As far as I know the SYTO dyes are propriatory and of undeclared structure. So all the following remarks are speculative! HOWEVER dyes which are selective for RNA/nucleoli are often intercalators. So in live cells they intercalate in dsRNA, but not in dsDNA as this would  require unwinding from the histones, which is not likely [see DeL Castillo et al. 2010. Biotechnic & Histochemistry 85: 247-256]. But you are staining fixed cells, where such factors may no longer apply. So the dye can intercalate with dsDNA. And presumably the binding affinity for dsDNA is greater than for dsRNA. By the way, the Hoechst dye stains dsDNA because it is not an intercalator, but binds in the minor groove.

Hi, Michael, I actually developed this dye when I was in R&D at Molecular Probes (I'm now in Tech Support).  It only works if you label live cells, then fix with methanol.  You can't label already-fixed cells and still get the same selectivity.  Did you label live then fix?

Also, what concentration and label time did you use?  This dye does have some low level of dsDNA labeling, and at too high of a concentration or time it will lead to nuclear labeling.

Hey Jason, thanks so much for responding. I tried labeling fixed cells according to the protocol provided on the Molecular Probes website (10 minutes in -20C MeOH). Then I stained in 500 nM for 20 minutes at room temperature.

So, I was attempting to label imaginal wing discs from drosophila larvae, as I'd seen this dye used in tissue before and the imaginal disc is more-or-less a monolayer. I tried staining unfixed discs and imaging them right away, but saw no signal.

Do you have any recommendations? Thanks again!

It may be a penetrance problem, where the dye may not be able to diffuse through the tissue.  You may be better served by a different protocol where you fix, permeabilize lightly, then label with a different RNA-selective dye, like Ethidium homodimer-2, RiboGreen, or YOYO-1.  You'd have to try a range of concentrations and label times, though.  

There's also the dye used in the Quant-iT RNA kits, which is very selective for RNA, but we've only tested it for solution assays in those kits and not for cell labeling (the dye standalone is Q32884): 

https://tools.thermofisher.com/content/sfs/manuals/Quant_iT_RNA_Assay_UG.pdf

Thank you much, Jason

Jürgen Rybak Likely they would label, though perhaps not with an intensity bright enough over background. SYTO RNASelect Green will penetrate tissues, but we don't know how deeply. So you'll need to determine these things with controls, empirically.

Hi Jason A Kilgore , I am trying to stain purified RNA with RNASelect to measure the electrophoretic mobility of complex via capillary zone electrophoresis. I used a concentration of RNASelect (~10uM) and RNA of (~6ng/ul). I am getting a lot of particle-like objects (maybe precipitates) with no clear signal from RNA (comparing with/without RNA solutions). what am I missing? is RNASelect can be used for purified RNA?

Thank you in advance.

Mahmoud Atta , certainly will label purified RNA. It was used for figure 1 in the manual: http://tools.thermofisher.com/content/sfs/manuals/mp32703.pdf. However, that's not its intended usage; the intended usage is for labeling RNA in cells. But you shouldn't see precipitation at 10 uM. Chance of precipitation will increase if the stock is very old, or if water has been introduced into the DMSO of the stock solution (including due to humidity and repeated openings). The dye is somewhat hydrophobic, so it should be used immediately after dilution into the staining solution buffer. If it sits for very long, it will start to precipitate out of solution. Do you think one of these possibilities might explain the issue?

Hi Jason A Kilgore

Thank you for prompt reply and sorry for late response, yes, one of these reasons might appear where, I am staining the purified RNA in a solution of (25 mM Hepes, 50 mM Imidazole and 10 uM RNASelect in water) and wait until preparing the capillary for electrophoresis (~20 minutes and more). should I change the protocol and use it immediately (within few minutes?)?

Hi Jason A Kilgore I am facing a problem with staining live bacterial cells. I tried different concentrations of the stain but somehow there is little to no fluorescence in live growing cells. I added the dye to the cells, let it incubate for a bit, and then put the cells on an LB agar pad on the glass slide, and place a cover slip on top.

The protocol mostly mentions about eukaryotic cells, can you help me with some tips regarding bacterial cell staining.

[Apologies for hijacking this thread]

Hi, Ishita. I'm afraid we've never tested that on bacteria. It may not be able to get through the cell wall. And, unfortunately, I don't know a better option. Some of the other SYTO dyes have been used for bacterial labeling, but how RNA-selective they are in live bacteria is not clearly known.

I think a better option would be to fix and permeabilize them first, DNase-treat them, then label with an RNA-selective dye in the wavelength you want. Again, we haven't tried it here at Molecular Probes, but some that have been useful for this in eukaryotic cells include Ethidium homodimer-1 (red, around 300 nM), RiboGreen (green, around 1:30 dilution), POPO-1 (blue, around 700 nM), or TOTO-3 (far red, around 1 uM).

Mahmoud Atta Yes, I'd recommend using right away. Be sure to check the pH, too. The HEPES should buffer it to physiological range, but distilled water can run acidic.

Jason A Kilgore Well in that case it should be mentioned in the dye description, that it might not work with bacteria and/or it hasn't been tested on bacteria or cell wall containing cells. Thank you, I'll have to look at something else now.

HI Jason A Kilgore , I tried your advice by using the RNASelect immediately for capillary zone electrophoresis, but I succeed once and fail tens. I am still seeing the particle-like objects even if I used it immediately. I am wondering what I am doing wrong? have you ever tried capillary zone electrophoresis with RNASelect? what buffer is good for hybridization with RNA? how long should I incubate for hybridization (20min is the cell incubation time in the protocol)?

Thank you.

Hi, Mahmoud Atta . Unfortunately, we have only validated it for labeling cells and not for your technique. So I don't know what else to suggest. But I can say that we have only used physiological buffers, such as PBS or HBSS.

Hi, Jason A Kilgore , Thank you so much! I think I could get reasonable results by optimization of the dye/RNA concertation. If I used very high concentration of the dye, I see a particles-like objects in the channel. If it is low concentration, The signal is very week. eventually, I got good signal at 5uM of RNASelect with RNA sample of ~30ng/ul, there might be another values that gives good results too. Thank you..

Mahmoud Atta Thank you for sharing that, Mahmoud. I'll add it to our Tech Support notes for others.

Jason A Kilgore I am also having difficulty with using this stain on bacteria. In the provided protocol it clearly states: 'Bacterial cells can also be stained with the RNASelect green fluorescent cell stain. Working concentrations of the dye should be kept below 1 μM. The 500 nM labeling solution, as prepared in step 2.3, is a good starting point for determining the optimal staining concentration.' Based on your discussion with Ishita, this appears to have never actually been tested? Please could you comment on this, or provide any insights into getting this dye working with bacteria (fixed or unfixed)

It appears I am mistaken. Thank you for pointing that out, Josh. Looking again at the product manual, there is a section that was added at the end of it that briefly mentions that, which you quoted. I did not test that usage when I developed the dye in R&D, so I'm not sure who may have tested it. Likely the product manager, aware of papers, added the note. I have found at least a couple sources, linked below. So you may just need to optimize the concentration and time, but there is certainly the possibility that not all bacterial types will label efficiently. Paper 1: https://www.cell.com/cell/pdfExtended/S0092-8674(14)00739-9 Article: https://exrna.org/extracellular-rna-bacteria/

Hi Jason A Kilgore I am trying staining the non-adherent cells (K562) with CellLight ER, RFP. I tried 30PPC and 50PPC to check the optimum concentration but the signal is really weak (~100 in a scale of 4095 with 20x lens and maximum camera gain). I was wondering whether the transfection process is less efficient with floating cells such as K562 or this is the normal level of signal? Can I get higher signal with higher concentration ~100PPC?

Mahmoud Atta K562 are hematopoeitic cells, correct? CellLight reagents don't work with hematopoeitic cell lines. You could try our BacMAM enhancer solution, B10107, but likely it wouldn't be much better.

Thanks Jason A Kilgore , Yes, it seems that K562 is a hematopoietic cell line. I am wondering why are CellLight reagents not suitable with those cells? I will check whether to go with the enhancer solution or try my experiment with another cell line such as HeLa cells.

Mahmoud Atta I'm not sure why BacMAM technology, like CellLights, doesn't work well with hematopoietic lines, but it is widely observed. Cheers.

Hi Jason A Kilgore

I have observed an interesting physics (New cellular heterogeneity) during observation of RNA extraction from single cells (staining with RNASelect). I interpreted the particle-like objects as mitochondria referring to the manual of RNASelect. However, the reviewer suggested to confirm that those particles are Mitochondria. I tried dual staining with RNASelect and MitoTracker deep red to confirm the co-localization but RNASelect signal is altered by the dual staining. How could you confirm that RNASelect signal from the cytosol was predominantly associated with Mitochondria?

Thanks,

Mahmoud

Mahmoud Atta It's not just mitochondria. There can be localized areas of RNA in the cytoplasm, as well as a diffuse RNA population. There is wide differences between cell types, too. You are already doing what I would suggest by colocalizing with a mitochondrial dye or fluorescent protein.

Thanks Jason A Kilgore , You are right. I reduced the concentration of MT dye (to the minimum level) and get better colocalization with sometimes extra bright dots in the cytosol in RNASelect images, which may be a localized areas of RNA in the cytoplasm but not mitochondria.

I too have double labeled RNASelect with Mitotracker (100nm) in human Neural Stem Cell lines and saw puncta spots INSIDE the mitochondria. I was wondering if those are true RNA signal or could it be also labeling mtDNA? Could you please advise kindly

It's expected to see some mito signal. We show an example of it, with the punctate spots, on the dye product page at Thermofisher.com, and we say in the manual " Maximal fluorescence is observed in the nucleoli, with faint fluorescence throughout the nucleus. Weak fluorescence is also seen throughout the cytoplasm, predominantly associated with mitochondria." We also state in step 1.4 (the step regarding optional methanol fixation) " With methanol fixation, the labeling pattern should be the same as with unfixed cells, except that the punctate mitochondrial signal disappears upon fixation." Since RNase and DNase washing controls are only possible with fixation, and the mito signal disappears with fixation, it isn't possible to confirm whether it is DNA or RNA. My suspicion, though, is that the bright punctate spots may actually be DNA, based on my prior mito DNA labeling with other dyes, and mito RNA is a more diffuse signal within the mito.

Thank you very much for your kind response Jason. Indeed I agree that likely the labeling is mitoDNA. Do you have any suggestions to reduce mtDNA labeling for live cell imaging?

Reducing the concentration and/or label time of the dye may reduce the signal, but I am doubtful it can be removed entirely.

If you were to fix and permeabilize the cells first, you may be able to use different RNA-selective dyes, though. If that is of interest to you, let me know and I can find options for you.

Yes, I would be grateful if you could suggest other RNA-selective dyes. Are there any you may know that also work for live cells?

Way back in 2003 I did a screen of all of the nucleic acid dyes that we had at Molecular Probes at the time, on live cells and on fixed, when I developed SYTO RNASelect Green. For fixed and permeabilized, I found the best were POPO-1 (violet), Ribogreen (green), TOTO-1 (yellow), Ethidium Homodimer-1 (red), and TOTO-3 (far red). See attached for recommended label time and concentrations. (the "XF" number on the images I reference is the catalog number for Omega Optical filter sets that I used. I also show the exposure time).

Also, see attached for a nucleic acid selectivity chart for SYTO RNASelect Green, determined in solution, not on cells. You'll see it labels DNA, too, just not nearly to the same extent (thus the "select" in the name, to reflect selectivity and not specificity). Unfortunately, there's no mitochondrial DNA test included.

Jason A Kilgore great info. Did you determine if any of those stains worked in live cells? Thanks!

Allyson McCune All of the dyes I mention in that gallery are NOT cell permeant, so you would have to fix and permeabilize to label. Other SYTO dyes would be recommended for live cell imaging, but we haven't characterized them very carefully for RNA selectivity, and likely wouldn't be as selective for RNA over DNA

as SYTO RNASelect Green.

Jason A Kilgore thanks! I’m just trying to stain extracellular vesicles. It’s not terribly important for the stain to be selective for either one. I’m more concerned about residual aggregates in the EV sample and stain leaking out of them after ultracentrifugation.