Electrospray: from ions in solution to ions in the gas phase, what we know now.

There is an advantage for users of electrospray and nanospray mass spectrometry to have an understanding of the processes involved in the conversion of the ions present in the solution to ions in the gas phase. The following processes are considered: Creation of charge droplets at the capillary tip; Electrical potentials required and possibility of gas discharges; Evolution of charged droplets, due to solvent evaporation and Coulomb explosions, to very small droplets that are the precursors of the gas phase ions; Production of gas phase ions from these droplets via the Ion Evaporation and Charge residue models; Analytical uses of ESIMS of small ions, qualitative and quantitative analysis; Effects of the ESI mechanism on the analysis of proteins and protein complexes; Determination of stability constants of protein complexes; Role of additives such as ammonium acetate on the observed mass spectra.

Gas phase ion thermochemistry of tetraaza complexes such as cyclamM2+ with H2O, CH3OH, NH3, and other ligands, where M2+ = Mn2+, Ni2+, Cu2+, and Zn2+.

Tetraaza complexes with M(2+) were produced in the gas phase by Electrospray (ESI) of solutions containing salts of M(2+)dinitrates and a tetraaza compound such as cyclam. The complex CyclM(2+) formed in solution and transferred to the gas phase via ESI was introduced into a reaction chamber containing known partial pressures of a ligand L. Equilibria between CyclM(2+) and L establish CyclML(n)(2+) = CyclML(n-1)(2+) + L and the equilibrium constants K(n,n-1) are determined with a mass spectrometer. Determinations at different temperatures lead to not only the DeltaG(0)(n,n-1) values but also the DeltaH(0)(n,n-1) and DeltaS(0)(n,n-1) values. Data for n = 1, 2, and 3 were obtained for L = H(2)O and CH(3)OH. The DeltaG(0)(1,0), DeltaH(0)(1,0) as well as DeltaG(0)(2,1), DeltaH(0)(2,1) values, when M(2+) = Mn(2+) and Zn(2+), were larger than those for Ni(2+) and Cu(2+). The ligand field theory and the Irvine-Williams series predict a reverse order, i.e., stronger bonding with Ni(2+) and Cu(2+) for simple ligand reactions with M(2+). An examination of the differences of the reactions in solution and gas phase provides a rationale for the observed reverse order for the CyclM(2+) + L reactions. Differences between gas phase and solution are found also when M(2+) = Cu(2+), but the tetraaza macrocycle is changed from, 12-ane to 14-ane to 15-ane. The strongest bonding in solution is with the 14-ane while in the gas phase it is with the 15-ane. Bond free energies, DeltaG(0)(1,0), for CyclCu(2+) with L = H(2)O, CH(3)OH, NH(3), C(2)H(5)OH, C(3)H(7)OH, (C(2)H(5))(2)O, and CH(3)COCH(3), are found to increase in the above order. The order and magnitude of the DeltaG(0)(1,0) values is close to DeltaG(0)(1,0) values observed with potassium K(+) and the same ligands. These results show that the cyclam in CyclCu(2+) leads to an extensive shielding of the +2 charge of Cu(2+). Ligands with gas phase basicities that are relatively high, lead to deprotonation of CyclM(2+). The deprotonation varies with the nature of M(2+) and provides information on the extent of electron transfer from the N atoms of the cyclam, to the M(2+) ions.

Ion-ion and ion-molecule reactions at the surface of proteins produced by nanospray. Information on the number of acidic residues and control of the number of ionized acidic and basic residues.

Mass Spectra of charge states of folded proteins were obtained with nanospray and aqueous solution containing 20 microM the protein (ubiquitin, cytochrome c, lysozyme) and one of the NaA salts NaCl, NaI, NaAc (acetate) (1-10 mM). At very low collision activated decomposition (CAD), the mass spectra of a protein with charge z exhibited a replacement of zH+ with zNa+ and also multiple adducts of NaA. Higher CAD converts the NaA adduct peaks to Na minus H peaks. These must be due to loss of HA where the H was provided by the protein. The degree of HA loss with increasing CAD followed the order I < Cl < Ac. Significantly, the intensity of the ions with n (Na minus H) adducts showed a downward break past an n(MAX) which is equal to the number of acidic residues of the protein plus the charge of the protein. All the observations could be rationalized within the framework of the electrospray mechanism and the charge residue model, which predict that due to extensive evaporation of solvent, the solutes will reach very high concentrations in the final charged droplets. At such high concentrations, positive ions such as Na+, NH4+ form ion pairs with ionized acidic residues and the negative A- form ion pairs with ionized basic residues of the protein. Adducts of Na+, and NaA to backbone amide groups occur also. This reaction mechanism fits all the experimental observations and provides predictions that the number of acidic and basic groups at the surface of the gaseous protein that remain ionized can be controlled by the absence or presence of additives to the solution.

Sequential hydration energies of the sulfate ion, from determinations of the equilibrium constants for the gas-phase reactions: SO4(H2O)(n)2- = SO4(H2O)(n-1)2- + H2O.

Sequential hydration energies of SO4(H2O)(n)2- were obtained from determinations of the equilibrium constants of the following reactions: SO4(H2O)(n)2- = SO4(H2O)(n-1)2- + H2O. The SO4(2-) ions were produced by electrospray and the equilibrium constants Kn,n-1 were determined with a reaction chamber attached to a mass spectrometer. Determinations of Kn,n-1 at different temperatures were used to obtain DeltaG0n,n-1, DeltaH0 n,n-1, and DeltaS0n,n-1 for n = 7 to 19. Interference of the charge separation reaction SO4(H2O)(n)2- = HSO4(H2O)(n-k)- + OH(H2O)(k-1)- at higher temperatures prevented determinations for n < 7. The DeltaS0n,n-1 values obtained are unusually low and this indicates very loose, disordered structures for the n > or = 7 hydrates. The DeltaH0n,n-1 values are compared with theoretical values DeltaEn,n-1, obtained by Wang, Nicholas, and Wang. Rate constant determinations of the dissociation reactions n,n - 1, obtained with the BIRD method by Wong and Williams, showed relatively lower rates for n = 6 and 12, which indicate that these hydrates are more stable. No discontinuities of the DeltaG0n,n-1 values indicating an unusually stable n = 12 hydrate were observed in the present work. Rate constants evaluated from the DeltaG0n,n-1 results also fail to indicate a lower rate for n = 12. An analysis of the conditions used in the two types of experiments indicates that the different results reflect the different energy distributions expected at the dissociation threshold. Higher internal energies prevail in the equilibrium measurements and allow the participation of more disordered transition states in the reaction.

Features of the ESI mechanism that affect the observation of multiply charged noncovalent protein complexes and the determination of the association constant by the titration method.

Several factors, attributable to the ESIMS mechanism, that can affect the assumptions of the titration method are examined: (1) The assumption that the concentrations in solution of the protein P, the ligand L, and the complex PL are proportional to the respective ion intensities observed with ESIMS, is examined with experiments in which ion intensities of two non-interacting proteins are compared with the respective concentrations. The intensities are found to be approximately proportional to the concentrations. The proportionality factors are found to increase as the mass of the protein is decreased. Very small proteins have much higher intensities. The results suggest that it is preferable to use only the intensity ratio of PL and P, whose masses are very close to each other when L is small, to determine the association constant KA in solution. (2) From the charge residue model (CRM) one expects that the solution will experience a very large increase of concentration due to evaporation of the precursor droplets, before the proteins P and PL are produced in the gas phase. This can shift the equilibrium in the droplets: P + L = PL, towards PL. Analysis of the droplet evaporation history shows that such a shift is not likely, because the time of droplet evolution is very short, only several micros, and the equilibrium relaxation time is much longer. (3) The droplet history shows that unreacted P and L can be often present together in the same droplet. On complete evaporation of such droplets L will land on P leading to PL and this effect will lead to values of KA that are too high. However, it is argued that mostly accidental, weakly bonded, complexes will form and these will dissociate in the clean up stages (heated transfer capillary and CAD region). Thus only very small errors are expected due to this cause. (4) Some PL complexes may have bonding that is too weak in the gas phase even though they have KA values in solution that predict high solution PL yields. In this case the PL complexes may decompose in the clean up stages and not be observed with sufficient intensity in the mass spectrum. This will lead to KA values that are too low. The effect is expected for complexes that involve significant hydrophobic interaction that leads to high stability of the complex in solution but low stability in the gas phase. The titration method is not suited for such systems.

Hydration energies in the gas phase of select (MX)mM+ ions, where M+ = Na+, K+, Rb+, Cs+, NH4+ and X- = F-, Cl-, Br-, I-, NO2-, NO3-. Observed magic numbers of (MX)mM+ ions and their possible significance.

The sequential hydration energies and entropies with up to four water molecules were obtained for MXM(+) = NaFNa(+), NaClNa(+), NaBrNa(+), NaINa(+), NaNO(2)Na(+), NaNO(3)Na(+), KFK(+), KBrK(+), KIK(+), RbIRb(+), CsICs(+), NH(4)BrNH(4)(+), and NH(4)INH(4)(+) from the hydration equilibria in the gas phase with a reaction chamber attached to a mass spectrometer. The MXM(+) ions as well as (MX)(m)M(+) and higher charged ions such as (MX)(m)M(2)(2+) were obtained with electrospray. The observed trends of the hydration energies of MXM(+) with changing positive ion M(+) or the negative ion X(-) could be rationalized on the basis of simple electrostatics. The most important contribution to the (MXM-OH(2))(+) bond is the interaction of the permanent and induced dipole of water with the positive charge of the nearest-neighbor M(+) ion. The repulsion due to the water dipole and the more distant X(-) has a much smaller effect. Therefore, the bonding in (MXM-OH(2))(+) for constant M and different X ions changes very little. Similarly, for constant X and different M, the bonding follows the hydration energy trends observed for the naked M(+) ions. The sequential hydration bond energies for MXM(H(2)O)(n)(+) decrease with n in pairs, where for n = 1 and n = 2 the values are almost equal, followed by a drop in the values for n = 3 and n = 4, that again are almost equal. The hydration energies of (MX)(m)M(+) decrease with m. The mass spectra with NaCl, obtained with electrospray and observed in the absence of water vapor, show peaks of unusually high intensities (magic numbers) at m = 4, 13, and 22. Experiments with variable electrical potentials in the mass spectrometer interface showed that some but not all of the ion intensity differentiation leading to magic numbers is due to collision-induced decomposition of higher mass M(MX)(m)(+) and M(2)(MX)(m)(2+) ions in the interface. However, considerable magic character is retained in the absence of excitation. This result indicates that the magic ions are present also in the saturated solution of the droplets produced by electrospray and are thus representative of particularly stable nanocrystals in the saturated solution. Hydration equilibrium determinations in the gas phase demonstrated weaker hydration of the magic ion (NaCl)(4)Na(+).

Prediction of the charge states of folded proteins in electrospray ionization.

Earlier work from this laboratory dealt with the observation that the charge states of non-denatured proteins can be decreased by use of buffer salts in which the gas-phase basicity of conjugate base B, GB(B), of the buffer cations is high. A theoretical model was developed and applied to several small proteins. The predictions of the charge states were found to be in good agreement with those observed experimentally. Because the computational model is based on the charge residue model (CRM), the observed agreement lends support for the CRM. In the present work, the same model is applied to recent data by Catalina et al. who showed that very large charge reductions are achieved with very high GB(B) proton sponges. Their data included lysozyme but also the very much larger proteins, p-hydroxybenzoate hydroxylase (PHBH), 90 kDa and glutamate synthase (GLTS), 166 kDA. The present work examines the performance of the model for the much stronger bases and the very much larger proteins. It is found that the predictions of the charge states agree well for the small protein lysozyme but somewhat less well with the experimental results for PHBH and GLTS. The causes for the lack of good agreement with the large proteins are examined.

Effect of buffer cations and of H3O+ on the charge states of native proteins. Significance to determinations of stability constants of protein complexes.

The progressive reduction of charge in charge states of non-denatured proteins (lysozyme, ubiquitin, and cytochrome c), observed with nanospray in the positive ion mode, when the buffer salt ammonium acetate is replaced by ethylammonium acetates (EtNH(3)Ac, Et(2)NH(2)Ac and Et(3)NHAc) is rationalized on the basis of the charge residue model (CRM). The charge states of the multiply protonated protein are shown to be controlled by the increasing gas-phase basicities, GB(B), of the bases(B) NH(3), EtNH(2), Et(2)NH and Et(3)N. Charge states derived from evaluated apparent gas-phase basicities GB(app) of the basic side-chains of the protein and the known GB(B) of the above bases are found to be in agreement with the experimentally observed charge states. This is a requirement of the CRM, because in this model the small positive ions (the buffer cations in the present case) at the surface of the electrospray droplets are the excess ions that provide the charge of the final small droplet that contains the protein molecule and on evaporation of the solvent transfer the charge to the protein. The observed charge states in the absence of buffer salts, i.e. pure water, are attributed to excess H(3)O(+) ions produced by the electrolysis process that attends electrospray. A proposed extended mechanism provides predictions of factors that determine the sensitivity for detection of the multiply protonated proteins. Consideration of restraints imposed by the CRM lead to some simple predictions for conditions that should be present to obtain accurate determinations by electrospray and nanospray of stability constants for the protein-complex equilibrium in aqueous solution.

Charged states of proteins. Reactions of doubly protonated alkyldiamines with NH(3): solvation or deprotonation. Extension of two proton cases to multiply protonated globular proteins observed in the gas phase.

The apparent gas-phase basicities (GB(app)'s) of basic sites in multiply protonated molecules, such as proteins, can be approximately predicted. An approach used by Williams and co-workers was to develop an equation for a diprotonated system, NH(3)(CH(2))(7)NH(3)(2+), and then extend it with a summation of pairwise interactions to multiply protonated systems. Experimental determinations of the rates of deprotonation of NH(3)(CH(2))(7)NH(3)(2+) by a variety of bases B, in the present work, showed that GB(app) = GB(NH(3)) = 196 kcal/mol. This result is supported also by determinations of the equilibria: NH(3)(CH(2))(p)NH(3)(2+) + NH(3) = NH(3)(CH(2))(p)NH(3) x NH(3)(2+), for p = 7, 8, 10, 12. The described experimental GB(app) is 14 kcal/mol higher than the value predicted by the equation used by Williams and co-workers but in agreement with an ab initio result by Gronert. Equations based on electrostatics are developed for the two proton and multiproton systems which allow the evaluation of GB(app) of the basic sites on proteins. These are applied for the evaluation of GB(app) of the basic sites and of N(SB), the maximum number of protons that the nondenatured proteins, carbonic anhydrase (CAII), cytochrome c (CYC), and pepsin, can hold. The N(SB) values are compared with the observed charges, Z(obs)'s, when the nondenatured proteins are produced by electrospray and found in agreement with the proposal by de la Mora that Z(obs) is determined by the number of charges provided by the droplet that contains the protein, according to the charge residue model (CRM). The GB(app) values of proteins have many other applications. They can be compared with experimental measurements and are also needed for the understanding of the thermal denaturing of charged proteins and the thermal dissociation of charged protein complexes.