Anal. Chem. 1999, 71, 3785-3792
Evaluation of Electrospray Mass Spectrometry as a Technique for Quantitative Analysis of Kinetically Labile Solution Species Hongjun Wang and George R. Agnes*
Department of Chemistry, Simon Fraser University, Burnaby, BC, Canada V5A 1S6
The extent that a kinetically labile equilibrium reaction was perturbed by passage through the electrospray ion source has been measured. The reaction studied was strontium ion chelation by EDTA (Sr2+ + Y4- h SrY2-) in 100% aqueous solutions. The forward reaction rate is very fast (109 M-1 s-1) and the reverse rate is very slow (100 s-1) relative to the time scale of the ES process (∼10-2 s). The SrY2- detected with the mass spectrometer were expected to be representative of thermodynamic equilibrium within the droplets, but the position of the equilibrium shifted to the right relative to the solution equilibrium position. Given the current status of understanding of the ion generation process in the electrospray ion source, the degree that the [Sr2+] changed due to passage through this ion source was smaller than expected, which is fortuitous with respect to the quantitation of such species. The pH of each calibration set determined the fraction of strontium that was uncomplexed in solution. The equilibrium shift induced by passage through the ion source was constant for solutions at constant pH but differed for solutions at different pH. Decreasing the solution pH generated smaller equilibrium shifts as measured by the change in the [Sr2+]. In solutions with free Sr2+ and excess EDTA at equilibrium, the free [Sr2+] decreased by ∼100 and ∼10% in solutions at pH 5.85 and 4.98, respectively. Quantitation of kinetically labile species in complex, multicomponent systems will be straightforward with ES-MS, provided all species involved in the equilibria can be identified and monitored. The electrospray ion source transfers preformed ions in solution to the gas phase with no fragmentation,1,2 though there is ample evidence that neutrals can be ionized electrolytically.3 Consequently, the electrospray ion source (ES) is often viewed as a mass spectrometric tool through which solution chemistry can be probed. However, to what degree a mass spectrum produced using an ES source reflects the solution-phase chemistry is a frequently asked question. The uncertainty in correlating (1) Kebarle, P.; Ho, Y. In Electrospray Ionization Mass Spectrometry: Fundamentals, Instrumentation, and Applications; Cole, R. B., Ed.; John Wiley and Sons: New York, 1997; pp 3-63. (2) Kebarle, P.; Tang, L. Anal. Chem. 1993, 65, 972A-986A. (3) Van Berkel, G. J. In Electrospray Ionization Mass Spectrometry; Fundamentals, Instrumentation, and Applications; Cole, R. B., Ed.; John Wiley and Sons: New York, 1997; pp 65-105. 10.1021/ac9813742 CCC: $18.00 Published on Web 08/03/1999
© 1999 American Chemical Society
solution versus gas-phase ion abundance is related directly to the fact that though there is general agreement regarding the major steps in the ES process, the details of the ion transfer from solution to the gas phase remain unresolved. Because we are interested in applying ES-MS to the study of kinetically labile solution species,4 it was necessary to determine whether a quantitative relationship between solution-phase ionic abundance and gasphase ionic abundance could be obtained for labile solution species. To query this problem, we chose to study the wellcharacterized solution equilibrium reaction between an alkaline earth metal ion (Sr2+) and EDTA. In this study, the potential for gas-phase reactions to skew the solution equilibrium is negated because all reactants and products are nonvolatile, ionic species. Interest in the quantitation of kinetically labile solution species stems from the fact that biological function is achieved with said interactions. With respect to metal ions, enzymatic activity in living organisms depends on regulation of the essential and nonessential metals. Insufficient metal ion concentration inhibits biological reactions, whereas excess concentrations cause unwanted side reactions that are toxic to the organism.5 Within these systems, the utilization of the metal ion in all aspects of the uptake, transport, and release is governed by dynamic processes.6 It is now recognized that organized structures in environmental matrixes also function similarly to living organisms, in that the systems must be interpreted with kinetic models.4 Due to the complexity of natural systems and that the number of techniques capable of measuring labile species is limited, detailed studies of such systems have been limited.4,7-12 Because of the merits of the electrospray source, mass spectrometry could prove well suited for quantitative investigations of kinetically labile solution species, provided equilibria positions are not extensively perturbed by the ES process. The objective of this work was to measure the shift induced on a labile equilibrium by passage through an electrospray ion source. (4) Langford, C. H.; Cook, R. L. Anal. Chim. Acta 1995, 120, 591-602. (5) McRae, D. E. Nature Struct. Biol. 1998, 5, 8-10. (6) Gray-Owen, S. D.; Schryvers, A. B. Trends Microbiol. 1996, 4, 185-191. (7) Langford, C. H.; Gutzman, D. W. Anal. Chim. Acta 1992, 256, 183-201. (8) Gutzman, D. W.; Langford, C. H. Environ. Sci. Technol. 1993, 27, 13881393. (9) Millward, G. E. Anal. Chim. Acta 1995, 120, 600-614. (10) Daniele, P. G.; Prenesti, E.; Aigotti, R.; Ostacoli, G. J. Inorg. Biochem. 1995, 58, 139-146. (11) Kramer, J. R.; Gleed, J.; Gracey, K. Anal. Chim. Acta 1994 284, 599-604. (12) Miller, L. A.; Bruland, K. W. Anal. Chim. Acta 1994, 1994, 573-586.
Analytical Chemistry, Vol. 71, No. 17, September 1, 1999 3785
Briefly, the major steps of the ion-transfer process in the ES source involve (1) generation of the droplets with a net negative or positive charge by electrical atomization of a liquid jet, (2) evaporation of solvent from the charged droplets, (3) Coulomb fission, a process to release excess charge from a desolvated droplet, and (4) production of individual ions that are likely solvated. A fifth step, related to mass spectrometric analysis, is the sampling of the gas at atmospheric pressure into a mass spectrometer. These factors that lead to ion generation suggest that the mass spectrum of ions sampled from an ES source could reflect the aqueous solution chemistry within the droplets.13 The problem with respect to quantitation is that it is not possible to compare ion intensities in the electrospray mass spectrum with the solution chemistry of the droplet during the critical moment of ion formation.14-17 Rather, it is only possible to quantitatively compare the distribution of ions in the bulk solution prior to being introduced to the ES source, with ion signal intensities in the mass spectrum. On the basis of the presumption that solvent evaporation is a necessary step in the ion generation process, a kinetically labile equilibrium such as Sr2+ + Y4- h SrY2should be shifted far to the right, relative to the original solution conditions. In addition, these experiments were performed in the negative ion mode and the electrolytic increases in the solution pH will also cause this chemical equilibrium to shift to the right.3 In related work, there have been several studies that have compared gas-phase to solution-phase ion abundance for small molecular ions present in thermodynamic equilibrium in solution. Leize et al. studied the competition of Na+, K+, Rb+, and Cs+ for cryptate 222 and the crown ether 18C6.18 They found that ES-MS ion abundance was in good agreement to the calculated solutionphase ion populations. On the other hand, Gatlin found that the dissociation equilibriums of Fe2+(bpy)3 and Ni2+(bpy)3 complexes (bpy ) 2,2′-bipyridyl), as measured by ES-MS, were much greater than expected on the basis of solution-phase equilibrium calculations.19 They attributed this increase in degree of dissociation of the complexes to an increase of droplet acidity caused by the ES source. The Fe2+(bpy)3 and Ni2+(bpy)3 complexes in solution at pH 6.7 gave rise to spectra, that as measured by the degree of dissociation, were compatible with a droplet acidity of pH 2.63.3. These authors concluded that this 103-104-fold increase of acidity over the neutral bulk solution was localized on a surface layer several nanometers in thickness and that the residence time of the complex in this layer is sufficiently long for it to dissociate, prior to becoming a gas-phase ion. In addition, numerous cautionary reports continue to appear in the literature regarding the correlation, or lack thereof, between solution-phase charge distributions of proteins, and that observed in the mass spectrum,20-24 (13) Chowdhury, S. K.; Katta, V.; Chait, B. T. J. Am. Chem. Soc. 1990, 112, 9012-9013. (14) Smith, R. D.; Loo, J. A.; Edmonds, C. G.; Barinaga, C. J.; Udseth, H. R. Anal. Chem. 1990, 62, 882-899. (15) Van Berkel, G. J.; Zhou, F.; Aronson, J. T. Int. J. Mass Spectrom. Ion Processes 1997, 162, 55-67. (16) Thomson, B. A.; Iribarne, J. V. J. Chem. Phys. 1979, 71, 4451-4463. (17) Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Mass Spectrom. Rev. 1990, 9, 37-70. (18) Leize, E.; Jaffrezic, A.; Dorsseler, A. V. J. Mass Spectrom. 1996, 31, 537544. (19) Gatlin, C. L.; Turecek, F. Anal. Chem. 1994, 66, 712-718. (20) Meunier, C.; Jamin, M.; Pauw, E. D. Rapid Commun. Mass Spectrom. 1998, 12, 239-245.
3786 Analytical Chemistry, Vol. 71, No. 17, September 1, 1999
Figure 1. Diagram depicting the pneumatically assisted ES source and the quadrupole mass spectrometer. The tip of the ES capillary was positioned 2.5 cm from the counter electrode and 0.5 cm offaxis. Flow rates of the gases used in the ion source: nebulizer gas, 700 mL/min; curtain gas, 800 mL/min; and discharge suppressing gas, 700 mL/min.
and how instrumental factors, such as high-pressure ion sampling influence the appearance of mass spectra.25 Because of the considerable questions that presently exist between ionic solution-phase equilibria and ionic gas-phase abundance as measured by ESMS, we choose to examine a wellcharacterized kinetically labile chemical system for the purpose of measuring the change in the equilibrium species distribution caused by this ion source. The complexation of Sr(II) by EDTA proceeds at near diffusion-limited rates in solution. As such, the forward reaction can respond rapidly to maintain a thermodynamic abundance of SrY2- in response to electrolytic pH changes within the capillary and to the system volume decrease due to evaporation of solvent from the droplets. Methodology is described by which linear calibration curves for labile solution species can be obtained with ES-MS. Through comparison of the ion abundance in the ES-generated mass spectrum to the calculated solution-phase equilibrium species concentration, the degree to which a labile equilibrium is shifted by the ES process has been quantitated. EXPERIMENTAL SECTION Mass Spectrometry. An in-house-constructed pneumatically assisted ES source was used.26 The dimensions of the inner and outer capillaries were 0.1 mm i.d. × 0.21 mm o.d. and 0.28 mm i.d. × 0.51 mm o.d., respectively. The inner capillary protruded from the larger sized outer capillary by 0.3 mm. The potential applied to the ES source was -3200 V (Stanford Research Systems, model PS350). Sample solution was delivered to the ES source at a rate of 5 µL/min. by a syringe pump (Harvard Apparatus, model 22). Three separate gas streams, all at ambient temperature (293 K), were necessary to enable studies with 100% aqueous solutions. The curtain and nebulizing gas flows of N2 are indicated in Figure 1. The flow of O2 was required to minimize electrical discharge that otherwise precluded analysis of species of
