about LCM

[johnmay]

Hello all,
Recently I am set to perform LCM (laser capture microdissection) on PFA-fixed midbrain sections (VTA) of TH-GPF transgenic mice and encountered some problems. The quantity and quality of RNA isolated from LCM cells were too poor to do the downstream analysis like real-time PCR. Because the GFP expression in the cells of interest can help visualize the cells, I don’t have to perform a staining step. Also, I didn’t do the dehydration step since the alcohol could quench the GFP signal remarkably. Then it was difficult for us to capture a single cell using IR laser on LCM machine, what we can do was just to use UV laser to cut the cell patches, which was not we want to do.
(1) Is the dehydration step very important for cell capture? If yes, how can I perform this step? (the literatures refers the graded ethanols and xylene, but ethanol could quench my GPF signal)
(2) How many cells do I have to collect for downstream analysis like real-time PCR?
(3) What kind of slide is better for single cell capture by IR laser? Glass or membrane? And the thickness of the section? I use Arcturus PicoPure RNA Isolation Kit (#KIT0204) to isolate RNA from LCM cells, is it better to add proteinase K to the Extraction Buffer to help extraction since the proteinase K could help extract RNA by reversing the extensive crosslinking network of aldehyde?

Any reply will be highly appreciated,
Zhong

[CrowSan]

Hi - we do LCM quite a lot using a Palm microscope from Zeiss. For RNA I would go with a membrane slide (if your LCM is a 2 laser system - one laser cuts and the second "pings" the sample into the collection device). For membrane slides using the above system (rather than an ablation isolation slide/system which will "fry" your RNA) I would recommend that the samples are dry and fixed (as moisture on the membrane will reduce laser cutting efficency). If you want RNA and do not want to use ethanol then you could try "RNAlater" from Ambion (I think). This gives similar results in qRT-PCR similar to ethanol fixation (and is designed to preserve RNA). If for some reason this does not work then use 4% PFA (made fresh from powder - NOT formaldeyde solution which is stabilised with methanol). The slides we use are from MMI (molecular machines and industries).
This is how we do it:
1) Seed cells on the membrane and allow to adhere (4 hrs or overnight)
2) Remove media, rinse 1x PBS and add RNAlater to slide and place at 4oC overnight (or shorter).
3) Remove RNA later and rinse slide in H20 (PBS crystals are often a probelm in cell isolation). Air dry slide.
4) Place in LCM and cut cells.

We can get sufficient RNA from a single cell to run mutliple rt-PCR reactions. However if I was you I would start with a higher number alongside i.e. cut 50 cells, 10 cells, 10x 1 cell and compare the RNA/results etc.
When cutting around the cells leave a space (~5 microns or the radius of the cell) to avoid cooking the RNA with the laser.
We do not add proteinase k.
Hope this helps.
CrSn

[CrowSan]

Hi - I just re-read what you wrote and realised I maybe did not answer your question...(I was thinking about single cells rather than tissue sections). As the tissue is already PFA fixed then it may be that (depending upon storage) you might not get any decent RNA out of a single cell - unless during embedding/fixing care was taken to try to preserve RNA then it may already be too degraded for RNA analysis.
(this paper discusses best fixatives for tissues for later RNA extraction: http://onlinelibrary.wiley.com/doi/10.1 ... 0/abstract)
I am guessing that the section is blocked in wax? If so then you will probably have to remove this from the section prior to isolating the cells. Normally the section would have been dehydrated etc prior to embedding in wax so really the reverse if this is required (re-hydration). As during wax embedding the tissue would have been subject to graded ethanol baths (70 - 100% ETOH) then I am suprised there is still GFP signal there to be honest.
To re-hydrate a section I would normally place a section on a slide into heated Histoclear (a non-toxic xylene substitute) to remove the wax then re-hydrate (90% ETOH, 70% ETOH, 50% ETOH, H2O) at 2-5 min per bath.
The heating steps above may prove problimatical if using a membrane slide (as the membrane may melt).
As regards section sizes - as the wax will be removed during the clearing (above) then this should not affect cutting BUT bear in mind that cells (rounded) range in size from around 10 microns to 20-25 microns. This means that sections less than 30 microns will probably not contain very many intact cells but instead will be cell "slices".
At best and remaining RNA in your fixed tissues will probably be <700 bp (DNA which is "tougher" is usually reduced to 500-700bp fragments by fixation/dehydration/embedding).
In brief: The approach you are taking is unlikely to work unless the samples were fixed and blocked in wax following a protocol designed to preserve RNA integrity (see paper ref above which fixes in methacam which would probably destroy any GFP signal anyway).
What you could do is try to see if you can get a protocol that gives at least some info: (assuming the sections are in wax and you have an LCM with seperate cell cutting and "pinging").
i.e: Cut 50 micron sections. Place on membrane slide. Isolate 100 cells, 10 cells and 10x 1 cell directly into seperate collection devices (tube or lid). Place RNA lysis buffer (with no enzyme in) into tubes and heat to 70oC (to melt residual wax in the sections) and isolate RNA using desired columns etc. Perform rt (reverse transcription/pre-amp) PCR using control primers only (not sample) with the primers for the gene (such as 28S) being no more than 200 -300 bp apart.
This will tell you a rough limit of detection and if this has any chance of working.
If any of the assumptions I have made are wrong then let me know.

[CrowSan]

Me again - I just thought that maybe the tissue you are using has been snap frozen rather than fixed.
This chapter (found online) is using a micro-beam LCM and frozen tissue to isolate RNA
http://www.abdn.ac.uk/ims/microscopy/do ... ter_13.pdf

CrSn

[LeserattePD]

Dear JohnMay,

As far as I understand you want to look at gene expression in EGFP expressing cells from FFPE (formalin-fixed, paraffin-embedded) animal tissue. Is there a particular reason why this tissue has to be FFPE? It is really not easy to get good quality RNA out of FFPE? Please remember not to rely on oligo dT primers for the reverse transcription step if you get good enough RNA out of FFPE tissue (formaldehyde modifies the adenine in poly-A tails and therefore inhibits binding of oligo dT primers).

I'm asking because with EGFP expressing cells, it would be quite easy to take fresh tissue, dissociate it and then FACS sort using the EGFP as marker directly into RNAprotect Cell Reagent (Qiagen). That way you should get a high number of cells = lots of RNA too!

Just a suggestion..

[Manuora]

Hello,
sorry, I don't have any advices to submit but a quite similar problem and need help.
I try to knock down a gene in vivo by stereotaxic injection of lentiviral vector in rat brains. Then, I use LCM to select
the injected region and validate the inhibition of expression by qPCR. The vector used express eGFP and after cutting
I actually see fluorescence. However, as I use fresh tissu sections of brain, a drying step is necessary for LCM.
Is there any other way to dry tissues besides ethanol which quench the GFP signal?

Thank you for any kind of help.

Laura

[CrowSan]

Hmm - there is none that readily spring to mind. However you could try one of two things:

a) Grow the cells on the LCM slide (may need to coat with proteins etc).
Remove the media, rinse PBS and fix in 4% PFA for 10 min. Remove PFA and air-dry the slide. PFA should preserve fluoresence of GFP.
b) This way is a bit of a "cheat" and is based upon the fact that most LCM's have to know precisely where the gates (or elements) are set on a slide. Grow cells on the slide. Remove all but a thin layer of media from the slide. Carefully position the slide in the LCM. Identify positive cells and draw a "gate/element" around the cells. Remove the slide. Remove media, fix in EtOH, remove ETOH and allow to air-dry. Place slide back on the LCM in the same orientation as before. The gates should still bea round the cells originally identified even if they are no longer glowing. Cut and keep.

Not sure if the above will work for you but it's probably worth a try. i know from personal experience on our Palm LCM that I can take a slide I used a week before, re-import the gates I had drawn and saved (elements) and they still match the holes left by the cells I had cut out a week before. Impressive little thing.
CrSn.