Detection of Alternative Splice Variants at the Proteome Level in Aspergillus flavus

Abstract

Identification of proteins from proteolytic peptides or intact proteins plays an essential role in proteomics. Researchers use search engines to match the acquired peptide sequences to the target proteins. However, search engines depend on protein databases to provide candidates for consideration. Alternative splicing (AS), the mechanism where the exon of pre-mRNAs can be spliced and rearranged to generate distinct mRNA and therefore protein variants, enable higher eukaryotic organisms, with only a limited number of genes, to have the requisite complexity and diversity at the proteome level. Multiple alternative isoforms from one gene often share common segments of sequences. However, many protein databases only include a limited number of isoforms to keep minimal redundancy. As a result, the database search might not identify a target protein even with high quality tandem MS data and accurate intact precursor ion mass. We computationally predicted an exhaustive list of putative isoforms of Aspergillus flavus proteins from 20 371 expressed sequence tags to investigate whether an alternative splicing protein database can assign a greater proportion of mass spectrometry data. The newly constructed AS database provided 9807 new alternatively spliced variants in addition to 12 832 previously annotated proteins. The searches of the existing tandem MS spectra data set using the AS database identified 29 new proteins encoded by 26 genes. Nine fungal genes appeared to have multiple protein isoforms. In addition to the discovery of splice variants, AS database also showed potential to improve genome annotation. In summary, the introduction of an alternative splicing database helps identify more proteins and unveils more information about a proteome.

Discovery and Mass Spectrometric Analysis of Novel Splice-junction Peptides Using RNA-Seq

“…Another important issue in the discovery of alternative splice forms at the protein level is the low number of splicespecific peptides actually identified, an issue that has been revealed by work reported in the literature (17,19,22,24,25,30,31). Part of the reason for the low number of alternative splice variants detected are the technical differences between RNA-Seq and bottom-up proteomics, namely sequence coverage and detection sensitivity.…”

“…In this approach, exon coordinates are first determined by obtaining exon sequences from databases such as Ensembl or by using ab initio computational algorithms to predict the location of putative exon boundaries. Next, these exon sequences are assembled into all theoretical exon-exon (and exon-intron) combinations, and then the sequences are translated into polypeptide sequences and used for MS-based searching to discover novel splice variant peptides (27)(28)(29)(30). To extend this approach, several research groups have restricted their exon-exon database to include only those sequences corroborated with transcript expression data (31)(32)(33), thereby eliminating spurious sequences.…”

Most Highly Expressed Protein-Coding Genes Have a Single Dominant Isoform

“…11 However, many experiments identify many fewer alternative isoforms than would be expected, even if the low peptide coverage is taken into account. 6,12–14 …”

Accurate Peptide Fragment Mass Analysis: Multiplexed Peptide Identification and Quantification

“…Many reasons have been suggested for the low proteome coverage including poor quality MS/MS spectra 8, 9 and incorrect genome annotation. 10, 11 There is also strong evidence that the primary reason for the low coverage is the DDA paradigm. 12, 13 DDA methods have been shown to limit the dynamic range of the analysis 14 leading to inadequate sampling of the proteome, even for relatively simple organisms.…”