This seems to have generated some variety of responses. I forwarded an e-mail to Dr. Lyon with an article showing 3-5 weeks in formalin (not 5-7 I had quickly mis-read) with reduced apoptotic signal in ISEL (I let the subtle difference between that and TUNEL slide). Is The Histochemical Journal vol. 27, number 12, December 1995 pp 983-988 F.D.Davison, et al. I actually did find a few articles dealing with DNA strand breaks and formaldehyde but all I believe are outside the realm and having little to do with this discussion.
Again, my experience is that formalin fixation is good for TUNEL. Of course the devil is in the details and I hear about non-reactivity, background, false positives and I do not deny those things can occur but I believe there are ways to avoid them. Just doing TUNEL out of a kit is tantamount to getting an all-prediluted, IHC kit. It can work but then you start running into problems and you are stuck with the directions of the kit. Depending on my model, tissue, fixation I certainly amend the TUNEL protocol to fit my needs. Has little to do (I hope) with any superstitious allusion. You can certainly, and I have, altered the amount of Tdt in the mix. They use way to much in kit protocol but they want it to work, FOR SURE, and also sell a lot of expensive Tdt so really pour it on. Don't need that much. Tdt is template independent DNA polymerase so I vary time, I think they state way too high but again, they absolutely want staining there of some sort. Mix concentration I var
y. Any who have done PCR knows the wild results you can get varying [Mg++]. Proteinase K, certainly there are other ways to get after the nucleoproteins (histones, etc) that will block the DNA enzyme from creating a longer strand. Someone mentioned citrate heat, its great. So I don't just use "a TUNEL kit". It can stain some stuff fine as is, but with a bit of ingenuity and thinking, always have been able to modify pretreatment, time, Tdt conc, Mg++ conc, to fit my model, tissue and type and time of fixation. My bet is that the drop off the authors saw in their study was due to the extended fixation of formalin affecting the nucleoproteins but them maybe not compensating by extending their pre-treatment. Just a guess.
Someone mentioned survivin - great. Many mentioned cleaved caspace-3 and that is a fantastic antibody. With the giant caveat that it is well known that not all types of apoptosis go through the canonical caspace cascade. So depending on your model- you can miss apoptosis completely relying solely on caspace-3. Is great but not the total gold standard.
I still believe TUNEL to be a fantastic tool which if used properly and with proper controls and variance of procedure, can lead to interpretable, no background results with with little worry of false positive and false negatives, IF you set things up properly. Then getting back to what I think was the original question. Someone had ?tissue in alcohol for extended periods. Have done this, just avoided a harsh pretreatment of long proteinase K which just chews up everything and with no formalin to have fixed it, your tissue ends up a mess instead of partially and correctly digested.
I also am hoping I didn't spend my years suffering in grad school, eating cup-o-noodles every day with coffee after coffee to stay awake to study molecular immunology, molecular and cellular biology and signal transduction if sacrificing a chicken under a full moon (I hope I got the TUNEL protocol correct) was the simple answer on how to make this work.
-------------- Original message --------------
From: "Goodpaster, Tracy A"
> We use Biocare's cleaved caspase-3 (cat# CP229B) at 0.21 ug/ml for 30
> minutes. We use it on formalin fixed paraffin embedded sections after
> high pH steam antigen retrieval for 15 minutes with 15 minute cool down.
> Trilogy or pH 6.0 citrate buffer also work for antigen retrieval.
> Tonsil is a great control.
> Good Luck,
> Tracy Goodpaster
> -----Original Message-----
> From: email@example.com
> [mailto:firstname.lastname@example.org] On Behalf Of Roger
> Sent: Tuesday, September 04, 2007 1:33 PM
> To: Cheryl Cross; JR R
> Cc: email@example.com
> Subject: Re: [Histonet] TUNEL, Formalin, DNA strand breaks
> Never had the problem with TUNEL on formalin fixed tissue sections. And
> I have also used the anti-cleaved caspase 3 technique. Both worked
> beautifully, but at this point, due to the fact the the IHC anti-cleaved
> caspase 3 method is simpler, that's the way I'd go. I am racking my
> retired brain to come up with names of vendors. So far no joy. Two
> months of retirement (as of today) seem to have put my memory in
> neutral. If I come up with them I'll get back to you. All the catalogs
> and methods are still with my former place of employment.... I have
> done the TUNEL stain to verify results, and the pathologist wanted to
> compare apples to apples, as it were. All the new work is being done
> with the caspase 3 IHC.
> Roger C. Moretz, Ph.D. (Ret.)
> --- Cheryl Cross wrote:
> > Hi all -
> > I have been round and round with this issue myself; I was basically
> > told by people who've attempted it that TUNEL on FFPE sections is not
> > specific enough due to the formalin issue.
> > If you are trying to nail apoptosis, what about anti-active caspase-3?
> > mind you, that staining can be a bit of a booger too, but it should be
> > more sensitive and specific for apoptosis (no references to offer, i'm
> > just repeating what i have been told).
> > Cheryl Cross, DVM, Dipl. ACVP
> > Researcher
> > University Corporation for Atmospheric Research College of Veterinary
> > Medicine University of Tennessee Department of Pathology
> > 2407 River Drive, Room A201
> > Knoxville, TN 37996-4542
> > (423) 967-2724
> > fax: 865-974-5616
> > firstname.lastname@example.org
> > On Sep 4, 2007, at 2:46 PM, JR R wrote:
> > >
> > > Hi Ray,
> > >
> > > Sheesh-- I read (and then lost) the paper maybe
> > 5-7 years ago, back
> > > when I was tearing my hair out over high
> > background, or false
> > > positives for TUNEL stain in formalin fixed,
> > paraffin embedded
> > > arterial sections. I can troubleshoot any
> > immunostain, but the
> > > TUNEL and ISEL assays bedeviled me.
> > >
> > > The paper was titled something to the effect of
> > "formaldehyde
> > > causes single and double stranded DNA breaks," and
> > was pretty
> > > emphatic. I'll look for the article.
> > >
> > >
> > >
> > > I think the assay could work in whole cells for
> > flow, or maybe in
> > > frozen tissue. My hunch at this point is that
> > this is one of those
> > > rare cases where lots and lots of peer reviewed
> > articles are just
> > > plain wrong.
> > >
> > > Hey, if someone has a good protocol, I'd be
> > willing to try it out,
> > > and I would be delighted if I turn out to be
> > mistaken. For now, I
> > > think TUNEL is the assay of the beast.
> > >
> > > Oh, here is a thought--say you are looking at a 5
> > micron section
> > > through a nucleus. The microtome blade pretty
> > much had to create a
> > > lot of double stranded DNA breaks, no?
> > >
> > >
> > > Jerry L. Ricks
> > > Research Scientist
> > > U.W. Medicine at South Lake Union
> > > 815 Mercer Street
> > > Seattle, WA 98109
> > > (206)-685-7190
Histonet mailing list