Skip to main content
. 2012 Aug 1;17(3):460–470. doi: 10.1089/ars.2011.4184

FIG. 4.

FIG. 4.

Exposure to hyperoxia induces mitochondrial matrix H2O2 in FPASMC. (A) The redox sensor roGFP was used to assess oxidant levels in cytosol and mitochondrial matrix. Surface-exposed cysteine residues are sensitive to the local redox environment. After obtaining ratiometric measurements of roGFP, the sensor was calibrated by maximally reducing it with dithiothreitol (DTT, 1 mM) and maximally oxidizing it with t-butyl hydroperoxide (TBH, 1 mM) to yield percent oxidation. FPASMC were plated on collagen-coated cover slips and were exposed in a flow-through chamber on an epifluorescence microscope to 95% O2/5% CO2 for 30 min, followed by a 30-min recovery period in 21% O2/5% CO2. Data shown are from 19 cells from four coverslips for mito-roGFP and 7 cells from five coverslips for cytosolic roGFP. *p<0.05 vs. FPASMC baseline oxidation in 21% O2/5% CO2; #p<0.05 vs. FPASMC maximal oxidation in 95% O2/5% CO2. (B) FPASMC were loaded with MitoSOX, a fluorescent mitochondrially targeted superoxide sensor, prior to 30 min of exposure to 95% O2/5% CO2 (n=10, measured in duplicate). (C) FPASMC were exposed to 95% O2/5% CO2 for 30 min, and intracellular H2O2 was quantified with Amplex Red fluorescence (n=8, *p<0.05 vs. FPASMC in 21% O2/5% CO2).

HHS Vulnerability Disclosure