Q1. Try glow-discharging the formvar-coated grids to make them more hydrophylic, and let sample settle down for at least 5mins before continuing with Uranyl Acetate staining. Don't wash off any access UA, that may dislodge the stained sample, but gently pull off the drop of UA with filter paper held through the tips of your forceps which clasp the grid.

Q2: Virus-sized balls from a virus sample are most probably your virus particles. UA can crystallize if used at higher concentrations, but crystals do not appear as aggregates of spheres...

Q1. Try glow-discharging the formvar-coated grids to make them more hydrophylic, and let sample settle down for at least 5mins before continuing with Uranyl Acetate staining. Don't wash off any access UA, that may dislodge the stained sample, but gently pull off the drop of UA with filter paper held through the tips of your forceps which clasp the grid.

Q2: Virus-sized balls from a virus sample are most probably your virus particles. UA can crystallize if used at higher concentrations, but crystals do not appear as aggregates of spheres...

Thank you very much for your helpful advice. I will try it and let you know the result by posting here :-)

Dear Motoharu Ono,

it might sometimes be that virus sampling and negative staining on grids might yield only poor if no result for several reasons I will not go into detail. As of your question that you are examining lentivirus/adenovirus mixtures (or did I get this wrong and you are examining either Lenti- / or Adenovirus preps?) it might be that there occurs perhaps clumping of virus particle "balls" as you describe. Just as ONE possibility to check out your method I would like to point you to an article (I had in my collection) from 2003 dealing with purification and negative staining of lentiviruses, so you should be able to check your prep.-steps:

Lentivirus Vector Purification Using Anion Exchange HPLC Leads to Improved Gene Transfer. By:

Kaoru Yamada, Douglas M. McCarty, Victoria J. Madden, and Christopher E. Walsh

University of North Carolina at Chapel Hill, Chapel Hill,

NC, USA

BioTechniques 34:1074-1080 (May 2003)

From the Methods section:

BioTechniques 1075-1076, Vol. 34, No. 5 (2003)

I dont have any experience with mammalian viruses, however I work frequently with various plant viruses. You might try couple of trick to concentrate your virus on the grid. If you have an antibody, the easiest way would be to coat your grid with antibody first, then wash it on several drops of water or PBS and then incubate your virus prep and stain it as you are used to. It increases the number of our VLPs by at least one order of mangitude. There are couple of other tricks to concentrate virus for TEM such as the use of PEG (see Colombet et al, Journal of Microbiological Methods 2007). Another problem might be that your viruses are unstable at acidic pH, so you might want to try ammonium molybdate or phosphothungstic acid instead of UA.

Thanks for all of your useful advices. It looks like Glow-discharging method worked well in my case. I could have clear virus staining. SEM spheres seems a mixture of virus and NaCl crystal based on spectra analysis.

You could try diluting the sample in HEPES to get less NaCl crystals