Hi all, I’m a grad student trying to adapt the v1 RNAScope protocol my lab uses for adult mouse DRG to whole fresh-frozen spinal cords from P0-2 and P7-10 mice. None of us have worked with postnatal mice before, but my PI thinks I should be able to section the lumbar cord to visualize both the DRG and dorsal horn neurons.
Despite copious troubleshooting, I often lose much of my tissue during the experimental process. The bulk of section loss seems to happen during the 2 hour hybridization step, but I’ve noticed it as early as the EtOh dehydration.
I’ve called the company and they suggested (1) increasing the fixation time and (2) baking slides at 37-60C for 30-60 minutes. I’ve previously gleaned from lurking in this forum that (3) using thinner sections (14uM instead of 20uM), or (4) increasing the drying time after sectioning might help as well. Other than experimenting with these parameters, I always follow the original ACD protocol for fresh-frozen tissue, including use of Superfrost Plus slides.
I’ve manipulated all four aspects described above in a few different ways.
I’ve increased the fixation time from 15 to 30 minutes, and then to 45 minutes. This alteration seemed to help the most.
I’ve tried adding a pre-protease baking step at 37C for 30 minutes, 48C for 30 minutes, 48C for 45 minutes, and 55C for 30 minutes. Baking helps, but it’s unclear if using more time or a higher temperature is significantly better than the minimum.
I’ve tried cutting 14uM sections once-this didn’t seem to help with retention at all.
I’ve increased drying time from 30 minutes to 1-3 hours. I typically end up with a range of drying times based on when I finish making the slides; 2-3 hours is seemingly ideal.
Even with all the different permutations I’ve tried, I always lose at least half my sections before the end. There are also many different variables that I could alter that it’s hard to tell exactly what exactly is the best thing to change now.
If someone already knows a good procedure for whole postnatal cords, I’d love to hear about it.
If not, I’d still appreciate advice on how to develop a better procedure on my own. All of these lost sections are wasting us money, and I’m wary of overdoing a step and losing the signal completely.
@cehanson Thanks for your post. I’m sorry about your troubles. I’m a bit stumped. I was going to say, fix the tissue (2 hours would be my starting point), cut thin sections (12-14 um) on Superfrost slides, dry (I usually don’t need more than 30 mins at RT) and then do the gentlest pre-treat (fresh frozen).
There must be something different about young mouse tissue compared to adult, but I’m not sure, to be honest.
I should start by saying that I primarily have experience with fixed-frozen spinal cord/DRG/brain, and only with the version 2 kit. However, I am almost positive that your tissues are underfixed, which causes them to fall off the slide. I had the exact same problem the first time I did RNAscope with larger pieces of brain tissue that were (unbeknownst to me at the time) underfixed. Drove myself absolutely crazy trying different protease treatments, baking, all that stuff, before I finally added an on-slide fixation step which solved everything. Based on the info you’ve provided here, I would recommend the following:
If possible, rather than fresh-freezing your tissue, perfuse the animal with 4% PFA then dissect out the cord. Postfix the harvested cord in 4% PFA for an additional 2 hours or so at room temperature, then transfer to 30% sucrose in (RNAse-free) PBS overnight at 4 degrees to cryoprotect before cutting your sections and mounting onto slides. Use the fresh-frozen RNAscope pre-treatment protocol except omit the on-slide fixation step and start instead with the EtOh dehydration step. It’s very unlikely that sections prepared this way will fall off the slides, but if they do, add the on-slide fixation step back in.
If you can’t perfuse the animal with PFA and must harvest the cord fresh, still drop the freshly harvested cord into 4% PFA for 2 hours at room temp, then treat the same as above (sucrose overnight, then cut, then fresh-frozen pretreatment starting at EtOh dehydration). With a very young animal like that, the cord is skinny enough that immersion fixation alone is probably sufficient if you can’t perfuse.
If you can’t perfuse with PFA or immersion fix after dissection (i.e., you must use fresh-frozen tissue), keep your sections 14 uM or thinner and try increasing the duration of your on-slide fixation step even more, and/or doing the on-slide fixation at room temp instead of 4C.
The way I see it, at this point your goal is just to get the sections to stay on the slides. If I were in your position, I would not even use a probe while you’re troubleshooting this: as you said, you’re just throwing away money until you can get the sections to stay on there. When you get to that part of the protocol, just throw some PBS on the slide and stick it in the hybridization oven as if you were hybridizing a probe, then see if you still have sections by the end of it. Once you get to the point where you feel confident you won’t lose sections, then worry about the other stuff, like strength of probe signal (in my experience you would have to WAY overfix to lose signal if you’re starting from fresh or PFA-perfused tissue).
Hope some of that is useful; let me know how it goes!
Wow just wow! Super helpful and thorough response @liz. I knew you’d likely have an answer. Hopefully this helps @cehanson get the results she needs. Thank you so much for contributing!
Just to clarify - you are sectioning a DRG, nerve root, and spinal cord all in one block, correct? If that is the case, then my thought is that your tissue mass it too large. I see this with mouse brains, if they do not lie completely flat (and if you get a bubble underneath the section) while you are sectioning OR if you are putting too many sections onto one slide (as in your adjacent section is laying on top of some of the OCT from your other section), they will fall off. I don’t think increasing the fixation past the recommended 15 minute fixation time is going to change anything. If you end up increasing the fixation, you may end up needing to do antigen retrieval and optimizing more steps.
Getting that entire portion to lay completely flat seems challenging. If one tiny piece of that tissue isn’t laying flat on your slide or isnt completely dried to your slide, it will pull away from the slide and continue to pull on the rest of the section the moment it starts flying around in liquid during dehydration/washes.
I’m guessing you already know to used CHARGED slides?
I use fresh frozen tissues all the time for RNAScope (v1 kit), cut at 20um and dried in the cryostat for at least 20 minutes. I always do a 15 minute fixation step as the protocol states. I find that a 1-2 minute Protease IV treatment is best; for all of the fresh frozen tissue types (mouse and human) that I have tried, this is the only time that consistently works. I don’t do any other optimization, no antigen retrieval - just the normal fresh frozen v1 protocol.
Over-digestion using Protease IV can destroy your tissue, which also may be part of your problem. If your postnatal tissues are sensitive to the Protease IV, try Protease III since its half strength and do a shorter time. But I’d start with Protease IV at 1-2minutes since that works great for my mouse and human RNAscope.
updating for posterity- despite the many suggestions, we weren’t able to determine a way to reliably preserve whole cord sections. The resulting pictures from confocal imaging aren’t nice enough to justify such an inefficient method.
Our solution has been to stop using sampling in the early postnatal stage, and wait until the pups are big enough for me to use the same dissection protocols as I would for adult mice. The pups’ smaller size makes it functionally impossible to get high quality DRG and cord from the same animal, but maybe I’ll get better in time. For now, I just plan to get one tissue type from each dissection.
@cehanson Thank you so much for sharing the conclusion of your efforts. I’m really sorry nothing ended up working. I’m still puzzled as to what is different about early postnatal tissue such that it won’t stick to slides. If your scientific question can be answered though by circumventing the issue, do that. But I feel like there has to be a way to get this to work (although maybe not worth the continued effort)
AIBS Spinal Cord atlas uses P4 mice. I went and looked at their methods
Doesn’t look so different than the things we discussed, but I guess it’s proof of principle
Anyway, good luck with your ongoing work. Keep us posted if you have any new developments.
