Assessment of purity and screening of peptide libraries by nested ion mobility-TOFMS: identification of RNase S-protein binders.

Combinatorial peptide synthesis in combination with affinity selection and high-resolution ion mobility/time-of-flight mass spectrometry (IM/TOFMS) analysis has been used to investigate the binding of a series of 96 related eight-residue peptides (with the general sequence NH2-GX1X2FX3X4X5G-CO2H, where X1 = L, F, V, Y; X2 = N, F; X3 = E, V, T; X4 = V, L; X5 = V, L) to the ribonuclease S protein. A key advantage of this strategy is that the IM/ TOFMS approach allows the relative abundances of individual library components (including numerous sequence and structural isomers) to be characterized before and after screening. The relative binding interactions of different sequences are assessed by comparing IM/TOFMS data for those components that pass through the column (as well as those that bind) to data for the library prior to screening. The high-affinity sequences that are found in this study are compared with those selected from much larger combinatorial libraries. The results suggest that many expected sequences in the large libraries may be missing (e.g., due to issues such as failure of specific steps during the synthesis or differences in solubility). Comparison of the binding sequences obtained in these studies and those reported previously indicates that screening results from large libraries should be interpreted with caution.

Determining synthetic failures in combinatorial libraries by hybrid gas-phase separation methods.

A combinatorial tripeptide library having the general form D-Glu-Xxx-Xxx-CONH2 has been synthesized using a standard mix and split synthetic protocol that is expected to produce 676 components. All components of the mixture were analyzed using a new high-resolution ion mobility/time-of-flight mass spectrometer coupled with an electrospray ionization source. In this approach ions are separated by differences in their gas-phase mobilities prior to being introduced into the mass spectrometer for mass-to-charge analysis. The peptide library includes a wide range of different sequence, structural, and stereo isomers; trends in the number of expected and resolved isomers that are observed at each m/z ratio allow specific synthetic steps that have failed to be identified, even in the presence of other isomers. Information about the relative abundances of different isomers should dramatically improve the reliability of binding affinity studies from direct analysis of mixtures.

Gas-phase separations of electrosprayed peptide libraries.

High-resolution ion mobility spectrometry has been combined with time-of-flight mass spectrometry for analysis of a combinatorial peptide library that is expected to contain 676 components. In this approach, the components of a mixture of three residue peptides, having the general form (D)Phe-Xxx-Xxx-CONH2 (where Xxx is randomized over 26 residues including 10 naturally occurring amino acids and 16 synthetic forms) were ionized by electrospray ionization. Ion mobility/time-of-flight distributions have been recorded for all ions using a nested drift(flight) time technique. The improvement in resolving power [(t/delta t) = 100-150 for singly charged ions] was illustrated by analysis of a mixture of tryptic digest peptides using high- and low-resolution instruments. The approach allows many components of the library (e.g., structural, sequence, and stereo isomers) that cannot be distinguished by mass spectrometry alone to be resolved. Impurities due to side reactions appear to be minimal, comprising < 10% of the total ion signal. Direct evidence for approximately 60-70% of the expected peptides is found. Variation in ion abundance for different components indicates that there are differences in solution concentrations or ionization efficiencies for the components.

High-order structure and dissociation of gaseous peptide aggregates that are hidden in mass spectra.

Injected-ion mobility and high-pressure ion mobility techniques have been used to examine the conformations of bradykinin, insulin chain A, and several other peptide ions in the gas phase. Under the experimental conditions employed, evidence for multimer formation in the mass spectra of peptides is minimal or absent altogether. However, ion mobility distributions show that aggregates of peptides (containing a single charge per monomer unit) are observed at the same mass-to-charge ratios as the singly charged parent ions. Collision cross sections for these clusters show that they have tightly packed roughly spherical conformations. We have bracketed the average density as 0.87 < p < 1.00 g cm-3. In some cases, specific stable aggregate forms within a cluster size can be distinguished indicating that some high order structures are favored in the gas phase. Multimer formation between different sizes of polyalanine peptides shows no evidence for size specificity in aggregate formation. Collisional and thermal excitation studies have been used to examine structural transitions and dissociation of the multimers. Aggregates appear to dissociate via loss of singly charged monomers. The observation that peptide multimers can be concealed in mass spectral data requires that fragmentation patterns and reactivity studies of singly charged monomers be undertaken with care.