HE is not really cytosolic, because of its positive charge, marking it easy to penetrate membrane and prone to bind DNA. Prolonged incubation of HE results eventually in a heightened background of fluorescence that is not contributed by specific reaction with superoxide. Therefore overincubation should be avoided. HE preferentially localizes to nucleus in the end points. 

MitoSOX has been recommended by Invitrogen to be used at doses between 1-5 uM. At 5uM, for over 30 min, the probe becomes visiaully toxic to some cells, and triggers off more ROS production, artificially. 

In imaging, MitoSOX also contributes to some (nuclear) DNA staining if incubated for 30 min or longer. 

In theory, an ROS probe can work like a scavenger towards some ROS, esp. when its probe concentration is relatively low to yield a detection fluorescence (e.g. below 1 uM).

MitoSOX is recommended to be used at about 2-.2.5 uM, 15 min to 20 min. 

Mitochondria have their own DNA, so specificity of MitoSOX has been in debate. 

See: 

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2228267/

http://www.pnas.org/content/103/41/15038.abstract

http://www.nature.com/nprot/journal/v2/n9/full/nprot.2007.327.html

HE is not really cytosolic, because of its positive charge, marking it easy to penetrate membrane and prone to bind DNA. Prolonged incubation of HE results eventually in a heightened background of fluorescence that is not contributed by specific reaction with superoxide. Therefore overincubation should be avoided. HE preferentially localizes to nucleus in the end points. 

MitoSOX has been recommended by Invitrogen to be used at doses between 1-5 uM. At 5uM, for over 30 min, the probe becomes visiaully toxic to some cells, and triggers off more ROS production, artificially. 

In imaging, MitoSOX also contributes to some (nuclear) DNA staining if incubated for 30 min or longer. 

In theory, an ROS probe can work like a scavenger towards some ROS, esp. when its probe concentration is relatively low to yield a detection fluorescence (e.g. below 1 uM).

MitoSOX is recommended to be used at about 2-.2.5 uM, 15 min to 20 min. 

Mitochondria have their own DNA, so specificity of MitoSOX has been in debate. 

See: 

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2228267/

http://www.pnas.org/content/103/41/15038.abstract

http://www.nature.com/nprot/journal/v2/n9/full/nprot.2007.327.html

Yes, I have observed similar results with aspirin in yeast cells.

Is the drug dose you're administering to the cells lethal, or do the cells survive?

If the cells survive and assuming the results are correct, it might it be possible that the cells increase their mitochondrial manganese superoxide dismutase (MnSOD) activity as a stress response to the drug. This would effectively quench the superoxide radicals being generated by the drug within the mitochondria (presumably in the matrix, where there is 1) MnSOD and ii) mitochondrial DNA for any oxidised mitoSOX product to bind with and become highly fluorescent), consequently lowering the signal of the MitoSOX probe.   

Also bear in mind that superoxide radicals derived from mitochondria (typically from some induced malfunction of the electron transport chain), may also be generated on the other side of the inner membrane, facing the inter-membrane space (IMS). Superoxide radicals generated in the IMS might potentially leak out into the cytosol through the more permeable outer mitochondrial membrane, which might account for the increased DHE fluorescence in your cells in response to the drug. 

Although, I suspect that the mitochondria constitute the critical target of your drug (as it is for many drugs with anti-neoplastic properties, including aspirin) I would also not exclude the involvement of other (non-mitochondrial) sources of superoxide radicals such as NADPH oxidases. 

The above is speculation of course but I hope that it may help to some degree. 

As a suggestion, might it be worth testing for mitochondrial (and perhaps even cytosolic?) SOD activity using SOD activity gels or assays to complement your ROS studies? 

I hope this helps. 

Gianluca 

Dear Han Shen,

it may also depend on what type of drug you are using - where exactly the ROS are generated. If in the cytosol, or even outside cells, then the results you are getting is quite plausible. If you are expecting an increased mitochondrial respiration, then you should be getting an increase in Mitosox red fluorescence. 

Martyna