Effect of water treatment on analyte and matrix ion yields in matrix-assisted time-of-flight secondary ion mass spectrometry: the case of insulin in and on hydroxycinnamic acid

A systematic study was performed to identify the origin of surprisingly high analyte-to-matrix yield ratios recently observed in time-of-flight secondary ion mass spectrometry (TOF-SIMS) analysis of oligo- and polypeptides mixed in matrices of alpha-cyano-4-hydroxycinnamic acid (4HCCA). Several sets of samples of porcine insulin in 4HCCA (1:3100 molar) were prepared from liquid solutions by a nebuliser technique, with more than one order of magnitude variation in sprayed material (substrate silicon). Following different periods of storage in air and/or vacuum as well as exposure to high-purity water, TOF-SIMS analysis was performed under oblique impact of 22 keV SF5+. Treatment with water involved either deposition of a droplet covering the whole sample for times between 1 and 20 min or spraying with water in droplet equivalent quantities. The analyte and matrix molecules were detected as protonated molecules (insulin also in doubly protonated form). Even the as-prepared samples usually showed insulin-to-4HCCA yield ratios exceeding the molar ratio of the mixed material. Upon ageing in vacuum the matrix ion yields remained constant but the analyte yields decreased, partly due to break-up of intrachain disulfide bonds. Water treatment resulted in a pronounced decrease in the 4HCCA yield, typically by a factor of five, in parallel with an increase of the insulin yield, by up to a factor of four. Evidence is provided that these changes occur concurrently with a partial dissolution of 4HCCA at the sample surface. The enhanced insulin yield was not correlated with the Na+ yield. The typically 20-fold increase in the insulin-to-4HCCA yield ratio, generated by water exposure of the samples, provides the explanation for the high yield ratios observed previously with water-treated samples. Spraying with water or repeated exposure to water droplets caused a pronounced degradation of the insulin parent yields in combination with an increasing appearance of signals due to the B- and A-chains of insulin. To clarify the issue of surface segregation, a few samples were prepared by spraying acetone-diluted solutions of insulin on previously deposited layers of 4HCCA. Whereas the insulin yields from as-prepared samples were rather low, the yields observed after water treatment were comparable with those observed with samples of insulin in 4HCCA. The results suggest that a large amount of insulin is present at the surface of samples prepared from liquid mixtures of insulin in 4HCCA. With both methods of sample preparation, however, high secondary ion yields of insulin were only obtained after exposure of the samples to water. The chemical changes responsible for this beneficial effect still need to be identified. Copyright (C) 2002 John Wiley Sons, Ltd.