Free Floating ISH will be very expensive because you need al lot of reagents like probes etc.. Do you have a reliable protocol for your ISH? What Kind of instruments do you like to use for the free Floating ISH?

Dear Lena,

it may be a little tricky in the beginning; however, I had a rather good experience of doing ISH and especially FISH on free-floating adult zebrafish brain sections. You can find details of my protocol in the methods section of my paper (doi:10.1111/ejn.12692) here -> https://www.researchgate.net/publication/264942028_Emotions_and_motivated_behavior_converge_on_an_amygdala-like_structure_in_the_zebrafish

Good luck!

Jakob

Article Emotions and motivated behavior converge on a amygdala-like ...

Dear Lena,

It is really easier to make ISH on free floating sections (using a vibratome). You will have a nicer labelling and you will not use so much regaent. If you need more details, just ask

Best wishes

Nico

Dear Jakob,

thank you for pointing out your article. I went through the materials and methods section. I also went to Michael Levins paper, as you referring to it. What was the reason for you to choose a mixture of formaldehyde and glutaraldehyde? However, I think I will start with ISH before I get my hands on FISH. I will certainly get back to you then with more detailed questions.

Dear Nicolas, Dear Jakob and Dear Stefan,

yes I would like to have more detailed information. The following:

1) I assume that you fix the brain in 4% PFA. over night at 4 degrees. Do you dehydrate and rehydrate in Methanol and if so, what is the stepsize and incubation time? And is it possible to store the brain in Methanol?

2)I am planning to make these little baskets described in the Thisse protocol (basically a cut 10 ml syringe fused to mesh. They apparently fit into a 24-well multiwell plate. I do have a 65 degree water bath I can use. I plan to let the multiwell plate float in the water bath for the hybridization step. Do you think that works

3) What thickness of sections do you recommend? Nicolas, how do you embed the brain tissue? I cut the sections and collect them with a paint brush and transfer them into PBS, right? I subsequently need to digest the samples with Proteinase K, Jakob, you do half an hour, right? Then I postfix them with PFA and proceed. Now, that could give me a little trouble because PFA and the material of multiwell plates do not go well together. How do you get around that problem?

I think those are my questions for now. 

Thank you a lot for your time and input!

Dear Lena,

if you want to do classical ISH you do not need the protocol described in my paper. This particular protocol was designed for multicolor FISH on adult zebrafish brain sections and it took me about a year to establish it. In the end, however, it turned out that it was definitely worth it. Perhaps you want to give FISH a try?

Thanks for your question – it highlights a key point. The particular mixture of glutaraldehyde and formaldehyde that was inspired by Levine's paper was used in oder to be able to hybridize the probe on sections rather than on the whole brain in toto as it is commonly done. Hybridization of the probes directly on the sections rather than on the whole brain in toto greatly increases the signal and, consequently, the signal-to-noise ratio. – In classical ISH the brains are imbedded in glutaraldehyde, but glutaraldehyde has a very strong autofluorescence making FISH impossible. Formaldehyde, on the other hand, does not have this strong autofluorescence but it melts during the hybridization procedure at 65°C and thus you subsequently loose your sections. Hence, I developed a specific mixture of glutaraldehyde and formaldehyde that allowed my to stabilize the sections sufficiently and eliminate autofluorescence.

Concerning your other questions.

1) Like you said; yes, dehydrate and rehydrate; 5 minutes in 30%; 50%; 70%; 90% methanol in PBST; 2 x 100% methanol; store in 100% methanol at –20°C for maximum 3 months (may be even longer).

2) I never used baskets, but used a 48–well plate; all steps done very carefully by hand, no automatization. But should be possible, I guess.

3) I cut at 50 µm, but it depends on your needs and what you are investigating. I cut in PBS and transferred with paint brush into PBST – 2 sections per well. Wash again in 
PBST. Permeabilize 30 minutes with 10 µg/µl Proteinase K, yes. Wash 2 times in PBST to get rid of the the Proteinase, then refix with 4% formaldehyde for 20 min at room temperature (without shaking). Wash 3 x 5 minutes (or longer) in PBST (with shaking) … Did not have any problems with PFA and material of the well plates.

Good luck!

Jakob

Dear Jakob,

(Salut Nico ;))

Thank you for the details about your FISh protocol, that I also would like to give a try. In your paper there are some points I would like to clarify with you:

- The 2% H2O2 incubation is described just after the dehydratation and before the rehydratation. Is that correct? I suppose the H2O2 solution is diluted in water? Or is it diluted in 100% Methanol?

- After sectioning and during the prehybridation at 65°C, did you encounter problems due to the melting of the surrouding agarose. Like slices destroyed very easily, sticking on the surface  or background due to the melted agarose from the surrounding? In what type of container are you doing the pre/hybridation (plates, tubes) and what are the volume you are using?

-I used actually protK at 10µg/ml. So it means that you use the ProtK at the concentration of the stock solution (10mg/ml) is that correct?

-Do you include Dextran Sulfate in the hybridization buffer to increase the sensitivity?

 Thanx

Olivier

Dear Oliver,

Thanks for your questions.

1.1 Remove brains from −20 °C and wash once in 100 % methanol at room temperature

1.2 Incubate brains in 2% H2O2 for 35 min in 100% methanol

2. The brains are not imbedded in agarose, since the agarose is melting at 65°C – at least in my hands. Instead, they are imbedded in a specific bovine serum albumin/gelatin type A 
mixture that is stabilized through 6.47% of formaldehyde and 0.075% glutaraldehyde. – This specific type of embedding medium and the process of how the brains are imbedded in it is the key to success of FISH on adult zebrafish brain sections (see also Lena's question). As said before, it took me about a year to play with it before getting it right.

2.1 Use 48-well plates (see above); not more than 2 sections per well. Like this you can fit about 1 telencephalon (0,8 – 1 mm) per row, i.e. 6 telencephalii per plate. Use 200 µl per well for pre– and hybridization solution.

3. You are right: it's 10 µg/ml not 10 µg/µl.

4. Yes, use dextran sulfate in hybridization buffer and to reveal the probe(s).

You are welcome.

Jakob

Pardon - Olivier!

Merci Jakob for all the détails!! This is indeed a lot of work from your side, and the idea of albumin/gelatin stabilized is excellent!

Avec plaisir. Initially, it was intended to publish also a methods paper on it. However, time was short and, unfortunately, I have not been given a chance to do that (yet).

The whole protocol requires some pratice to get it right. – Perhaps you could say hello to Masanari from me if he's still around in Karlsruhe? Merci.

Merci Jakob,

I will give it a shot and get back to you in case I am facing troubles.

Just one more clarification and two more questions.

First the clarification: By loosing the sections you mean that they disintegrate and not that they detach from the glass, right?

Second the questions (that directly relates to the previous):

1) You mount the slices after developing the stain, right? 

2) Do you place the multiwell plate in a waterbath or in an oven for the hybridization step?

Dear Lena,

some general remarks: I would strongly recommend that you take a practical approach. This means that you let yourself guide by the scientific questions that you wish to answer and then decide on the methodological approach that lets yourself solve these questions in the most promising and easiest way. If you do not need FISH to answer your specific question then use a classical ISH. (Personally, I had no other choice than to develop a new FISH protocol on adult zebrafish brain sections to solve the questions I was facing, but before I thought and tried all possible ways to get around just that.) There exists a number of good protocols for classical ISH on adult zebrafish brain sections; for example this one: Kuhn, E., and Koster, R.W. (2010). Analysis of Gene Expression by In Situ Hybridization on Adult Zebrafish Brain Sections. Cold Spring Harb Protoc; 2010, doi:10.1101/pdb.prot5382 – nevertheless, be aware that in the above cited protocol hybridization is done as usually on the whole brain in toto rather than on the sections themselves, i.e. a procedure that I cannot recommend, especially for a sensitive detection of only weakly expressed probes. Again, the advantage of FISH is that it is much more versatile and allows for three-dimensional quantitative co-localization analyses of the results among other things that are not possible with classical ISH. However, it is also slightly more complicated, and if you do not necessarily need these kind of analyses you may be better off if you first start with classical ISH.

After these general remarks now to your specific questions …

By "loosing sections" I mean that – after the hybridization over night at 65°C – the brain tissue is now longer embedded in the embedding medium; as a consequence the quality of your sections is no good and, in the subsequent steps (washing et cetera), you loose your sections by sucking them up with a pipette, or they adhere to the walls of you wells, or a multitude of other problems occur (everything you can and cannot think of).

1) Once the staining is complete I usually leave the sections in PBST at 4°C over 1–2 days. This is done to have a more beautiful staining with less background. Then the sections are mounted on microscope slides and imaged via confocal microscopy.

2) The 48–well plate(s) are placed in a waterbath (but not directly in the water) for the hybridization, yes (and a number of the subsequent washing steps)

Dear Jakob,

at the moment I just want to look at three specific isoforms of the gene I am working on. The problem is, that my antibody does not distinguish the three isoforms, because it is a polyclonal antibody generated using the full length protein. So I wanted to see that if I use in situ probes against the differing regions of the mRNA I can get a more clear picture. Though in the long run it will be worth doing FISH with the antibody I have and the three different in situ probes to see whether all of them are actually translated. 

Dear Lena,

the application of FISH to solve the specific type of problem that you are describing may indeed be useful if you manage to generate probes that are specific of the three different isoforms of your gene of interest. In principle, this problem could be solved in a single experiment: you do a triple FISH for the isoforms A, B, C and combine it with your antibody D (and perhaps a DAPI (E) counterstain). Your can then do a quantitative coexpression analysis of the probes A, B, C together with your antibody D to determine which of the isoforms are actually translated; however, the expression dynamics of your gene of interest on the RNA and protein level may be differentially regulated.

Dear Jakob,

thanks a lot for your input! Its really motivating to have someone discussing thoughts and ideas at times! So, let me get this right, triple FISH plus AB plus DAPI. 

That means for example the following dye combinations:

Alexa Fluor 488 (492/520 nm), Alexa Fluor 555 (555/565 nm), Alexa Fluor 594 (590/615 nm), for the FISH, we have a Dylight 649 that I can use for antibody staining and DAPI.

I will need to check with our microscopes whether we have the matching filters for that. 

Do you have a specific FISH kit which you recommend?

I guess the last point you mention about the RNA and the protein points to the striking difference of Cfos protein and cfos mRNA expression that you observe in the zebrafish telencephalon. 

Dear Lena,

here's what worked for me: Antibody D (405 nm); FLUO (488 nm; probe A [weaker probe]); TAMRA (543–568 nm; probe B [weak to middle probe]); Cy5 (633-647 nm; probe C [strong probe])

or

DAPI (405 nm); FLUO (488 nm; probe A [weaker probe]); TAMRA (543–568 nm; probe B [weak to middle probe]); Cy5 (633-647 nm; probe C [strong probe])

Thus, you will have to use either the antibody D or the DAPI counterstain (and what I said above was therefore not entirely correct, unless you manage to use a different combination of excitation lasers and filters).

FLUO and TAMRA are tyramide-conjugated and custom-made (e.g. http://wiki.xenbase.org/xenwiki/index.php/Flourescin_Tyramide_Synthesis); Cy5 congugated tyramide is available from PerkinElmer. To my knowledge there is no FISH kit – but perhaps there is? I never used one, but instead set up my own protocol.

Yes, since cfos is an immediate early gene, its mRNA and protein dynamics are differentially regulated, see e.g. Clayton, D.F. (2000). The genomic action potential. Neurobiol Learn Mem 74, 185–216.

Salut Jakob,

Thermo Scientific offers this one. I don't know, which of the two options (the one you describe or the Thermo Scientific kit) is the cheaper solution. How much did you roughly spent for the custom-made probes?

https://www.thermofisher.com/order/catalog/product/F32956

Dear Lena,

you could try the kit that you mentioned; however, did you check whether it uses highly sensitive tyramide signal amplification?

Custom-made tyramide-conjugated fluorophores should be a whole lot cheaper than any kit, and you'll have enough tyramide to "wholemount a cow" (see http://wiki.xenbase.org/xenwiki/index.php/Flourescin_Tyramide_Synthesis and Lauter, G., Söll, I., and Hauptmann, G. (2011). Multicolor fluorescent in situ hybridization to define abutting and overlapping gene expression in the embryonic zebrafish brain. Neural Dev 6, 10.)

Since a technician in the lab synthesized the tyramide-conjugated fluorophores (FITC and TAMRA), I have no overview of these costs.

Article Multicolor fluorescent in situ hybridization to define abutt...

Dear Lena,

out of curiosity - how did it go with your FISH experiments?