Isotype controls for cytokine staining

Mario Roederer roederer at
Mon Jun 4 19:51:36 EDT 2007

Isotype controls definitely have their value.  The reason I recommend  
against them is simply that people over-rely on them when they use  
them.  I'd rather that people not use them, and instead, think the  
problem out more carefully.

Specifically, isotype controls can identify IF there is a problem  
with nonspecific binding.  However, they cannot be used to identify  
HOW MUCH of a problem there is.  (Because of all of the already  
heavily-discussed issues, like F/P ratios, concentration, etc.)

Thus, isotype controls can act perhaps as a "red flag".  But setting  
gates (discrimination of positive vs. negative) based on isotype  
controls will only lead you down to false positive or false negatives.

Don't use isotypes for gating.	Use them to gain information about  
your staining.

Better yet... don't use isotypes.


On Jun 3, 2007, at 11:39 PM, Stuart Berzins wrote:

> Dear All,
> At the risk of re-igniting the whole topic of isotypes controls, am I
> correct in arguing that isotype controls ARE important (or at least  
> helpful)
> when examining antigen expression by stimulated vs non-stimulated  
> cells (for
> example, cytokine expression by T cells)? That is, providing they  
> are used
> alongside all the other antibodies in the normal cocktail to	
> control for
> compensation issues.
> I understand the limitations of isotypes relating to F/P ratios  
> etc, but
> equally, by not using isotype controls, isn't it akin to throwing  
> the baby
> out with the bathwater because alternate controls (eg examining  
> expression
> on known non-expressing cells) do not adequately control for Fcr- 
> binding and
> other 'stickiness' issues that might be unique to the stimulated  
> cell type?
> Am I right or wrong? I have searched the archives, but arguments  
> against
> isotypes seem to rely heavily on having very similar cells  
> available that
> are known not to express the antigen in question (eg for CD34  
> staining).
> That is not the case with activated T cells.
> Dr Stuart Berzins
> NHMRC RD Wright Fellow
> Department of Microbiology and Immunology,
> The University of Melbourne,
> Parkville 3010,
> email: berzins at
> Ph (office): +61-3-8344-5706
> Ph (lab): +61-3-8344-5704
> Fax: +61-3-9347-1540
> Mobile: 0427 849 123

More information about the Cytometry mailing list