Abstract
Proteomics is already invaluable for both applied studies, e.g. the discovery of novel biomarkers, and in fundamental investigations of cellular processes. However, there is a clear need to increase its sensitivity and selectivity, while retaining a high throughput of samples. We report here two general approaches: (i) systematic use of orthogonal mass datasets via methylation and (ii) enhanced chips for Desorption/lonisation on Silicon (DIOS), which go some way to address these issues. In the first procedure, samples for peptide mass fingerprinting are split and one half is methylated at COOH groups. The masses of the native and methylated peptides are then measured and used to search sequence databases in conjunction with a modified mass fingerprinting algorithm. Such orthogonal mass datasets efficiently identify proteins even when very limited peptide mass data is available. The second approach uses a novel DIOS chip fabricated in single crystal silicon to circumvent the problem of chemical noise inherent in matrix-assisted laser desorption ionisation mass spectrometry (MALDI-MS). The chips have a hitherto unreported columnar structure with an extremely high aspect ratio that can be reproducibly fabricated in localised areas by masking. The columnar structure has been realised without the use of pore initiation techniques, which require high definition lithography, and is quite different from porous silicon obtained by electrochemical etching techniques. The use of both approaches in several biological settings, including the analysis of the pathogen Mycobacterium tuberculosis, is described.