Dear Alba Jimeno, well, this is quite a good and perhaps a remarkable question...

unfortunately you don’t say anything about the material (cell culture, tissue, organ, thick sections, [ultra-]thin sections etc.) you have to examine or analyze by x-ray microanalysis (nor the specific method you will use for that).

So I will try to give it just my best but trivial shot on this matter:

i) first of all, you would like to fix chemically as usual, but simply omitting OsO4-postfixation.

ii) if you say “not applicable for our demand” – guessing you MUST see chemically fixed - almost perfect structures as usual - you perhaps could try a double fixation using buffered GA + tannic acid *) (e.g. 1%, LOW molecular weight TA!, sonicated and filtered prior to mixing with GA), followed – after at least several washes in Aqua bidest. (if not using the Malonate/maleinic acid buffer sequence for washing) by 1% Uranyl acetate*). Such has been said to be [ “perhaps”] a suited substitute for an “ordinary” osmium postfixation.

Since the radioactive uranyl acetate at some places has been banned you could try (admittedly this would be an experiment worth) also to add samarium triacetate*) or gadolinium triacetate*) (substitutes for uranyl acetate, see also a specific post in ResearchGate or find at: http://www.ncbi.nlm.nih.gov/pubmed/22146677

Nakakoshi M, Nishioka H, Katayama E: New versatile staining reagents for biological transmission electron microscopy that substitute for uranyl acetate.

J Electron Microsc (Tokyo). 2011 Dec;60(6):401-7. doi: 10.1093/jmicro/dfr084.)

to the initial fixative or as postfixative after GA-tannin primary fixation. For reasons of lengthiness I will not cite more articles except an additional one, where one could find some more specific information on technique one could think of:

:

Berryman MA, Porter WR, Rodewald RD, Hubbard AL (1992). Effects of tannic acid on antigenicity and membrane contrast in ultrastructural immunocytochemistry. Histochemistry and Cytochemistry 40, 6, 845-857

(open access at: http://www.ncbi.nlm.nih.gov/pubmed/?term=Berryman+MA%2C+Porter+WR%2C+Rodewald+RD%2C+Hubbard+AL+(1992),

free .pdf see: http://jhc.sagepub.com/content/40/6/845.long.

*) not knowing whether these substances interfere with energy peaks around 8-9 keV you are fighting against.

best wishes and good luck

Dear Alba, you have got an excellent advice from Wolfgang, I just want to add. In some cases you may try to skip postfixation completely. I often do not use OsO4 at all; actually right now I am processing new (for me) specimens in two batches, with and without Os, to compare their quality. You will have some problems with staining, since both Pb and W have peaks at 8-9 kV, so only U and Sm will be usable for you.

Dear Wolfgang,

Sorry for being so vague in my description of samples. We work with a wide variety of tissue samples, from different animal species, both vertebrates and invertebrates.

I will be trying to substitute OsO4 with tannic acid as you suggest and compare the resulting membrane fixation, as it is an issue of major concern for me. Will also check how the different stainings behave in relation to Os and tannic acid postfixation.

Vladimir,

I did try to skip postfixation with OsO4, and as you are doing now I compared samples that have undergone Os fixation and ones which didn't. For my samples at least, tissue organisation and general membrane integrity were totally lost when not postfixing (taking into consideration that I was using Glutaraldehide as main fixative). I hope you get better results.

As for staining, I skip that step completely (and when I absolutely have to, I do as you suggest and use U). I have found that for my species, Os postfixation stained the membranes so clearly and accurately that I could completely avoid staining whithout compromising image sharpness, or identification ability.

Thanks for your advice.

Dear Alba, thank you for your reply...as with all complex systems, I forgot to ask you (and you didn't say in your reply to Vladimir) about the buffer system you use for the Glutaraldehyde (GA)-fixation.

It could be that Na-Na-phosphate-buffered GA fixation has a more beneficial effect with regard to morphology and contrast gain than using Na-cacodylate.

And - as always - it might turn out that the addition of Tannic acid -based solely on the fact "to add tannic acid to the fixative" might be a little bit shortsighted. So I would recommend to google or search in PubMed for some articleswith recipes/method schedules for this task.

Lt us know about either the success or also about failing of your fixation experiments, I gretaly should acknowledge that (also for other readers of this thread).

Best wishes, good luck and warm regards,

Wolfgang

Dear Wolfgang,

I know my colleagues work with Na-cacodylate as buffer for the GA fixations, and thus far I am not aware of any morphological disturbances occuring in their samples (mainly fish liver, or whole fish larvae).

As for me, I use filtered seawater as buffer for the GA fixation, as I have found out that it works best in preserving the morphology of bivalve mollusks tissue samples. However, it makes the postfixation step mandatory. I will definitely check how samples turn out using Na-Na-phosphate buffered GA both with and without postfixation compared to my current seawater-buffered GA.

I will updater this question with the results of said trials, but I have to acknowledge that it may take some time to get them done, as we don't have any experiment running at the moment that requires sample processing for TEM.

Best wishes,

Dear Alba - firstly just to be honestly acknowledging your really informative reply on this ! - secondly: I understand that your colleagues using Na-cacodylate as buffer vehicle for the GA fixation don't report any problems with it - concerning the procedure and outcome (preservation of structural components) since usually Na-cacodylate is a very common medium if not the "buffer of choice" in TEM, at least, because it makes/induces no major troubles [ e.g. by producing "salt & pepper" contaminations as seen after staining and examining the ultrtathin sections as usually the use (Na-)Na-PO4 buffer can induce, especially when you dehydrate rapidly = out of fixation - washes or no washes and dehydrating by 70% ethanol as the first step= precipitation of PO4 within the tissue spec). I understand also that the buffer vehicle you use (filtered seawater) is very special in that case (one should know exactly the pH, the tonicity as well as perhaps the concentrations of + and ++ [cat]ions contained in this). But I think you might change AFTER fixation that to a Na-Na-PO4 buffer (if the Ca-concentration in your seawater is not too high...otherwise: P-precipitation with Ca!) which has been adjusted in pH as well as molarity to the previously used seawater conditions. Naturally this would be more work and experimental fiddling around with the conditions of spec preps for seawater-molluscs. If you need tables for the making of the Millonig's Na-Na-PO4-buffer (pH as well as different molarities) please don't hesitate to request them (I have some in my files). Unfortunately I have to admit that I never had any seawater specimens in my lab and therefore I don't have any experience about how one could overcome the problems I indicated above. All the discussion (Na-Na-PO4) only with the background that it has been reported to be the buffer of choice for (at least) human tissue specs, superior to (staining/contrasting) results achieved with "normal" Na-Cacodylate-buffer vehicle. I and others perhaps will be pleased to read at a certain time again, what has happened meanwhile and whether you were able to overcome the problem. Very best wishes and regards, sincerely yours, Wolfgang

If you wish to avoid osmium, I would suggest ruthenium tetroxide. X-rays are at 19.2 (K alpha) and 2.5 (L alpha). RuO4 will be more reactive than OsO4, so a little R&D may be required.

http://www.nature.com/jid/journal/v104/n3/pdf/5610961a.pdf

I feel sorry that I did not see this question earlier. Not that I have a good answer, rather some suggestions, especially considering that you work with sea water organisms...

We have tried to use KMnO4 instead OsO4 on our neuronal cultures. It fixes membranes great, though usage of KMnO4 leads to the considerable extractions of the cytoplasmic components, thus you have nicely visible membranes and homogeneously light grey cytoplasm... If you are interested in membranes only this could be an option. Please, have look on the attached image. You can use KMnO4 at ~1-1.5% made on 160 mM NaCl or even on PBS, but not on organic buffered solutions. I even fixed the samples (neuronal culture and acute brain slices) directly in permanganate solution, even without primary fixation with glutaraldehyde - orks well, the ultrastructure looks the same as on the sample image.

Concerning using Tannic Acid - I was advised to use gallic acid instead as it is not a mix like TA, but pure one, thus considered to be better (if you are interested you can contact Heinz Schwarz: https://www.researchgate.net/profile/Heinz_Schwarz/). Though, never tried myself. Another point: John Heuser once told me, that if you want to fix not only plasma membrane with TA but also the internal membranes you'd need to use a little bit of detergent in your fixing solution to let the TA penetrate... (e.g., have a look here: http://www.heuserlab.wustl.edu/procedures/glutaraldehyde-OsO4.shtml) Again - never followed this advice myself as we always make post-fixation, most of the time with OsO4.

Good luck.

Have you tried chlorpaladious acid? Paladium chloride in acid solution has a similar reactivity (for carbon-carbon unsaturated bonds) to Os and is used in a similar way.