Nicole wrote:
>> Gabriel ,
> > I have been trying to stain fixed/permeabilized cells grown on coverslips
> > with IgM mouse monoclonal antibodies.
>> Are you trying to stain for intra- or extra-cellular antigens? IgM is a
> bad choice for intra-cellular staining. It's just too big to
> effectively get into a permeablized cell.
Some IgMs have been published for internal staining, and we have had
some success with suspension staining with one IgM against a nuclear
antigen.
Not all antigens survive fixation and/or permeabilization. Formaldehyde
can destroy some epitopes and alcohol fixatives or permeabilizers have a
relatively high incidence of destroying native epitopes. For
formaldehyde fixed samples, antibodies against native proteins MAY be
more effective, for precipitating/denaturing fixatives antibodies raised
against peptides or known to work in westerns may be best. (Watch me
grossly over-generalize!)
More information about the antigen and the methods tried would make it
easier for posters to provide to better (more specific) advice.
>> >I use biotin horse anti-mouse
> > followed by streptavidin-CY3. I get background - red specks all over.
>> Can you do this in fewer steps? Each step has the potential for its own
> background staining. What do your controls look like? Isotype
> control? Biotin HAM, and possibly even the strep-av CY3 will
> nonspecifically bind to the cell.
>> Stain with your usual protocol, just the biotin HAM followed by strep-av
> -CY3 and just the strep-av-CY3 to check background.
>> What kind of cells are you staining? This may affect the "stickiness"
> of the antibodies.
>> Do you see increased staing in a known positive? If so then this may
> just not be a very sensitive system. If not, then the problem may be
> more serious.
>> Nicole
Nicole is giving excellent advice. Implied in her comments, also, is
that you should have known negative and positive cells type you can
analyze (if at all possible). I am amazed at how often people just go
for it with a positive target when the antibodies aren't well tested for
the specific application.
Good luck
Tom
