Paired Ion Electrospray Ionization for Trace Anion Analysis

July 14, 2015

Article

Using ESI-MS to perform quantitative binding analyses and determine association constants depends on the ability of the ionization process to preserve the system equilibrium.

When I started my independent academic career, I planned that a primary component of my group’s research would be the development and application of mass spectrometry (MS)-based methods to study noncovalent interactions. We were successful obtaining funding to do this and have pursued this line of research for approximately 10 years, even though in the recent years that effort has slowed a bit. We have been particularly interested in the development of flow injection–based methods to increase throughput for quantitative noncovalent binding determinations. We have developed several “dynamic titration” formats for this work (1–4). While we have applied these methods to a number of systems, the underlying question is always as follows: Does the electrospray ionization (ESI) process effectively preserve the solution equilibrium, such that the measured relative ion intensities for free unbound and complex ion forms are indicative of relative solution-phase equilibrium concentrations for these species? To perform quantitative binding determinations to determine association constants for a system by ESI-MS, this is a necessity. There are other considerations (response factors, ESI source settings, types of noncovalent forces, and so forth), but it is not my intention to expound these here.

The degree to which the ESI process alters solution equilibria is system dependent. As a very high level generalization-assuming the other miscellaneous factors above have been accounted-if a noncovalent complex present in solution is stable on the time-scale of the ESI process (100 μs–1 ms), then there is a good chance that the relative ion signals will reflect relative solution equilibrium concentrations. However, if association or dissociation kinetics are relatively fast, then the shrinking-droplet ESI process can alter equilibria. This is not good for studying noncovalent complexation in solution, but it is central to the success of a technique called paired ion electrospray ionization (PIESI).

Originally reported by Dasgupta and Armstrong in 2007 (5), PIESI involves the addition of multiply charged ion-pair reagents to solutions of anion analytes that are then passed through the ESI source. For example, a dication can be added to a sprayed solution containing monoanions that are desired to be detected. The dication is designed to be surface active (that is, one side is charged, whereas that other side is hydrophobic) when it ion pairs with the monoanion. The resulting formed complex has an overall positive charge. In ESI, it is fairly well known that positive ionization mode is more robust and less noisy than negative ionization mode. In negative ionization mode, there is a greater chance of discharge in the source that can lead to unstable ion signals (6). Additionally, many anions have low m/z values, and in the low mass region, many mass analyzers are subject to greater noise. When a cation reagent complexes with the anion target, the resulting complex is moved to a high m/z range, which is better for MS detection. What is interesting is that these dications and anions only associate very weakly in solution. The ESI process is primarily responsible for the enhancement in complexation observed in the mass spectra.

We worked together with the Armstrong group to study the mechanism associated with ultratrace detection of anions using PIESI (7). In fact, PIESI has been shown to be much more sensitive than many other anion detection strategies. We studied the binding in solutions using nuclear magnetic resonance (NMR)-based titration techniques. We then investigated binding using our ESI-MS–based dynamic titration method. Measured association constants were 3–4 orders of magnitude higher using the MS method compared to NMR. This is an excellent example of how the shrinking-droplet ESI process can enhance the abundance of complex ions. Pairing between the dication agent and the anion is essentially diffusion limited, meaning that complexation is very fast. Thus, the moment that they come into contact in the ESI droplet, a complex is formed. Importantly, this formed complex is highly surface active, as shown by surface tension measurements. Thus, the formed complex moves very quickly to the surface of the droplet, where it can be efficiently transferred into the gas phase. This makes PIESI extremely useful for liberating hydrophilic anions into the gas phase for mass analysis and very sensitive determination. The complexes can be measured directly, or strategies such as selected reaction monitoring (SRM) can be used to enhance sensitivity. Incidentally, we have previously reported other peptide-based binding systems that also exhibit a higher degree of complex formation in ESI mass spectra compared to solution binding measurements (8)-it has always been a critical point to compare ESI-MS–determined binding constants to those determined by traditional solution binding techniques to check accuracy of the former.

PIESI benefits from the shrinking-droplet process and the high surface activity of formed complexes. A very large list of organic and inorganic anions has been reported determined using PIESI-MS (5,9,10). Trications and tetracation reagents can be used to measured dianaions and triananions, respectively. A variety of different PIESI reagents have been developed and demonstrated for use. Many have been generated out of research efforts focused on the generation of new ionic liquids. The cationic reagents for PIESI are prepared with fluoride counterions, since virtually all other anions pair more strongly with the reagent during ESI. Generally, cationic reagents with flexible structures perform best, and several of the best performers are even available commercially for purchase. It is possible to couple anion determination with ion-exchange chromatographic separation of ions for improved speciation in complex samples. While primarily ion-trap technology has been used, the use of triple-quadrupole MS is ideally suited for SRM detection, since many of the cation–anion complexes are easily dissociated to create precursor–product ion detection pairs. We have recently demonstrated the use of PIESI and triple-quadrupole MS for the determination of anionic surfactants (11). Even though surfactants are inherently surface active, which would generally lead to good ESI sensitivity, the PIESI method was still shown to provide equivalent or better sensitivity than most of the methods reported in the literature.

Overall, I believe that PIESI should be considered when sensitive detection of anions is necessary. While we have been more attuned to systems where noncovalent complexes are well preserved, so that binding affinities can be measured, these systems for PIESI analysis show a novel advantage of the shrinking-droplet ESI process. More work is needed to demonstrate the viability of the technique for ultratrace determinations in highly complex biological systems.

References

(1) K.A. Schug, C.A. Serrano, and P. FryÄák, Mass Spectrom. Rev.29, 806–829 (2010).

(2) P. FryÄák and K.A. Schug, Anal. Chem.80, 1385–1393 (2008).

(3) P. FryÄák and K.A. Schug, Anal. Chem.79, 5407–5413 (2007).

(4) I.C. Santos, V.B. Waybright, H. Fan, S. Ramirez, R.B.R. Mesquita, A.O.S.S. Rangel, P. Frycak, and K.A. Schug, J. Am. Soc. Mass Spectrom.26, 1204–1212 (2015).

(5) R.J. Soukup-Hein, J.W. Remsburg, P.K. Dasgupta, and D.W. Armstrong, Anal. Chem.79, 7346–7352 (2007).

(6) G. Wang and R.B. Cole, J. Am. Soc. Mass Spectrom.4, 1050–1058 (1993).

(7) Z.S. Breitbach, E. Wanigasekara, E. Dodbiba, K.A. Schug, and D.W. Armstrong, Anal. Chem.82, 9066–9073 (2010).

(8) M.A. Raji, P. Frycak, M. Beall, M. Sakrout, J.-M. Ahn, Y. Bao, D.W. Armstrong, and K.A. Schug, Int. J. Mass Spectrom.262, 232–240 (2007).

(9) R.J. Soukup-Hein, J.W. Remsburg, Z.S. Breitbach, P.S. Sharma, T. Payagala, E. Wanigasekara, J. Huang, and D.W.Armstrong, Anal. Chem.80, 2612–2616 (2008).

(10) Z.S. Breitbach, M.M. Warnke, E. Wanigasekara, X. Zhang, and D.W. Armstrong, Anal. Chem.80, 8828–8834 (2008).

(11) I.C. Santos, H. Guo, R.B.R. Mesquita, A.O.S.S. Rangel, D.W. Armstrong, and K.A. Schug, Talanta143, 320–327 (2015).

Previous blog entries from Kevin Schug:

The LCGC Blog: Forensics, Lawyers, and Method Validation-Surprising Knowledge Gaps

The LCGC Blog: A Disconnect Between Science Research and Science Education Research

The LCGC Blog: A New Consortium of Researchers for Environmental Analysis and Remediation

The LCGC Blog: How to Get the Most Out of Your First Conference Experience

The LCGC Blog: Five Steps in the Evolution of an Instrumental Analysis Course for Enhanced Student Preparation

The LCGC Blog: Insights on Increased Efficiency for Superficially Porous Particles Among Other Things

The LCGC Blog: Responsible Unconventional Oil and Gas Exploration in Colombia

The LCGC Blog: Intact Protein Separations: Some Education is Missing

The LCGC Blog: Evaluating the Impact of Unconventional Oil and Gas Extraction on Groundwater