Hi Debosree,

Sense vs. antisense probes are good to run when working up an uncharacterized gene. However, I think that there's a misconception that the sense probe should be clean and give no signal. In my experience that may or may not be the case, it really just depends on the gene and whether things such as antisense transcription are involved. 

As far as designing in situ probes, I typically TA clone 1-1.5kb probes. I prefer targeting the 3' end of the gene and try to anchor at least 1/2 or 2/3 of that probe's length to be in the 3'UTR. This will typically give a more unique probe if you are looking at a gene that has paralogs (gene family members) in the genome you are working with. Again this is just a guide and some probes I wind up cloning other regions (full length cDNA, or 5' regions) that work best.

Hope that helps, best of luck with your experiments!

Ian

You should conclude that the probe is not working or that the transcription levels of your gene of interest are to low if your positive controls on consecutive slides are working. You should design the probe antisence to the mRNA. Probe size can very allot and can be between 50 to 2000bp. The bigger the probe the more signal per transcript you will have. Unfortunately bigger probes are more prone to not give a signal at all.

Hi,

If you have to make your probe "from scratch", the best way is to PCR a fragmen and clone it into any PCR cloning vector that contains Sp6 and T7 promoters (like pGEM-T). The framgent you may want to use as a probe should be based on cDNA sequence (without the introns!) or can be selected from a UTR region of the gene (3' usually is the longer one). UTR regions are very good for ISH probes as they are pretty much unique for a gene (they accumulate mutations). The length should be anything between 100 - 1000 base, I use ~500 if possible (longer probes are more specific but shorter ones penetrate better). Design your primers based on published genomic DNA sequence. When cloned into a plasmid, send for sequencing from Sp6 (or T7). When checking the orientation of your insert (PCR product will insert in a plasmid in a random orientation as it's not-directional), BLAST against mRNA. Your antisense probe must be a reverse complement to published mRNA sequence. Sense (the control) - logically - the same as published mRNA :) So, if your sequencing result shows plu/minus when done from Sp6 - use Sp6 to make (and label) an antisense (working) probe. If it comes out as plus/plus - use the oposite promoter (T7 in pGEM-T).

Sense probe should not give you any specific signal, may give some background. If you are unsure about your protocol for ISH - let me know and I'll send you mine (working well).

I hope it'll help!

Hi,

Designing a antisense probe would be a good idea. I would personally recommend a probe astarting from 200 bp to your mRNA. See if you can buy a full length sequence if you dont have it and then make the probe from 3' end if the sequence is not too long.  Run a positive and negative control to see if there is any problem with the protocol.

Hi,

She said she is making sense and antisense. You have either a bad probe design or morelikely a bad protocol for the in situ. If you post some images I can tell you if you have bad background problems. Is this a routine method in your lab? It can take a bit of work to get in situs working well and the method can vary depending on the developmental stage. Background can come from having repetitive elements within your probe, a top blast hit to your gene of interest does not tell you about where else it can bind. Background can also come from poor blocking and can happen at either the RNA probe or Ab steps. Please provide details of you protocol and reagents you are using. Your cloning statment is a bit odd, are you saying you are using genomic DNA to generate your probe? You need to use spliced mRNA converted into a cDNA clone. 

Just a comment: you don't need to use cDNA to make a probe as long as you are within one of the UTRs or, in general, within a coding sequence. However, using just single exons is not recommended as it may not be RNA specific.

Although almost every point has already been mentioned in the above answers, especially in the ones of Dorota,  Brian and Subrot, it will be good to summarize;

1- make sure you design from a cDNA (no introns)

2- make sure you avoid repeatative regions as much as possible

3- make sure you have a working probe as a positive control of the technique

4- make sure when you BLAST your sense probe, it does match with your gene's published mRNA, but nothing else.

5- Try to design probes for different regions of your gene

6- Try to keep the length of the probe in the range of 400-600 bases

7- Try to increase the hybridisation temperature if you see any background or the same pattern in antisense and sense of your stainings.

8- Try to increase the stringency of post-hybridisation washes

9- Try to increase the post-antibody washing time.

*** For Drosophila embryos I was washing them 6 times with 99,5% EtOH after 3 times post-development washes. This decreases background and enhances the contrast of the staining. I don't have experience with mice tissues. 

10- If the problem continues, I think this is the time you should check the sense strand for any unknown coding gene. 

Good luck! 

Alll of the above are true but your protocol could have problems as wel.  Your post indicates that you are doing whole mount in situ reactions. How long are you developing your embryos?  Developing should occur in the dark if using an AP developing technique.  Background can sometimes be reduced by including levamisole in the predeveloping washes and in the developer.  If your gene is expressed at really low levels, the addition of polyvinyl alcohol into the developing solution helps enhance a weak signal that may come across as background with a long developing time.  If you are developing for several days I recommend changing the developing solution every 24 hrs.  Additional considerations: do you have an ectodermal or mesodermal signal?  For whole mount applications if the probe does not penetrate into the embryo due to lightly digesting with proteinsae K you can also see high background.  If this is a technique new to you do you realize that a whole mount approach in the mouse is only really effective on embryos upto E11 otherwise the tissue is too dense and will usually turn color equally with either probe.  For embryos greater than E11 you should really do some partial subdissecting of the embryos to get a feel for expression in various organ compartments.  In the past I used to develop embryos in plastic nunc tubes which worked great.  However some manufacturers I found coat the inside of their tubes with something during processing that would fairly rapidly produce a blue blackground on the embryos regardless of probe direction.  Since that time I have always developed in glass vials.

Hope these ideas help.

Dear all,

Thank you very much for the suggestions. I am a first timer with this technique and as a matter of fact its the first time in our lab and I guess I need to consider a lot of things like everyone of you have mentioned.

I am seeing the location of a non-coding RNA, the probe sequence is present towards the 3'end and I have cloned it form the mRNA sequence. The RNA is an intronic one and dosen't have any exon-intron structure. The probe length is around 450bp and it dosen't show any repititve elements as seen by Repeat Masker. I have used Pax7 as a postive control with the same probe length as my experimental and I got quite a good pattern in 10.5dpc embryos with Pax7. 

One thing could be a low endogeneous level for my RNA as suggested by PCR. But, I will try to design multiple probes and try including levamisole and polyvinyl alcohol in the devloping mix to attain a better signal.

Hi Brian,

Please find attached the images and the detailed protocol. Hope you can help me with them.

Thanks

Hi,

You have a bit of background staining. Your protocol is very similar to the one I use, and it is good to see you ran a positive control. The embryos are getting pretty big to use for whole mount, I typicall stopped at E9.5 for whole mount staining. You should consider sectioning the embryos as others had suggested above. If you want to continue with the whole mount staining I would suggest this change.

I use a 3 x 30 minute wash at 70 degrees post probe hybridization, you have two. (Likely not a big deal)

For the blocking pre antibody I use a buffer called MAB, 100mM Maleic acid, 150 mM NaCl, 2mM Levamisole, 0.1% Tween-20, use NaOH to pH to 7.5, make fresh everytime it goes off after 12-24 hours, reacting with air. I then mix MAB buffer with 10% sheep serum and also add 2% blocking agent from Boehringer Mannheim (cat# 1 096 176, 124 DM ). After blocking for 2-3 hours I then, dilute the preabsorbed antibody in to MAB with 2% blocking and 1% sheep serum. Then continue as normal except all washes are done in MAB buffer not PBT, we also do an overnight wash in MAB at 4C (this adds another day to the protocol). Then you can wash into NTMT buffer and stain. I also use BM Purple from Boehringer as the stain. Check and stop the reaction as needed. 

Try to image against a white background. to get a better image of the expression pattern. 

Probe design is the next possible problem. I usually liked to use a minimum 250 bases and up to 1 kb. Other have made many comment on this part above. 

Best

Brian

Dear Brian,

Thanks a lot for your suggestions. Actually I tried a protocol by Nagy et al where they use MAB buffer but I didnt know that I should be makng it fresh everytime. Also, yes even I think I should go ahead with the sectioning. Thanks again!

We find the MAB to give far lower background, the overnight wash is an essential part as well. 

I have attached a protocol for RNA insitu on sections, just in case.

Best 

Brian

Thanks a lot Brian for your kind help! Is the hybridization buffer given in the protocol same as what you use for whole in situs also?

The ab hyb buffer is tris based but has levamisole like the other protocol. the thin sections are less prone to trapping that is common to whole mount. I never tried the MAB buffer on slides but it might work as well. 

Best

I think you should give the whole mount one more try, but with the MAB buffer, levamisole and the overnight wash. Run your control again as well. I think there could be real signal be it is hard to see against the high non-specific background.

Sure, got it! Thanks for all your help!