Automated Querying and Identification of Novel Peptides using MALDI Mass Spectrometric Imaging

Abstract

MSI is a molecular imaging technique that allows for the generation of topographic 2D maps for various endogenous and some exogenous molecules without prior specification of the molecule. In this paper, we start with the premise that a region of interest (ROI) is given to us based on preselected morphological criteria. Given an ROI, we develop a pipeline, first to determine mass values with distinct expression signatures, localized to the ROI, and second to identify the peptides corresponding to these mass values. To identify spatially differentiated masses, we implement a statistic that allows us to estimate, for each spectral peak, the probability that it is over- or under-expressed within the ROI versus outside. To identify peptides corresponding to these masses, we apply LC-MS/MS to fragment endogenous (nonprotease digested) peptides. A novel pipeline based on constructing sequence tags de novo from both original and decharged spectra and a subsequent database search is used to identify peptides. As the MSI signal and the identified peptide are only related by a single mass value, we isolate the corresponding transcript and perform a second validation via in situ hybridization of the transcript. We tested our approach, MSI-Query, on a number of ROIs in the medicinal leech, Hirudo medicinalis, including the central nervous system (CNS). The Hirudo CNS is capable of regenerating itself after injury, thus forming an important model system for neuropeptide identification. The pipeline helps identify a number of novel peptides. Specifically, we identify a gene that we name HmIF4, which is a member of the intermediate filament family involved in neural development and a second novel, uncharacterized peptide. A third peptide, derived from the histone H2B, is also identified, in agreement with the previously suggested role of histone H2B in axon targeting.

Imaging mass spectrometry statistical analysis

“…The advantage of histology-directed protein profiling is that the experimentalist is more free to optimize the sample preparation of the MALDI experiment (larger matrix droplets) for increased sensitivity, because the spatial resolution requirements are less stringent, and as data is only acquired from specific features of interest the experiments are faster and more focused on the pathological question. Bruand et al have reported a workflow to identify MS peaks with MS images localized to specific features of interest [56]. b) Histology-directed.…”

Multiple Changes in Peptide and Lipid Expression Associated with Regeneration in the Nervous System of the Medicinal Leech

“…For example, the peptide with the m/z of 2475, previously detected in the embryo nerve cord ( Figure 2A ) is also present in the adult regenerating ganglion with an anterior expression bias ( Figure 2B ), whereas it is absent in controls (non regenerating adult CNS segmental ganglia, Figure S1 ). This peptide, which has been recently identified as a fragment of a novel intermediate filament protein, Hm IF4 [10] and has the N-terminal sequence GTRTMERSVRTSSQYASGGPMPN, provides evidence for the idea that embryonic factors are re-expressed or up-regulated during the process of regeneration.…”

Potential of MALDI imaging for the toxicological evaluation of environmental pollutants

“…Nevertheless, unraveling the molecular complexity of tissue samples remains a daunting task. The use of high-resolution mass spectrometers [63] or coupling MS with ion mobility [64] holds great promise in helping overcome this issue.…”