Oxidative stress-induced mitochondrial protein degradation and peptide release, Released peptide data
The turnover of mitochondrial proteins under oxidative stress conditions was studied in vitro by incubating mitochondria isolated from potato tubers (Solanum tuberosum L.) in a medium containing a substrate cocktail and ATP (control), in the same medium plus FCCP (uncoupled control with low production of reactive oxygen species (ROS), or in the medium plus methyl viologen and KCN to block electron transport and maximize ROS production (oxidative stress). After 15 min incubation, the mitochondria were pelleted, digested with trypsin, iTRAQ-labelled and the samples pooled. The pooled sample was analyzed by liquid chromatography-mass spectrometry. We found that 69 tryptic peptides decreased in amount relative to other peptides in the same protein after the oxidative stress treatment, but not after FCCP treatment. This indicates that the peptides had been modified, probably by oxidation. The modified proteins represented a wide range of proteins, but the respiratory complexes, the tricarboxylic acid cycle enzymes and ROS-detoxifying enzymes were overrepresented. Many peptides were released from these POM and many more so in the presence of the pore-forming peptide alamethicin especially from the matrix and the inner mitochondrial membrane (IMM) consistent with the formation of large pores in the IMM. Twice as many peptides appeared from matrix and IMM proteins during the oxidative stress treatment pointing at an accelerated protein turnover. We then matched the sequence of the released peptides from respiratory chain complex components with that of their parent proteins (having a known IMM orientation), and this allowed us to identify the initial site of peptide release – matrix, IMM, intermembrane space, or cytosol/medium. The location of the cleavage site and its sequence signature also allowed us to predict which of the known ATP-dependent and ATP-independent mitochondrial proteases were likely responsible for the release. We conclude that isolated mitochondria respiring in vitro over 15 min have a detectable protein turnover. This protein turnover is greatly accelerated during severe oxidative stress leading to the release of many peptides that are potential retrograde signals to the nucleus.
Workflow is depicted in workflow.png:
1. Flow diagram for the treatment and proteomic quantification and identification. Isolated mitochondria from potato tubers were treated under three different conditions (control or KCN). The mitochondria were pelleted by centrifugation and the supernatant fractionated on a spin-filter allowing only molecules <10 kDa to pass through. The mitochondria were incubated for 1 min at the end of the incubation (immediately prior to pelletation) with the pore-forming peptide alamethicin, which is known to make the IMM permeable to smaller moleculesThe flow-through was up-concentrated by micro-column reverse phase chromatography before analysis by LC-MS and bioinformatic identification. The peptides were finally analyzed with LC-MS and identified using bioinformatics.
Samples containing biotinylated peptides were resuspended in 5 µl 0.1% formic acid and loaded on an EASY nLC system (Thermo Fisher Scientific) set up with two columns. Both the pre-column (100 μm inner diameter, 2 cm long) and the analytical column (75 μm inner diameter, 15 cm long) were in-house packed with C18 ReproSil reverse-phase material in fused silica. The flow of 250 nl/min was delivered and the peptides were separated using a 55 min gradient from 0-38% B buffer before washing with 100% B for 12 min (A buffer: 0.1% formic acid, B buffer: 0.1% formic acid, 90% acetonitrile). The flow from the analytical column was either coupled to an LTQ Orbitrap Velos Pro mass spectrometer (Thermo Fisher Scientific) for the analysis of released peptides or Q-Exactive Plus (Thermo Fisher Scientific) for the analysis of the digested pellets. The instruments were operated in positive ion mode with data-dependent acquisition. The Velos main settings were as follows: A full ion scan (from 350–1650 m/z) was acquired at resolution of 30,000 before the 10 most intense precursor ions with charge states larger than +1 and intensity above 15000 counts were selected for collision-induced disassociation (CID) fragmentation using a normalized collision energy of 35 %. Former target ions selected for fragmentation were dynamically excluded for 30 s. The Q-Exactive was operated with the following main settings: A full ion scan (from 400–1200 m/z) was acquired at resolution of 70,000 before the 12 most intense precursor ions with charge states larger than +1 were selected for higher-energy C-trap dissociation (HCD) fragmentation using a normalized collision energy of 34 % prior to detection in the Orbitrap with a resolution of 17,500. Former target ions selected for fragmentation were dynamically excluded for 20 s.
For iTraq data: https://doi.org/10.6084/m9.figshare.28607600.v1
