Figure 1.
Reconstitution of MDV formation in vitro.
A) EM sections of isolated mitochondria from bovine heart showing vesicular profiles. Scale bars are 500 nm in first panel, and 100 nm in 3 panels highlighting potential MDVs. B) Titration of cytosol with 5 mg of isolated mitochondria at 37°C for 60 minutes in the presence of an ATP regenerating system. Mitochondria were removed by centrifugation and 10% of the supernatant loaded onto an SDS-PAGE gel for transfer and western blotting, as indicated. 20 μg of starting mitochondria were loaded as an input control. C) The relative intensities of the cargo within the supernatant were quantified relative to that found within the 20 μg of starting mitochondria. Standard errors were calculated from three different gels loaded with the same reaction. The data is representative of at least 3 independent experiments. D) Time course of MDV formation under minimal conditions in the absence of cytosol. The reaction was incubated at 37°C for the indicated times in the presence of an ATP regenerating system and the supernatants were separated by SDS-PAGE and transferred for western blots as shown. E) Quantification of D), as in C).
Figure 2.
Mitochondrial stress triggers MDVs in vitro.
A) Isolated mitochondria were incubated in the absence of cytosol, with an energy regenerating system, for 60 minutes at 37°C (or 4°C) under the indicated conditions: 10 µM oligomycin, 50 µM Antimycin A, 50 µM Chloramphenical, 0.1 mM ATPγS, 200 µM xanthine and 0.4U xanthine oxidase. Along with 2 µg of the starting mitochondria, supernatants of the reactions were separated by SDS-PAGE for analysis by western blot, probing for anti-Core 2, a component of mitochondrial electron transport chain complex III, and the outer membrane channel VDAC. At the exposures shown here, the VDAC western blot does not detect the signal within the starting mitochondria. The data is representative of at least 3 independent experiments. B) As in A) except that 20 mM of NEM was added when indicated, and 0.1 mM GTPγS was used where indicated. C) Supernatants obtained from reactions performed in the presence of either 200 µM xanthine and 0.4U xanthine oxidase or 50 µM Antimycin A were fixed, and isolated in low-melt agarose and prepared for EM analysis. Images show thin sections from the pelleted MDVs, revealing both single and double-membrane bound vesicles. Note the differential electron electron density of material within the inner membrane compared to the space between the two membranes. Scale bars are all 100 nm.
Figure 3.
Stress-induced MDVs generated in vitro are enriched in oxidized cargo.
A) Reactions were performed as indicated and supernatants were separated by SDS-PAGE and transferred for western blots, including the matrix enzyme pyruvate dehydrogenase (PDH), and the indicated components of the electron transport chain complexes I through V. Note the absence of subunits within complexes I and V. B) Reactions were done as in B) and extraction of mtDNA was performed as described in materials and methods, and used as a template in a PCR reaction amplifying the mitochondrial DNA encoded bovine Cox1 gene. The starting mitochondria were used as a positive control, which efficiently amplified the Cox1 sequence, however there was no incorporation of mtDNA under any of the conditions tested. C) Normalizing the relative immunoreactivity of each cargo within the starting mitochondria (loading 2 and 5 µg), we quantified the amount of protein released into the trypsin-resistant supernatant fractions under the conditions indicated. Quantification of the indicated cargo was done from at least 3 independent experiments (error bars: SE). D) Reactions were performed under basal conditions, or upon treatment with Antimycin A as in A) except that the supernatants were not submitted to trypsin treatment. Free carbonyl groups within the fractions were derivitized with DNPH and quenched prior to separation by SDS-PAGE and blotting with an anti-DNPH Oxyblot antibody to reveal the oxidized proteins. The detection of protein oxidation from 20 µg material is shown, and VDAC was probed as a loading control for mitochondria and MDVs.
Figure 4.
Ultrastructural analysis of MDVs generated in vitro.
A) Supernatants obtained from reactions treated with Antimycin A were fractionated on a discontinuous sucrose gradient. After centrifugation, fractions were collected and analyzed by western-blot as indicated. 5 µg of starting mitochondria is included in the first lane. Two peaks of VDAC were noted, at the 20/30% interface and the 30/40% interface, as indicated by the arrows and numbers 1 and 2 above the figure. The heavier fractions also contained the inner membrane subunit Core2 of complex III. Tom20 was also seen in both fractions, although was less abundant within the lighter peak relative to VDAC (which was equally present in both peaks). The ER marker calnexin reveals some contamination in the heavier fractions. B) A repeat of A) probed for another subunit (NDUFA6) of Complex I, confirming it's absence within MDVs. Again, note the two peaks of VDAC, and Core2 present only in the heavier peak. C) Budding reactions were performed under either basal conditions, or in the presence of 50 µM Antimycin A. Following separation by sucrose density centrifugation as in A, B, proteins within peak fractions 1 and 2 were derivitized with DNPH, and conjugated proteins were separated by SDS-PAGE and revealed using anti-DNPH Oxyblot antibodies. The presence of VDAC in both peaks was used as an internal loading control. D) EM analysis from fractions in both peaks of a reaction performed in the presence of 50 µM Antimycin A was performed on fixed and low-melt agarose embedded samples. Scale bars are 100 nm.