The Section on Metabolic Analysis and Mass Spectrometry carries out research in areas of chemistry and biochemistry in which mass spectrometry is the primary analytical tool. Current research focuses on two separate areas: the energetics of hydration and protein characterization. The goal of this fundamental physicochemical research is to provide missing, critical information needed for the biophysical characterization of macromolecular and membrane interactions and a characterization of the energy required for noncovalent bonds between molecules or ions in solution. Much of the current activity in theoretical studies of molecular configuration is utterly dependent on the values assigned to solute/solvent or peptide/metal bonds, yet reliable quantities often do not exist. By adapting mass spectrometric concepts on cluster formation in systems of small ions and molecules, we are investigating the energetics of the interactions of some model solutes, as well as of amino acids and small model peptides, with water.

Hydration Thermodynamics
Vieira, Gilligan, Yergey
In the first area, we are determining the energy required to form and break noncovalent bonds between molecules and ions in solution. The particular bonds of interest are those involved in the interplay with water of solutes, including amino acids, small peptides, and lipids. The objective of our thermodynamic investigations is to determine solvation enthalpies and free energies of selected biologically interesting ions. Knowledge of the energetic requirements of these noncovalent bonds, particularly those involving water and biologically significant molecules, is fundamental to understanding molecular interactions and the changes in conformations that are integral to them.

Experimentally, we use the tools of equilibrium gas phase ion-molecule chemistry applied in a modifed electrospray ionization (ESI) source of a single quadrupole mass filter. In this work, we determine the thermodynamic quantities deltaH(std)298, deltaS(std)298, and deltaG(std)298 by using the approach of equilibrium ion-molecule reaction chemistry. We calculated hydration thermodynamics values from equilibrium constants measured over a temperature range of 0� to136�C at ion source and water partial pressures ranging between zero and 100 mtorr. We made equilibrium ion intensity measurements for at least four hydration states, i.e., zero through three water molecules associated with a core ion, at each of at least 60 combinations of water partial pressures and temperatures covering the ranges of experimental variables. The initial goal of our work is to determine the enthalpy of solvation of a series of alkylammonium ions, CnH(2n+1)NH3+. While the ions are clearly a model system, they offer the possibility of providing insights into the understanding of the relationship between hydrocarbon chain length and solvation as well as the hydration of lipids and membranes. Previous researchers determined thermodynamic parameters for about 25 percent of the hydrations we have studied, and results from our measurements are in close agreement with the published data. In the past year, we completed our measurements in the model system.

We have begun extending our work to biologically more significant systems. Observations by several investigators have shown different effects of 1,2-(OH)2-propane and 1,3-(OH)2-propane on membrane fusion and collagen self-assembly. It has been hypothesized that the differential effects might be attributed to differences in the organization of water around the two molecules. Our equilibrium ion molecule studies of these two simple diols points to substantial differences in their hydration thermodynamics. The step-wise addition of water to protonated 1,2-(OH)2-propane shows a trend of diminishing exothermicity for each addition. While we were unable to obtain a direct measurement for the addition of the first water, we were able to estimate an upper limit on the exothermicity for this process based on the water partial pressure and temperature. We conclude that the first hydration step is energetically very favorable. The decreasing trend in energetics seen for 1,2-(OH)2-propane was observed for the addition of the first two water molecules to 1,3-(OH)2-propane but was not maintained for the addition of the third water molecule. The addition of the third water to the complex was determined to be energetically more favorable than the addition of the second and furthermore was found to have a substantial decrease in the entropy for the process. These two observations led to our conclusion that the 1,3-(OH)2-propane trihydrate is an energetic and entropic favorable state. The existence of such a favorable state of hydration is consistent with the 1,3-diol incorporating into and disrupting otherwise stable biomolecular structures.

Protein Characterization
Backlund, Li, Gilligan, Vieira, Yergey
In the second area of interest, we conduct research on the mass spectrometric characterization of proteins. We carry out this work collaboratively with groups in NICHD as our first priority but also conduct independent investigations of mass spectrometric protein characterization. A major aspect of our work is the identification of proteins isolated in biochemical investigations.
In terms of the identification of unknown proteins, we use mass spectrometric data to query genomic databases to ask whether any of the protein sequences present in the database have expected proteolytic cleavage products with theoretical masses that match the empirically determined masses of the peptides generated from an unknown sample. We use both Matrix Assisted Laser Desorption Ionization (MALDI) with Time-of-Flight (TOF) mass analysis, and liquid chromatography followed by electrospray ionization with mass analysis in an instrument capable of using fragmentation reactions to generate peptide sequences, i.e., MS/MS. With the latter method, we can identify proteins in mixtures at levels of about 100 fmole when they are injected into packed capillary reverse phase columns operated at 400 nL/min flow rates. With these approaches, we are confident that, given a minimum of 100 fmole to be applied to the LC column from a gel band, we can make a positive identification for a protein that is described in a database.

Two principal areas of development will improve our protein characterization capabilities. First, we have begun addressing the question of providing sequence information on proteins that are not described in databases because of either gaps or errors in the database. We are taking the approach termed de novo sequencing of peptides; it requires detailed interpretation of individual mass spectra of peptides that have been subjected to fragmentation and mass analysis of the fragments. We are using a recently developed tandem TOF mass spectrometer in conjunction with interpretation software written by this section. To date, we have been reasonably successful in applying the approach to the sequencing of proteins isolated from sea urchin cortical vesicles (CV). Specifically, we have shown that components of several bands isolated from the CV of sea urchins have sequences corresponding to no known proteins yet demonstrate homologies to other proteins that may provide insights into fusion processes.

The second area of development involves the use of surface plasmon resonance detectors, as implemented by Biacore instruments, to isolate specific proteins. We have developed a model system in which we have demonstrated the isolation of 30 fmole of hen egg lysozyme in the presence of 30 pmoles of myoglobin. The isolated proteins are then recovered in small volumes so that concentrations are appropriate for mass spectrometric analysis. MALDI analysis of the recovered proteins shows strong signals from the isolated material and zero response from other proteins present.