Generation of a murine SWATH-MS spectral library to quantify more than 11,000 proteins

Abstract

Targeted SWATH-MS data analysis is critically dependent on the spectral library. Comprehensive spectral libraries of human or several other organisms have been published, but the extensive spectral library for mouse, a widely used model organism is not available. Here, we present a large murine spectral library covering more than 11,000 proteins and 240,000 proteotypic peptides, which included proteins derived from 9 murine tissue samples and one murine L929 cell line. This resource supports the quantification of 67% of all murine proteins annotated by UniProtKB/Swiss-Prot. Furthermore, we applied the spectral library to SWATH-MS data from murine tissue samples. Data are available via SWATHAtlas (PASS01441).

Fig. 1

Sample preparation and data analysis workflows used in the generation of the spectral library. L929 cell lysates were first fractionated with size-exclusion chromatography and digested with trypsin. The resulting peptides were fractionated with HILIC (Hydrophilic Interaction Liquid Chromatography). The tissue samples were digested with trypsin and the peptides were fractionated with high-pH chromatography. The peptide fractions were dissolved in 0.1% formic acid containing iRT peptides, which were analyzed using shotgun MS. The DDA files were searched with X!Tandem and Comet, and results were combined with iProphet. The combined results were filtered with 1% protein FDR and made a consensus spectral library with Spectrast software.

Fig. 2

Characteristics of the murine spectral library. (a) True positive (black) and all protein identifications (red) as a function of protein FDR. The vertical dashed line was protein FDR of 0.01 determined by MAYU software. (b) True positive (black) and all peptide identifications (red) as a function of protein FDR. The vertical dashed line was protein FDR of 0.01 determined by MAYU software. (c) Overlap of murine proteins in UniProtKB/Swiss-Prot, a subset annotated with protein-level evidence and the murine spectral library. (d) Coverage of the proteome for SWATH-MS spectral libraries of different species. The numbers of proteome coverage were directly taken from the cited publication^,,,. (e) The number of proteotypic peptides per protein in the murine spectral library.

Fig. 3

Analyzing tissue SWATH-MS data using the murine spectral library. (a) The numbers of quantified proteins at 1% global protein FDR in three technical replicates in seven tissue datasets. The mouse library and the internal libraries were used to analyze SWATH-MS data. (b) The numbers of quantified peptides at 1% global protein FDR in three technical replicates in seven tissue datasets. (c) Pearson correlation of protein intensities identified in two samples. (d) CV of log2-transformed intensities of quantified proteins in three replicates using L929 library. (e) The proteins with the top ten highest abundances in each tissue. The protein intensity was normalized with the sum of all protein intensities, and top ten proteins were shown. The tissue-function related proteins were labelled in red (protein entry names are from UniProt/SwissProt database).