What kind of excitation source do you use ?

Looking at your picture it seems that the entire region in the middle is affected. I encountered something more or less similar in the past, doing live imaging using a UV light source, and long exposure affected the cells integrity (several seconds exposition). Those cells died rapidely. It is possible that you might somehow have killed the cells with a too long exposition and they became leaky for what looked like a segregated GFP expression in the sourounding intact cells. Does this "leaky" fluorescent signal remains after a few minutes or does it "dissolve" away in the surrounding media ?

Was the requiered exposition time the same with both the old and the new filter ? Is it possible that when using the new filter (at a later time point in your experiment) the "natural" expression level of your green fluorescent probe was lower than in the early measurements, thus requiering longer exposition, resulting in cells destruction ?

Did you notice a change in morphology (in bright field) before exposition with the new filter and after exposition when you start to observe this leaky green signal ?

Dear Caspar,

to me this looks like you rather have trouble with your mounting media (or storage conditions of the sample), than with your filter settings. So how did you store your samples? Which mounting media did you use and how old is it? Did you carefully remove all aqueous liquids from your sample before adding the mounting media? Did you observe this before in any sample?

So I think, your mounting media does not function as a radical scevenger after the storrage, either due to wrong storage of your sample, to much water in the mounted sample, incomplete sealing etc...

Usually it is possible to store the samples at 4°C in the dark for at least some weeks.

Hope that helps, best,

GAD

Dear Frederic,

that were fixed samples, thus things happening in live cell experiments like changing the morphology is not really possible...

and when you read the question carefully, this also happend after storage of the samples with the old filter set...

Guido,

You're right... I skipped the fixed sample part when I read the question... My mistake !

Mea culpa maxima.

Make sure that you view before/after images on same fixed intensity scale for pixels. Commercial software will default to autoscale - it may be that you have simply bleached the dye and so towards the end of your acquisition you start to see the more persistent autofluorescence, which may appear to be more problematic than it is simply do to autoscaling, so check the absolute raw values of the pixel intensities first before anything else.

bw

m

Frederic,

Can you please clarify the nature of your secondary 488 antibody? Also, which mounting medium did you use?

Hi Caspar,

This is a weird one alright, I don't have the answer for you unfortunately. However, can you confirm which fluorophores you are using in the kit (or which kit)? My guess is it's probably one or more of the reagents?

Again, sorry for being no help, but the additional info might be a help,

Hope you get it sorted,

Omar

You might want to check your immersion oil. Some of them tend to auto-fluoresce. I have seen autofluorescent spots when mixing two immersion oils (old one left on a slide with new one added to it).

Hi Caspar- I do not know for certain, but I have observed the same phenomenon in my own work. In my case, it occurred when immersion oil came into contact with the anti-fade mounting medium. We use a "homemade" anti-fade that employs DABCO as free radical scavenger. When this happens, if I observe an area for a little while it steadily acquires a green fluorescence. I have not observed it in the red or blue channels. If this is the issue you encountered, then simply by scrupulously avoiding any contact of your mounting medium with immersion oil should solve the problem. Sorry I cannot give a more concrete suggestion.

Hi Caspar,

Have you tried the sample on another scope? That would at least restrict the problem to either the slide or the scope. I'd also check alignment and centering and the field stop for the fluorescent light; the fact that both of cubes had been moved and then the problem showed up still leaves the possibility of a mechanical issue.

Photochemistry there is very complex and depends on many factors, solutions and storage; there are layers of antibodies that are bound, degrade and re-bound again,

one possible way is to fix (to cross-link again) after the last antibodies (and hope that cross-linking would not create light sensitive pockets,

- other way - is to store freshly obtained digital data

Dear Caspar, photodegradation of the primarily red fluorescent fluorochome to a smaller, still fluorescent molecule may be the reason for your observation. As the resulting fluorochrome is smaller it emits fluorescence with a shorter wavelength, i.e. green fluorescence. I have observed this with other dyes, especially when they were photoactive like photosensitizers for photodynamic therapie. You can avoid this phenomenon by decreasing the exciting light intensity (filter or lower excitation power), while increasing the detector gain and shortening the excitation time, which, of course are general rules to diminish photochemical effects. If you work with fixed settings (i.e. no autoamplification of the detected fluorescence), I think the appearance of the green fluorescence is not due to autofluorescence that becomes more visible, when the red fluorescence is bleached. Hope this helps. Best regards, Frank.

Hi Caspar! I have experience only with plant cells. I encountered similar (but not so intense) phenomena when my samples were not properly fixed and stored for a relatively long time. Can you check if your fixation procedure is OK?

However I agree with everybody who wrote that photobleaching can be a problem, too.

Hi, and thanks all. Impressive with so many good replies overnight.

Vinod and Csaba: We receive the FFPE tissue from our pathological institute, where they allready fixed them using machine.. Thus the fixing should not be problematic... But it is never good to just assume that they perform an excellent fixing, so I'll just try to make a diplomatic question regarding the procedures of fixing :-)

Guido: Hmm... That might be some of the problem... I do store them in the dark at 4 degree celcius... But when I looked a bit closer, I saw that the sealant was not that good.... Definately something I'll correct next time :)

Mark: Oh I do know the troubles with automatic functions of the software... So I allready deactivated most of the automatic functions, so I could keep everything constant. But definately a good suggestion :)

Omar: I used Zytovision's "Zytolight SPEC HER2/CEN17 Dual Colour probe", where the ERBB2 (Her2) probe is labelled with ZyGreen (close to FITC) and CEN17 with ZyOrange (close to rhodamine). For the FISH itself I used DAKO's "Histolofy FISH accessory kit".

Adam: I was thinking the exact same, So I tried to look at some immersion oil alone, and there I did not see any fluorescence at all.

David: That is a good suggestion. As I also wrote to Guido, the sealant might have been poor, so I'll keep that in mind next time.

Jitte: I did not write it, so you did not miss it :) I'm using DAPI as a counterstain. I have noticed earlier that DAPI tend to aquire a green fluorescence if to high a concentration is used or in other conditions. But I would imagine that I just would see a strong green fluorescence of the nuclei, rather than all over the tissue? But thanks :)

Fabienne: Well that do sound familiar... Perhaps a combination of immersion oil spoiling the mouting media and then making it more susceptible to photobleaching... At least I can say that in the future I need to be more carefull (less slobby, it was late in the evening :-/ ) with the sealing, take the images the same day or the following day and be more carefull with the time under the fluorescent light... Although it was maximun 2-3 minutes totally....

Kira: Oh I would love to be able to do that. However, while we have several light microscope, we only have a single epifluorescence microscope... My assignment at present is to open the eyes of the people here in the building, to the wonders of fluorescent microscopy :-)

Mechanical failure... Well I wouldn't rule that completely out... I wouldn't hurt to give it a little fast service :)

Irina: Sorry, I should have mentioned that... I do not use antibodies for fluorescence... The FISH probes are directly labelled with fluorophores.

Frank: Interesting, I did not know that :-) I do work with fixed seting, since I hate not having the complete control of the microscopy.

Thanks for all the suggestions everybody :)

Best

Caspar

Hi Casper, I agree with David about the mounting medium. Do you seal your samples? Some mounting medium that has anti-fade will have an increase in the background fluorescence if not completely sealed from air. Also DABCO and other similar mounting mediums that need to polymerize first before being sealed get funky if they are not fully polymerized. Slowfade gold, which does not polymerize could be a short-term use solution, but samples are only good for I month. I used to save all my fixed samples (for years) but in general they will never look as good after the first couple of days.

I did use a coverslip sealant that was supplied in the kit. However, when looking at that following a few days, I saw that it did not seal proberly... So I think next time I'll go back to using the old fashioned nail polish :-) Then We'll see if the same issue comes up...

Hi Casper,

sorry to hear that you have a problem. At first it sounds like a filter problem. Switching the filters apparently did not solve the problem. What happens if using a different microscope, probably the same? As for the mounting medium I would use prolong-gold (ProLong ® Antifade Kit (P7481) from Moelcular Probes) which hardens. Advantage: You are reducing the possibility of oil sipping under your cover slip. Now it's easy to seal the coverslip with nail polish (if you want to be 100% sure that nothing gets to your sample).

I still have slides (stored at -20C) I prepared 5 years ago and the signals are still great. If you go for this option, be aware that you will have compression to some degree. As for conjugates I strongly believe in Alexa and Cy. They are more stable.

Hope that helps.

Ludger- You mentioned "compression" of your samples. I am curious exactly what you mean. I have wanted to try a permanent mounting medium but due to budget issues have stuck with a homemade DABCO mix. It works pretty well as an anti-fade but does not allow long-term storage. If you could clarify what the "compression" issue refers to I'd appreciate it. Thanks.

David, to address your concern: Well, we have the "luxury" to look at things in 3D, meaning that we take z-stacks, reconstruct, measure the position and intensity of every signal within the nucleus. We also did that with SIM (structural illumination microscopy) and noticed a certain flattening of the nuclei. Since it really does not matter at 60x I stuck to the mounting medium. Probably you will not notice any difference when looking at sections. As for the mounting medium I would never try to save a penny. In the end you pay more since you have to repeat the experiment and you end up purchasing the new mounting medium anyways. We stopped using the "home-made brew" a long time ago. Check around who has commercially available mounting medium, best is to go straight to Prolong Gold. You will love it. PS: Be aware that it is challenging to get the coverslip off. Here you will have to place the slide in buffer for quite some time. I later place them in 75C PBS since it does not affect our experiment at all. Actually it helps to get rid of the probes when re-gybing is done.

PS: As an alternative you can use Vectashield (Vector Laboratories, Inc. Burlingame, CA). If you are using DAPI you might find that the DAPI concentration is too high. Therefore we perform the DAPI stain independently, rinse off excess DAPI and mount with DAPI-free Vectashield (cat# H-1000).

Ludger- Thanks for the heads-up about the distortion. We capture z-stacks from time to time as well, and do keep our samples for up to a couple months at 4C, but have never attempted -20C storage. Perhaps the compression is due to shrinkage of the Prolong as it polymerizes? We're viewing very small cells so the compression would probably not be noticeable anyway, but good to be aware of known artifacts. Interesting that it would be noticeable in the z but not the x-y dimensions. I have not tried Vectashield but recently assistend another PI whose slides had been mounted with it, and I agree- the DAPI is indeed way too high a concentration. Thanks for the information.

David, you are most welcome. And YES, you are right that the compression is due to the drying process. However, if you are looking at small cells it probably might not really show. To put your mind at ease I would try both, Vectorshield and Prolong Gold and see what happens. If the a/c-ratio is unaffected then you are fine.

I have observed a broad band, non-specific fluorescence from tissue that has been fixed. This signal was not found in fresh, unprocessed tissue samples. We thought that its origin was the formation of cross-links especially in the structural matrix. Light might have similar effects due to photochemistry that takes place.

Well, I finally got the time to redo the experiment. And sure enough, same thing happened.

This time I was way more carefull with sealing the sides of the coverslip, so I'm positive that immersion oil did not leak into the tissue...

So I'm starting to consider the idea, of photobleaching - like Fabienne described.I'm just wondering how normal it is, that the photobleaching happens in a matter of seconds... I could understand if I have exposed the tissue to light for several minutes, but less than 10 seconds?

I have talked with others that have used the same setup, and sure enough, they experienced the same once in a while.... So I'm wondering if the probe itself or something in the kit is making the tissue more susceptible to light? I wouldn't think so, since it's a commercial kit for diagnostics.... Or if mammatissue or glandular tissue in generel just tend to be more susceptible.

Best

Caspar

Hi Caspar,

sounds to me, that the fluorophore you are using got old/unstable. If possible you should try a new one or an aliquot from a collegue (which is working).

Best,

Guido

Hi Guido.

Thanks for the answer... I do not hope that won't be the problem... I had the problem with the old probe cocktail. So I bought a new one a week ago, but still same problem.... The DAPI-mounting medium is two month old, but has been stored at -20 C... So I cannot imagine that it would give a problem.

Best

Caspar

it looks that your cells are autofluorescent strongly. or the method you use creates an autofluo material. some oxidization (during storage) should have to occur before it becames detectable, that is why you see after a week (or the Dapi excitation sometimes generates it for me).

i do not think it is a problem. you can image your samples next after the experiment. if it is problem, change the green stain to a higer excitation one being compatible your microscope if you have another possible channel. good luck.

I'm not sure what you described here. As I interpret what you see is an increase of fluorescence, or newly formed fuorescence substance, in the green channel. With photobleaching the fluorescence disappears and it does not increase. So, if you have seen an increase of fluorescence you can explain that by fluorescent dye getting dissolved in solution and deaggregating and thereby its fluorescent quantum yields increases. Such processes can be activated by light as well.

Hmm... That does sound interesting. It is true that I see an increasing fluorescens., so that could be a valid reason.

I have been in contact with the companies and neither have been able to explain the reason behind (besides the usual statement that no problem would be experienced if only we would purely use their reagents). Since it is a bit to expensive to continue trying to solve the problem, we have desided to use a complete kit from a single manufacturer... But I am of course still curious...

The individual fluorophores is incoorporated in ish DNA probes. The probes are commercial probes, so they should not loosen by themselves.. Perhaps a reagent in the ish solutions - comes from another manufacturer - interacts with the probes somewhat... Do you have any ideas how to test this?

Best

Caspar

My field is in vivo/human photodynamic therapy, where photosensitizers with complex structures constantly show aggregation and deaggregation with accompanying loss or increase in fluorescence levels. I am not so well known with the probes you use and the specific details. But it seems a light activated process. Can you reduce the illumination time ? or the fluence rate of the laser ?

Well, problems starts to occur within 1-2 seconds, and much of the settings are manually. So I do not believe that it will be posible to reduce illumination time.

I am able to reduce the intensity of light hitting the tissue... I have tried that and it does significantly reduce the problem.

However, for the diagnostic application, I would have to move around in the tissue and count manually number of signals. Thus, I will have to have a longer illumination time, so for the practical applications, this is just not feasible.

But thanks. It sounds like a resonable explanation, which would settle my mind :)

A lot of dyes stay dissolved in typical solvents such as dmso/alcohol ect. can you add just befor measuring a few drops of some kind of solvent to see if you can prevend the effect from happening.

Did you used anti-fade mounting medium such as Mowioll ? It helps to avoid photobleaching. using this kind of product may allow you to keep your slide for one month at 4°C