[Histonet] CD4 and CD8 background in mouse tumors frozen sections, help

[Erickson, Jamie E]

Hello Histonetters,
                                         First I would like to say happy holidays to you all, this is such a great resource to have.
Now for my question.
I'm rather vexed about this problem with staining CD4/Cd8 in mouse tumors I hope someone can shed light on because I'm stumped.
An investigator gave me some frozen mouse tumors and spleen to do CD4 and CD8 stain on and I expected this to be easy...
After staining the spleens and tumor with CD4 and CD8 at (1ug/ml and 5ug/ml respectively) I saw the problem, the isotype control. The isotype for these 2 Rat-mouse antibodies from BD biosciences is a Rat IgG2a.
The rat IgG2a on the mouse tumor and spleen slide were  negative and I was happy but when I looked at the other tumors  Rat IgG2a slide some were very positive. The CD4 was staining in some tumors but these tumors had background throughout the tissue as well. Very strange I thought, some tumors no background others lots of background..

As  I'm not a tumor biologist I'll just say these tumors vary in size from small to large,  large (>1cm) and some have necrosis others do not.
So what I see is that the isotype (rat IgG2a) stained in some tumors but not all of them and  the spleen and a few small tumors are negative.  I only did Neg controls on 2 tumors one large and one small tumor with a spleen from that tumor bearing mouse.

I first thought peroxidase or Fc binding as I don't do a protein blocking step with this system, typically. I added a protein block (Dako) and increase peroxidase time from 5 minute to 10 but saw no difference.
It looked like my titers were high for these tumors so I titered  the antibody out. I  titered my CD4 down to 0.25ug/ml for these tumors  and that did not help the Rat IgG2a staining but the CD4 staining looked better not as dark brown in the spleen.
So now I need to explain my staining protocol.. We use a Leica Bond RX system which I use a DAB kit to detect with and it works great.

Marker (antibody, Rat Anti-mouse CD4)                15
Rabbit Anti-Rat
vector lab (adsorbed to mouse)
as a linking step.                                                               20
polymer (Anti-Rabbit)                                                    8
Peroxide block                                                                  5
DAB                                                                                       10
Hematoxylin                                                                      5

I next looked at the linker step
slide #1  I did a slide with no CD4.... still had staining (background)
Slide #2 No CD4 and No linker ........No background.. clean....I think it is the linker..
Unfortunately I'm stuck using this linker (rabbit anti-Rat , 2ug/ml) for now..

What could be going on in these tumors ???  could I block this without changing my polymer (rabbit). If it is the linker then is it binding via Rabbit Fc or is the hypervariable region specific for something in these tumors i.e.... Anti-Rat tumor protein?
Can I pre-complex the CD4 and linker then bind up excess anti-Rat, any one done that???
If it is not complicated enough the linker is adsorbed to mouse.
I tried Biocares Rat-on mouse polymer but I did not see any reaction at all, ..but I haven't investigated that fully..

Sorry for the long email ..
Hope you can help...Happy holidays.

Jamie Erickson
AbbVie BioResearch Center
Pharmacology
100 Research Drive
Worcester, MA 01605
OFFICE    +1 508-688-3749
EMAIL  terry.melim <@t> abbvie.com