Quantitative inner membrane proteome datasets of the wild-type and the Δmin mutant of Escherichia coli

This article presents data that were obtained through measuring the impact of the Min oscillation on membrane proteins in Escherichia coli by quantitative protemoics analysis. We isolated inner membranes from the wild-type and mutant strains to generate proteomics datasets based on NanoLC-nanoESI-MS/MS mass spectrometry using the isobaric tags for relative and absolute quantitation (iTRAQ) method. The datasets included the raw spectral files from four sample replicates and the processed files using Proteome Discoverer that contained a total of 40,072 MS/MS spectra with confident peptide identifier (FDR<0.01) and the peak intensity of the reporter ions. The data was further filtered, which resulted in an inner membrane proteome of unique proteins with quantitation. Proteins of interest, that show significant difference in protein abundance of the mutant membrane, were isolated through statistical filtering. The data is related to “Quantitative proteomics analysis reveals the Min system ofEscherichia colimodulates reversible protein association with the inner membrane” (Lee et al., 2016 [1]).


a b s t r a c t
This article presents data that were obtained through measuring the impact of the Min oscillation on membrane proteins in Escherichia coli by quantitative protemoics analysis. We isolated inner membranes from the wild-type and mutant strains to generate proteomics datasets based on NanoLC-nanoESI-MS/MS mass spectrometry using the isobaric tags for relative and absolute quantitation (iTRAQ) method. The datasets included the raw spectral files from four sample replicates and the processed files using Proteome Discoverer that contained a total of 40,072 MS/MS spectra with confident peptide identifier (FDR o0.01) and the peak intensity of the reporter ions. The data was further filtered, which resulted in an inner membrane proteome of unique proteins with quantitation. Proteins of interest, that show significant difference in protein abundance of the mutant membrane, were isolated through statistical filtering. The data is related to "Quantitative proteomics analysis reveals the Min system of Escherichia coli modulates reversible protein association with the inner membrane" (Lee et al., 2016 [1]

Value of the data
Data provide an inner membrane proteome of E. coli that contains information of protein topology and identifies new membrane-associating proteins.
The iTRAQ datasets offer the signature peptides with spectral evidence for the inner membrane proteins of the Δmin mutant, which could be useful for the experimental design of the targeted quantitation of proteins by mass spectrometry.
Data show enrichment in peripheral membrane proteins and metabolic enzymes that could represent a correlation between protein topology and function.
The network created based on the proteins of interest could guide assessment of molecular and genetic interactions to uncover additional function of the Min system.

Data
The mass spectrometry data files were generated to address whether the membrane-associating oscillator of the E. coli Min proteins would have a general role in cell physiology by influencing topology of membrane proteins [1]. The presented data include the dataset files (noted in Supplementary Table 1), the combined peptide quantification file (Supplementary Table 2), a network built based on the filtered proteins of interest (Network 1), and a network of MinD and MinE (Network 2).

Labeling with isobaric tags for relative and absolute quantitation
We resolved the IM proteins on SDS-PAGE (4% stacking and 10% separation gels) for protein extraction. The proteins were reduced, cysteine blocked, and digested, before labeling with the iTRAQ s reagent (Roche, Basel, Switzerland) containing different isobaric tags (114, 115, 116, or 117) according to the manufacturer's instruction. Four sample replicates were prepared from protein digests derived from independent cultures; each was labeled with a different isobaric tag. The protein samples from different strains with different isobaric tags were mixed in equal amounts for the LC-MS/MS analysis (Supplementary Table 1).

Protein quantitation and statistical analysis
We analyzed the mass spectrometry raw data with the Proteome Discoverer™ Software (v 1.4.1.14; Thermo Fisher Scientific, Inc.) and used Mascot (v 1.4; Matrix Science, MA, USA) for protein identification. The parameters used for Mascot processing were: precursor mass tolerance window of 10 ppm, fragment tolerance window of 50 mmu, dynamic carbamidomethylation on cysteine, dynamic oxidation on methionine, dynamic N-terminal and cysteine iTRAQ labeling. The missed cleavage was set as 2. We used all peptides of the first rank in each spectrum with a Mascot significant threshold less than 0.05 for data analysis to limit the false discovery rate (FDR) based on the targetdecoy database algorithm. The raw intensity values of each reporter ion for all confident peptides, i.e. FDR o0.01, were exported from Proteome Discoverer and combined for further data processing by normalization and aggregation. The resulting datasets correlated to 4 sample replicates (Supplementary Table 1) and the quantification data of all replicates were combined in a file (Supplementary Table 2). The data were further filtered to obtain an IM proteome of 808 unique proteins with quantitation [1]. In Lee et al. [1], we compared the level of protein expression between two subjects in the datasets (YLS1 vs. MC1000 with different isobaric labels) by the paired-sample t test to examine if the mean difference between experimental subjects is different than 0. A cut-off ratio at Z0.5 and a p value at r 0.05 were applied to obtain 40 proteins of interest (POIs) [1]. Forty POIs were imported into Cytoscape (Version 3.1.0, supported by NIGSM & NRNB, US) to retrieve data for network analysis [1]. The refined network is presented as Network 1. The data relating to MinD and MinE were extracted to create Network 2.

Reference library of protein subcellular topology
The subcellular topology data from databases of STEP (Subcellular Topologies of E. coli Polypeptides) [6], PSORT (a database of protein subcellular localizations for bacteria and archaea) [7], ASKA clone (-) (A complete Set of E. coli K-12 ORF Archive; National BioResource Project, Japan), and Dynamic Localizome [8] were combined into the protein library of E. coli strain K12 (organism ID 83333, retrieved from UniProtKB) that was manually curated to exclude duplicates, disrupted proteins, pseudogenes, and mobile elements [1]. This protein library was used as the reference to refine the datasets and to search for protein topology.