Responses to many growth and stress conditions are assumed to act via changes to the cellular redox status. However, direct measurement of pH-adjusted redox state during growth and stress has never been carried out. Organellar redox state (E GSH) was measured using the fluorescent probes roGFP2 and pHluorin in Saccharomyces cerevisiae. In particular, we investigated changes in organellar redox state in response to various growth and stress conditions to better understand the relationship between redox-, oxidative- and environmental stress response systems. E GSH values of the cytosol, mitochondrial matrix and peroxisome were determined in exponential and stationary phase in various media. These values (−340 to −350 mV) were more reducing than previously reported. Interestingly, sub-cellular redox state remained unchanged when cells were challenged with stresses previously reported to affect redox homeostasis. Only hydrogen peroxide and heat stress significantly altered organellar redox state. Hydrogen peroxide stress altered the redox state of the glutathione disulfide/glutathione couple (GSSG, 2H+/2GSH) and pH. Recovery from moderate hydrogen peroxide stress was most rapid in the cytosol, followed by the mitochondrial matrix, with the peroxisome the least able to recover. Conversely, the bulk of the redox shift observed during heat stress resulted from alterations in pH and not the GSSG, 2H+/2GSH couple. This study presents the first direct measurement of pH-adjusted redox state in sub-cellular compartments during growth and stress conditions. Redox state is distinctly regulated in organelles and data presented challenge the notion that perturbation of redox state is central in the response to many stress conditions.

Sustained inflammation may increase the susceptibility of hepatocytes to apoptotic cell death and therefore exacerbate liver damage. Here we report that the pro-inflammatory cytokine IL-1β sensitizes primary murine hepatocytes to Fas ligand (FasL)-induced caspase-3/-7 activity. This process was dependent on JNK1/2 and the BH3-only proteins Bim and Bid. Mathematical modeling revealed that incubation of hepatocytes with IL-1β depleted the anti-apoptotic Bcl-2 protein pool and thus shifted hepatocytes to mitochondrial type II apoptosis following Fas activation. As a consequence, IL-1β and FasL treatment enhanced cytochrome c release. Surprisingly, despite increased caspase-3/-7 activation, FasL-induced cell death was reduced by IL-1β pre-treatment. This protective effect was independent of JNK1/2, Bim or Bid. Furthermore, elevated caspase-3/-7 activity upon IL-1β and FasL treatment did not result in enhanced PARP cleavage. The protective effect of IL-1β was seen after 3 h of pre-incubation, indicating an anti-apoptotic transcriptional response. Indeed, NF-κB DNA binding was increased in response to IL-1β plus FasL and gene-expression profiling of NF-κB regulated genes revealed a transcriptional and translational upregulation of the caspase-8 inhibitor A20. A mathematical model was developed to explain the contradictious occurrence of both increased caspase-3/-7 activity and elevated cell viability by including a heterogeneous distribution of Bcl-2 proteins and variations in Fas signaling resulting in different subpopulations of hepatocytes.

The autophagy-related ATG8 protein GABARAP has not only been shown to be involved in the cellular self-degradation process called autophagy but also fulfils functions in intracellular trafficking processes such as receptor transport to the plasma membrane. Notably, available mass spectrometry data suggest that GABARAP is also secreted into extracellular vesicles (EVs). Here, we confirm this finding by the immunoblotting of EVs isolated from cell culture supernatants and human blood serum using specific anti-GABARAP antibodies. To investigate the mechanism by which GABARAP is secreted, we applied proximity labelling, a method for studying the direct environment of a protein of interest in a confined cellular compartment. By expressing an engineered peroxidase (APEX2)-tagged variant of GABARAP—which, like endogenous GABARAP, was present in EVs prepared from HEK293 cells—we demonstrate the applicability of APEX2-based proximity labelling to EVs. The biotinylated protein pool which contains the APEX2-GABARAP co-secretome contained not only known GABARAP interaction partners but also proteins that were found in APEX2-GABARAP’s proximity inside of autophagosomes in an independent study. All in all, we not only introduce a versatile tool for co-secretome analysis in general but also uncover the first details about autophagy-based pathways as possible biogenesis mechanisms of GABARAP-containing EVs.

Subcellular structures containing autophagy-related proteins of the Atg8 protein family have been investigated with conventional wide-field fluorescence and single molecule localisation microscopy. Fusion proteins of GABARAP and LC3B, respectively, with EYFP were overexpressed in HEK293 cells. While size distributions of structures labelled by the two proteins were found to be similar, shape distributions appeared quite disparate, with EYFP-GABARAP favouring circular structures and elliptical structures being dominant for EYFP-LC3B. The latter also featured a nearly doubled fraction of U-shape structures. The experimental results point towards highly differential localisation of the two proteins, which appear to label structures representing distinct stages or even specific channels of vesicular trafficking pathways. Our data also demonstrate that the application of super-resolution techniques expands the possibilities of fluorescence-based methods in autophagy studies and in some cases can rectify conclusions obtained from conventional fluorescence microscopy with diffraction-limited resolution.