Perhaps you could perform an in vitro experiment using EPR spectroscopy to trap the free radicals in question. With the use of a particular spin trap, you can then identify the type(s) of free radicals that are involved in your mixture.

Dear Colleague:

I do not know how you could manage to fix the ROS elements as they could easily react with macromolecules of the cells. However, it is quite possible that the produced ROS can react with membrane structure and in particular with bi layer lipids and disturb the integrity of membrane. thus one of the best biomarker may be determination of lipid peroxidation end product (MDA).

good luck

Hu, Upendar, you can not fix ROS because of chemistry: very short living time. Also, DCFDA can not be fixed as well. But you can fix sites of ROS accumulations after NBT or DAB staining. ROS production usually associated with membranes, you are right. And, also, ROS is definitely a part of signalling pathways. Could you tell us whcih kind of cell have you used and all details of the staining? Good luck!

You cannot "fix" ROS (chemical impossibility), only their reaction products. What you are seeing in the fixed cells is previously oxidised DCF. Bearing in mind that DCF is a truly awful probe, and very non-specific, the first thing to establish is whether the signal you are seeing is actually due to ROS per se. I would suggest using a membrane permeable anti-oxidants such as ebselen and tempol to see if this reduces your response, and exogenous ROS (e.g. 100uM peroxide) to see if it increases (and if the distribution changes). If you are really seeing a peripheral accumulation of oxidised DCF, this could be due to limited access of DCF to the cytosol (exogenous ROS might tell you whether this is the case), or a plasmalemmal source of ROS, eg NADPH oxidase. As test, you could use DPI; though this hits all flavoproteins and is therefore completely non-selective, it would tell you that whether this is real ROS accumulation from a cellular source. You could also try VAS2870, which inhibits NADPH oxidase (at least, some isoforms). What sort of cells are they, and what stimulus are you using? Frankly, I would consider using a different probe, perhaps one that can be localised within the cell following expression (e.g. variants of RoGFP).

Undoubtedly highly reactive ROS interact with membrane lipids, triggering the reaction of lipid peroxidation, and oxidize membrane proteins. Normally (and even more so for various diseases), these processes trigger different intracellular signaling systems

So, ROS are known to be the important signalling molecules within cells, in particular - the cells of the immune system. More over, it was shown that lower level of oxygen radicals in NADPH oxidase-expressing cells leads to more severe of arthritis in animals [Hultqvist et al. (2004), http://www.ncbi.nlm.nih.gov/pubmed/15310853].

Other useful references:

http://www.ncbi.nlm.nih.gov/pubmed/22981911

http://www.ncbi.nlm.nih.gov/pubmed/9641271

http://www.ncbi.nlm.nih.gov/pubmed/15134513

http://www.ncbi.nlm.nih.gov/pubmed/11375748

http://www.ncbi.nlm.nih.gov/pubmed/16438798

http://www.ncbi.nlm.nih.gov/pubmed/17570640

I agree with Jeremy. The ROS is not fixed but react with the fluorophore. ROS is known to plays important role in cell proliferation. The signals you got may be produced by a membrane bound enzyme. Low concentration of ROS is known to be important role in multiple signaling pathways at least in mammalian cells. As Taras and Marina pointed out that NOX system is one of the candidate you should consider in your system.

Dear Ward J,

The dye is working fine as it had increased with H2O2 treatment and reduced the fluorescence with N-acetyl cystein (ROS scavenger/uncoupler). Actually I am looking this in yeast cells. ROS generation was examined in live cells only. But, for staining with other dyes like DAPI, I had to fix and process the samples. So, in this process, cells were processed after DCF staining by fixation and followed by alcohol, I got the DCF signal pattern like live cell staining. So, I am asking whether this membrane associated staining is because of ROS or ROS modified bi-products which can oxidise DCF-DA to active probe? I appreciate your Idea of using DPI for NOX inhibition (As we are already planned to check this). I hope the source for ROS is NOX only and that might be H2O2. Is there any possibility that these molecules can be accumulated in cell without reacting with any biomolecules (for signalling/mainitenance)?

Finally, I would like to tell you that I had also tried to use DCF-DA after fixation-alcohol permeabilization, which resulted in same staining pattern like that observed in live cells (same fraction were stained). This again forced me to ask the question that these ROS can be fixed or their biproducts were oxidising the dye?

Thank you

Hi, Upendar, if you stain cells with DCF_DA after fixation and have same patterns as in living cells, your staining is not a ROS...Moreover, please, take in to account that NOX produce extracellular superoxide and you have to use XTT for detection. You should not have direct accumulation of the H2O2 after NOX activation. DPI also (at specific dose) does not have an effect on intracellular H2O2.. Good luck!

Dear Upendar, even though antioxidants and peroxide are having an effect in the live cells, I agree with Taras that if you are getting the same pattern of fluorescence after fixing the cells then that is not due to ROS! However, I would dispute that NOX only release ROS to the outside of the cell - that is true of inflammatory cells and NOX2, but in other cell types (smooth muscle for example) there is lot of evidence that NOX1, NOX4 and NOX5 (possibly) increase cytosolic ROS - and NOX1 is certainly involved in intracellular signalling. NB a good extracellular indicator for peroxide is Amplex Red. Note that DPI will inhibit xanthine oxidase as well as NOX and the mitochondria ETC. I know little about yeast cells and the type of NOX and other, and my comments are based on our work with mammalian smooth muscle.

NB There is evidence that DCF may actually be detecting highly reactive products of the Fenton reaction (hydroxyl radicals) rather than peroxide itself.

Dear Dr. Taras,

Thank you for your suggestions. I would like to ask you that the staining of DCF after fixation is not ROS, then what is reacting with the probe and converting to fluorescent active state? And there is a difference in staining pattern (live and fixed) but the same cells were responding/showing positive to stain (not all). If DCDHF-DA is not specific then how only few cells were responding to it? I agree that NOX will not generate H2O2 directly. SOD1,2 which are transcritionally active (in my case), which converts these superoxide to H2O2. So, that's the reason I am suspecting NOX as a ROS source. For Mitochondrial permeability, I had used Mitotracker-red (introgen molecular probes), which is failed to show alterations in membrane permeability, which further forced me to suspect NOX.

If you suggest any H2O2 specific detection probes for yeast, that will help me a lot.

Thank you once again.

Hi, Upendar,

Reactive oxygen species are formed from different sources in living cells. Most of them have small lifespan because of high reactivity and high impact of the antioxidant defence system. So you cannot fix them as you can fix proteins and membrane structure.

The membrane-localized distribution of the dye that you observed and that is not sensitive to previous cell fixation (!) could follow from the process of diffusion of this fluorescent dye into the cells with following auto-oxidation proportionally to local dye concentrations.

Take into account, please, that auto-oxidation and redox cycling could also be inhibited by NAC et cet.

Superoxide reduces tetrazolium dyes to more or less insoluble colored products. The classic example is the nitrobluetetrazolium test, which visualizes superoxide-producing organelles in phagocytic cells.

Once used iodonitrotetrazolium to visualize the superoxide-producing system in mucus. Posted on my researchgate page at: https://www.researchgate.net/publication/22940820_A_superoxide-producing_system_in_the_conjunctival_mucus_thread . Don't know if it will work similarly on cell walls, membranes, etc. But this does seem possible.

Article A superoxide-producing system in the conjuctival mucus thread

Hi Upendar, have you tried to use spin traps and EPR? I think it would be a good option to detect the signal for each reactive species occuring in your systems.

Dear Dr.Macros,

Currently, there is a problem in machine. I have plans for it.

Is EPR detects ROS qualitatively or quantitatively?

how to differentiate the signals before and after ROS accumulation?

Thank you for your suggestion.

I mean, I am not an EPR specialist but I have read interesting articles showing which ROS are formed. I don't really know if it is possível to quantitate. Pehaps, it is possível to adjust the methodology. I also suggest you to talk a little with chemists.

The cell-permeant 2',7'-dichlorodihydrofluorescein diacetate (H2DCFDA) (also known as dichlorofluores cin diacetate) is a chemically reduced form of fluorescein used as an indicator for reactive oxygen species (ROS) in cells, for example to detect the generation of reactive oxygen intermediates in neutrophils and macrophages. Upon cleavage of the acetate groups by intracellular esterases and oxidation, the nonfluorescent H2DCFDA is converted to the highly fluorescent 2',7'-dichlorofluorescein (DCF). Versions of this indicator that have better cellular retention are available (C400, C2938).

Inasmuch a peroxidized phospholipid, such as a lecithin molecule carrying a hydroperoxide group at a linoleoyl residue, for instance, can be considered a ROS, then my answer is yes. In fact, we were able to build stable lipid vesicles containing hydroperoxidized lecithin molecules up to 100% by mole. In my opinion, and under given conditions (to be studied) these hydroperoxy groups, bound to oxidized PUFA, should be able to oxidize neighbour normal PUFA and extend membrane phospholipid oxidation to the whole membrane, including (enzymatic) proteins prone to oxidation.

at

Various peroxidized phospholipids were reacted with proteins under N2. In all cases, phospholipid is bound covalently to the proteins whose molecular size is increased. Both the amount of bound phospholipid and the increase in molecular size of the protein depends on the nature of the phospholipid. Ultraviolet (UV) absorption of the proteins is increased in qualitatively similar ways. Their difference spectra, which show a gradual increase in absorption from 400 nm toward shorter wavelength, differ from that of malonaldehyde-protein complexes. The various complexes of proteins and peroxidized phospholipids have similar fluorescence spectra showing two excitation maxima at 310–320 nm and at 340–350 nm, respectively, and emission maximum at ca. 400 nm. This is different from both fluorescence spectra of malonaldehyde-protein complexes and fluorescence spectra reported for proteins after reaction with peroxidized polyunsaturated fatty acids. Amino groups of the proteins are consumed in the reaction with peroxidized phospholipids. Blocking the amino groups decreases the binding of phospholipid considerably. Besides amino groups, other structures of the protein molecule react with the peroxidized phospholipids. The similar features of UV absorption, fluorescence, decrease of amino groups, and covalently bound phospholipid phosphorus of the various complexes suggest that they are formed by common type of reactions. The reactions seem to be different from those generally believed important between peroxidized lipid and protein. Important reacting species are compounds other than malonaldehyde.

The importance of understanding the mechanisms of modulation of cellular signaling cascades by the peroxidized membrane phospholipids (PLs) is well recognized. The enzyme-catalyzed peroxidation of PLs, as opposed to their oxidation by air and metal catalysis, is well controlled and rapid and yields well-defined PL peroxides which are highly desirable for biological studies.

Chemical/enzymatic synthesis of specific oxidized PL (OxPL) molecules proves to be the correct route to controlling/modelling oxidized membranes for biophysical/biochemical study. As for UV absortion of OxPL, the main tool is absorption at 235nm of the conjugated diene system, useful for moderately oxidized PL. Heavily oxidized PL, i.e. truncated FA forms terminating with CHO or COOH group originating after cleavage of the conjugated diene bonds, lack 235 nm absorption and present a slight absortion around 270nm, not so useful. In addition, malondialdehyde itself, under oxidizing conditions (Fenton), is subjected to further oxidation to malonaldehyde, escaping the TBARs assay

ROS simply cannot be fixed. They are too short lived. What you may have detected is continuous production. However, the question is not that simply answered. Irina and Francesco among others have addressed the major issue quite adequately. As for spin traps, they are complex to use and as I can only access a shared facility I cannot even time my experiments accordingly. Probes are the way to go for for me too and they are quite adequate. However if you wish to quantify, use a fluorescent plate reader instead.

It does surprise me that the cell, treated harshly as you have done, continue to give fluoresce. As for the glow confined to peripheri, the membrane NAD(P)H oxidoreductases could be a major source of ROS. You do need to do some major reading before you waste any more expensive probes there.

Do you have deconvolution software on your microscope? that localizes glow much more effectively than passthrough from the thickness of the cell. Personally, with so many sources or ROS, such great variety of ROS, non-specificity of probes (including DCFH-DA) and a multitude of scavenging systems, enzymatic as well as non-enzymatic, make this a horribly complicated area to evaluate. This must be realized at all costs and relevant controls setup.

I would also refer you to the molecular probes site that has a lot of data on probes and also offers a huge no. of references.

This is an old thread but if anybody is still following, I have a question too. I am trying to process several culture dishes simultaneously for ROS dyes followed by microscopy. Since the cells are in PBS, by the time I reach the last dish, the cells lose their morphology. My question is, can I use the dyes on live cells to stain for ROS, then fix the cells with Formalin? Basically, can I fix the oxidized dye and do microscopy later on? I understand that I can not detect ROS on previously fixed cells.

Hi Sanjay,

Is there a specific reason to use PBS? If you are worried about phenol red you can order phenol red free medium this should keep cell happy for longer. Similarly, you can use alternatives for FBS (serum-free media). I have reservations about keeping cells in PBS for extended periods of time especially when performing intracellular measurements like ROS, mitochondrial potential and Ca2+ signalling. 

Regarding fixable ROS dyes check these websites;

https://www.thermofisher.com/au/en/home/references/newsletters-and-journals/bioprobes-journal-of-cell-biology-applications/bioprobes-issues-2011/bioprobes-64-april-2011/cellrox-deep-red-reagent-for-ros-detection-april-2011.html

https://www.thermofisher.com/au/en/home/life-science/cell-analysis/cell-viability-and-regulation/oxidative-stress.html

Unfortunately, I have no experience in using these fixable probes but will love to hear about first-hand experiences. The best strategy is to get a sample first and test it against regular probes for your cell based system.

Hope this helps.

Cheers,

Rohit

Hi Rohit,

Thank you for the tips. I am dealing with some very moody cells which start loosing their morphology in about 30 minutes to 1hr. Secondly, I am going to try an unconventional ROS dye, directed specifically to selected ROS species. But you are right about keeping cells for too long in PBS. May be I will do  the treatments sequentially in a way so that all the culture dishes spend nearly same amount of time in PBS/HBSS/PR free media.

Thanks.

Best,

Sanjay

Superoxide directly reduces nitroblutetrazolium (NBT) to an insoluble blue compound.  This forms the basis of the NBT test for superoxide production in neutrophils.   Unfortunately,   the amounts required are well-beyond those involved in redox-signaling..

I think the following paper will response your questions

"Detection of reactive oxygen species during the cell cycle under normal culture conditions using a modified fixed-sample staining method."

https://www.ncbi.nlm.nih.gov/pubmed/24749949

Dear Dr. Tavakol

Thanks for the suggestion. This is a nice article. Will add some confidence to my thoughts and give strength to pursue my problem further.

I can only think about 8-OH-dG which will be a non-specific biomarker for ROS.