y.; Shiea, J. J. Proteome Res. 2005, 4, 606–612. (23) Lee, C. Y.; Shiea, J. Anal. Chem. 1998, 70, 2757–2761. (24) Shiea, J.; Hong, C. M.; Tsai, F. C. Anal. Chem. 2000, 72, 1175–1178. (25) Hsu, H. J.; Kuo, T. L.; Wu, S. H.; Oung, J. N.; Shiea, J. Anal. Chem. 2005, 77, 7744–7749. (26) Shiea, J.; Huang, M. Z.; Hsu, H. J.; Lee, C. Y.; Yuan, C. H.; Beech, I.; Sunner, J. Rapid Commun. Mass Spectrom. 2005, 19, 3701–3704. (27) Huang, M.-Z.; Hsu, H.-J.; Lee, J.-Y.; Jeng, J.; Shiea, J. J. Proteome Res. 2006, 5, 1107–1116.

Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

869

Figure 1. Schematic representation of the desorption and ionization process occurring during LIAD/ESI/MS analysis.

Aluminum foil (thickness: 13 µm, purchased from a local supermarket) was used as the substance supporting the analytes (solid state or solution) and as a medium to transmit the shockwaves generated by rear-side pulse of the IR lasers. The samples prepared for LIAD/ESI/MS analysis include (1) three lines (0.15 cm wide for each line) drawn on the surface of the aluminum foil using blue-, red-, and green-colored markers, (2) dried spots of peptide and protein standard solutions, and (3) protein standard solutions and biological fluids (cow milk, yogurt, whole human blood, and serum). Because of rapid evaporation of the organic solvent in the ink, the color bands dried within 1 min. The foil was then placed on the sample plate that was pushed by a syringe pump, operated at a rate of 8 mm/min. While the sample plate was moving, the other side of the aluminum foil was irradiated continuously with an IR laser beam. To prepare the sample in sample 2, a micropipette was used to withdraw the sample solution (10 µL), which was then deposited on the aluminum foil (ca. 0.5-0.8 cm2); after drying, the foil was examined with LIAD/ESI/MS. In sample 3, after the foil was applied with the sample solution, it was immediately examined with LIAD/ESI/MS. During the LIAD/ESI/MS analysis, the aluminum foil charged with the analytical sample was positioned on a 3D stage; the rear side of the foil with the sample spot (either dry or wet) was irradiated with a pulsed Q-switched Nd:YAG laser (LS-2130, LOTIS TII, Belarus, Russia) under ambient conditions. The laser was operated at a wavelength of 1064 nm, a frequency of 2 Hz (controlled by a sweep function generator), a pulsed energy of ca. 20-30 mJ, and a pulse duration of 9 ns. The laser beam was focused through an objective lens to create a spot the size of 5 × 10-3 cm2 at a maximum power density of 109 W/cm2. After the irradiation with the laser beam, the desorbed neutral molecules, particles, and/or droplets entered an electrospray plume located on top of the laser spot. The configuration of the ESI source for LIAD was similar to that used in ELDI, pyrolysis/ ESI, and FD-ESI sources.18-27 The capillary electrosprayer was aligned parallel to the aluminum foil, ca. 2-3 mm above the 870

Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

sample spot. The ESI plume was directed toward the ion-sampling orifice. The electrospray needle and the sample plate potentials were both held at 0 V; the sampling cone voltage in the mass spectrometer was maintained at -4.5 kV. The electrospray solution was composed of methanol and water (1:1, v/v) containing 0.1% acetic acid. The electrospray solution was delivered through a capillary by a syringe pump operated at a flow rate of 4 µL/min. A nebulizing gas, commonly employed in conventional ESI sources, was not used during LIAD/ESI analysis. The ions generated in this manner were characterized using a quadrupole/ time-of-flight (Q-TOF) mass analyzer (Bio-TOF-q, Bruker Daltonics, Billerica, MA). The mass spectra were recorded at a scan rate of ca. 2 s/scan. The deconvoluted spectra of the protein ions were obtained using a DataAnalysis Version 3.2 (Bruker Daltonics). RESULTS AND DISCUSSION Figure 1 provides a schematic representation of the assemblage of a LIAD/ESI source. The typical procedure for the analysis of a sample using LIAD/ESI/MS involves the rear side of the sample spot on the aluminum foil being irradiated with an IR laser beam and then the desorbed neutral species entering the ESI plume, becoming ionized to generate ESI-like mass spectra. The thickness of the aluminum foil used in this study is 13 µm; this is similar to that reported in previous studies by Shea et al.10 However, for nonmetal or liquid metal LIAD foils such as Si and Hg, much thicker foils (e.g., 500 µm) are usually used.1,6 In comparison to other metal foils (e.g., Ag, Au, Cu, Fe, Ti, and Ta) used for LIAD analysis, aluminum has relatively higher thermal expansion coefficient (2.4 × 10-5/°C) and thermal conductivity (2.37 W/cm · K).9-17 Therefore, desorption of the analyte in LIAD/ESI/ MS might be due to both thermal and acoustic mechanisms induced by irradiating the rear side of the sample foil with an IR laser. We believe that the ionization mechanisms in the LIAD/ ESI source were similar to those of FD-ESI and ELDI, i.e., the desorbed neutral molecules were ionized through fusion with charged solvent droplets, followed by electrospray ionization or

Figure 2. (a) Photograph of aluminum foil displaying the path of desorption of dry inks after irradiating the rear side of the foil using an IR laser. (b-d) Extracted ion chromatograms of three ink ions: blue (m/z 478.3), red (m/z 443.3), and green (m/z 363.2). (e-g) Positive-ion LIAD/ESI mass spectra of the inks in panel a.

reaction with charged solvent species in the ESI plume through ion-molecule reactions.19,26,27 Because the relative distance between the ESI capillary, sample foil, and the interface of the sample inlet on a mass analyzer was short, we had to ensure that the aluminum foil was not positioned so close to the ESI source that it would distort the electric field between the ESI capillary and the inlet of the mass analyzer, causing the ESI plume to twist. Another unwanted situation was for the ESI plume to be directed toward the sample spot on the foil such that desorption electrospray ionization (DESI) could have occurred.28 Therefore, successfully obtaining ion signals through LIAD/ESI/MS requires the relative positions of the sample spot on the aluminum foil, the tip of the ESI capillary, and the sample inlet of the mass analyzer to be adjusted carefully. In a typical configuration that yielded analyte ion signals, the distance between the ESI capillary and the metal foil was 2 mm, and the distance between the ESI capillary and the inlet of the mass analyzer was 7 mm (see Figure 1). To examine the capability of using LIAD/ESI/MS to obtain analyte ion signals, we drew three bands on the surface of a piece of aluminum foil using blue-, red-, and green-colored markers. The photograph in Figure 2a of the front side of the aluminum foil after LIAD/ESI analysis revealed that the inks on the scanning route of the laser beam were depleted during LIAD/ESI analysis. Figure 2b-d shows the extract ion chromatograms (EICs) of the three ink ions, revealing the presence of different chemicals in the inks: m/z 478 for the blue band (M+•, Basic Blue 7, C33H39N3), m/z 443 for the red band (M+•, Rhodamine 6G, C28H31O3N2), and m/z 363 for the green band (unknown green dye). The ions present in each color band were detected only when the represented band was irradiated with the IR laser. Figure 2e-g present the LIAD/ESI mass spectra of the dye ions recorded from each color band. We also chose three amino acids (histidine, lysine, and serine) as standards to test the applicability of LIAD/ESI/MS for characterizing small compounds. Similar to those of the ink, protonated molecules of the amino acid standards (28) Taka´ts, Z.; Wiseman, J. M.; Gologan, B.; Cooks, R. G. Science 2004, 306, 471–473.

(m/z 156 for histidine, m/z 147 for lysine, and m/z 106 for serine) were detected (data not shown). These experimental results suggest that LIAD/ESI/MS is a useful analytical technique for characterizing small organic or biological compounds under ambient conditions. Although previous reports have demonstrated that singly charged protein ions (MH+) could be detected through LIAD using a TOF mass analyzer, the detection sensitivity and mass range of the peptide and protein ions have been low. To examine the applicability of LIAD/ESI for the desorption and ionization of large biological compounds, we employed aqueous solutions of angiotensin I, insulin, hemoglobin, lysozyme, and albumin standards (10-4 M, 5 µL aliquots) as analytical samples. Figure 3 displays the positive-ion LIAD/ ESI mass spectra recorded from the dry sample spots applied on the aluminum foil. We detected multiply charged ions (i.e., ESI-like ions) for all of these peptide and protein samples. The deconvoluted mass spectra (insets of Figure 3) show the molecular ions of the protein standards (m/z 5733 for insulin, m/z 15 121 for hemoglobin R-chain, m/z 15 862 for hemoglobin β-chain, m/z 14 306 for lysozyme, and m/z 66 307 for albumin). Similar results were also obtained in our previous studies using ELDI and FD-ESI.19-21,26,27 We used solutions containing different amounts of the hemoglobin standard (from 10-4 to 10-7 M) to estimate the detection limit of LIAD/ESI/MS. The hemoglobin ion series (+16 to +24) were detected in the solutions containing 10-4 (Figure 3c) and 10-5 M of hemoglobin standard; they were barely detected in the 10-6 M solution, and were not detected at all in the 10-7 M solution (data not shown). The detection limit of LIAD/ESI/MS for the analysis of hemoglobin was then estimated to be between 10-6 and 10-7 M. When using larger proteins such as the albumin standard as the analyte, however, the detection limit of LIAD/ESI/MS is only 10-5 M (data not shown). The detection of multiply charged protein ions in LIAD/ ESI/MS indicated that, after desorption through LIAD, large biological molecules were readily ionized through ESI proAnalytical Chemistry, Vol. 81, No. 3, February 1, 2009

871

Figure 3. Positive-ion LIAD/ESI mass spectra of dry peptide and protein solutions of (a) angiotensin I, (b) insulin, (c) hemoglobin, (d) lysozyme, and (e) albumin. The inset to panel a displays the enlarged mass spectrum, revealing singly charged angiotensin I ions. The insets to panels b-e display the deconvoluted protein standard ion signals.

cesses. Previously, the detected mass range for analytes in conventional LIAD/TOF was estimated to be less than 20 000 Da.3,4 The failure of conventional LIAD/TOF analysis to detect large protein ion signals may be due to either the energy transferred by the laser-induced acoustic wave not being sufficiently high to desorb large protein molecules or the ionization efficiency of LIAD being too low to generate a sufficient number of protein ions. In contrast, in this study, we have demonstrated that the protein ions having mass greater than m/z 66 000 (e.g., albumin) were detectable when using ESI for postionization (Figure 3e). These results reveal that (i) large protein molecules are readily desorbed through the action of the laser-induced acoustic wave and thermal desorption and (ii) ESI-mediated postionization enhances the ionization efficiency of the desorbed analyte molecules to such an extent that large protein ions (e.g., albumin) can be formed and detected. As mentioned above, the relative distances between the ESI capillary, the sample foil, and the interface of the sample inlet on a mass analyzer were adjusted carefully so that the ESI plume was not directed toward the sample spot on the foil, causing DESI to occur. To exclude the possibility of DESI processes occurring during LIAD/ESI, we used a myoglobin solution (10-5 M) to draw a line on the aluminum foil. After drying, the aluminum foil was placed on the sample plate, which was set 2 mm under the ESI capillary and 7 mm from the inlet of the Q-TOF mass analyzer. The ESI capillary provided a continuous electrospray of the acidic methanol solution while the syringe pump pushed the sample plate along its path. The IR laser was turned on twice during analysis (Figure 4, parts a and b). We detected ESI-like myoglobin ion signals only when the IR laser was switched on (cf., Figure 4c-e). These results exclude the possibility of generating protein ions through DESI processes during LIAD/ESI analysis. 872

Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

Figure 4. Photographs of the (a) front and (b) rear sides of an aluminum foil, revealing the path of desorption of a dry myoglobin solution line after irradiation of the rear side of the foil using an IR laser. (c) Total ion current trace for the analysis in panel a. (d and e) Positive-ion LIAD/ESI mass spectra of the background and myoglobin spots, respectively.

Conventionally, the LIAD source is operated under vacuum. Therefore, because of rapid evaporation of solvents under vacuum, LIAD can be used to analyze only solid or dry samples; the practicability of performing desorption/ionization through LIAD to detect analytes in solution has not been reported previously. In this study, we operated the LIAD/ESI source under ambient conditions; therefore, we suspected that liquid samples could be analyzed directly without the need for drying. We tested this hypothesis using aqueous solutions containing myoglobin and cytochrome c. Figure 5 displays photographs taken during desorption of a dry sample spot and a solution in a LIAD/ESI source. For comparison, Figure 5a provides a photograph of the electrospray plume in the absence of laser irradiation (i.e., when the laser power was switched off). The photograph in Figure 5b of the LIAD/ESI system analyzing a dry blue ink reveals that smoke containing the gaseous blue ink molecules was produced when the rear side of the sample spot was irradiated with the IR pulsed laser. This smoke moved upward rapidly to join the ESI plume for ionization (see Figure 2e for the resulting mass spectra). In contrast, we did not observe (Figure 5c) any smoke or droplets during analysis of a myoglobin spot dried from a 10-5 M solution, presumably because of the nonvolatility of myoglobin molecules. Because myoglobin ions were detectable by LIAD/ESI/MS (Figure 4e), these findings suggested that the protein molecules were desorbed during LIAD, but the size of the desorbed matter was too small to be observed by the naked eye. The photograph taken during LIAD-mediated desorption of the protein solution clearly reveals several bright trajectories leaving the sample solution (Figure 5d). These trajectories represented moving droplets induced by the action of the laser acoustic wave from the rear side of the aluminum foil. By

Figure 5. Photographs of the desorption and ionization of various samples: (a) ink was drawn on the surface using a blue marker, ESI was on, and laser was switched off; (b) ESI and laser were on dry ink; (c) ESI and laser were on a dry myoglobin spot; (d) ESI and laser were on myoglobin solution.

Figure 6. Positive-ion LIAD/ESI mass spectra of myoglobin and cytochrome c in the solid state and in aqueous solution.

carefully examining the surface of the sample solution with the naked eye, we observed numerous, but faint, fine droplets generated at the surface of the sample solution (Figure 5d). Figure 6 presents the positive-ion LIAD/ESI mass spectra of the protein solution; for comparison, the results obtained for the dried protein spot analyzed by LIAD/ESI/MS are also included. The myoglobin and cytochrome c ions detected from the dry spots were both denatured (Figure 6, parts a and c); we determined the conformational status of these ions from the charge distributions of the protein ions and the calculated molecular weights (see the insets to Figure 6). For example, the m/z values of native and denatured myoglobin ions are 17 567 and 16 951, respectively [i.e., loss of a heme molecule (m/z 616) from the native myoglobin ion]. Although the value of m/z remains the same for the native and denatured cytochrome c ions, the charge distribution of the native cytochrome c ions (from +8 to +10) was much lower and narrower than that of denatured cytochrome c (from +11 to +21). When analyzing protein solutions (liquid samples) using

Figure 7. Positive-ion LIAD/ESI mass spectra of (a) human serum, (b) human whole blood, (c) yogurt, and (d) cow’s milk in liquid condition. The insets to panels a, b, and d display the deconvoluted protein ion signals (by DataAnalysis version 3.2, Bruker Daltonics) detected in the sample.

LIAD/ESI/MS, we detected both the native and denatured myoglobin ions, but only the native cytochrome c ions (Figure 6, parts b and d). We suspect that a surrounding of water molecules protects the proteins from denaturation during desorption from the aluminum foil under the influence of the acoustic wave. Figure 7 displays the mass spectra obtained when using LIAD/ ESI/MS to analyze several biological fluids: human serum, human whole blood, yogurt, and cow’s milk. These biological fluids were Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

873

applied directly to the sample plate using a micropipette prior to LIAD/ESI/MS analysis; no other sample pretreatment was performed. The insets in Figure 7 display the deconvoluted protein ion signal detected in the biological fluids. Since these biological fluids contained mixtures of proteins and were not subjected to chromatographic or electrophoretic separation, we detected only the most prominent proteins/lipids present in the samples. For examples, we detected albumin (m/z 66 747) from the human serum, hemoglobin (m/z 15 130) from the human whole blood, lipids from the yogurt, and R-casein (m/z 23 986) as the only protein ion from the cow’s milk.

the conventional ionization methods such as MALDI, the need of sample preparation for a LIAD/ESI/MS analysis is minimal. LIAD/ESI/MS takes advantage of desorption and ionization under ambient conditions, which extends the sample conditions from solid to liquid states. This will also expand the practical applications of LIAD to the direct detection of the chemicals in biological fluids. In comparison to ESI, using LIAD/ESI for sample analysis may provide at least the following benefits, that (1) sample switching is rapid, (2) both liquid and solid samples can be analyzed, and (3) sample solution composition does not affect the ion intensity.

CONCLUSION LIAD/ESI/MS, a technique combining LIAD with ESI, allows the characterization of small organic and large biological compounds such as peptides and proteins. In comparison to

Received for review May 1, 2008. Accepted December 3, 2008.

874

Analytical Chemistry, Vol. 81, No. 3, February 1, 2009

AC800896Y