Adrian Smith a.smith at centenary.usyd.edu.au
Thu May 14 07:52:21 EDT 2009

Hi Jennifer,

Thanks for sharing that - I think it is very helpful.

Might it better to say in your last sentence that if you insist on  
using isotypes then they are best used with FMO controls, ie a sample  
will all isotypes is not particularly useful... or less useful than  
one where an isotype is used on an FMO (... although less than close  
to zero is not really that different I guess :)


Adrian Smith
Centenary Institute, Sydney, Australia

On 14/05/2009, at 3:29 AM, Jennifer Wilshire wrote:

> Hi Flowers,
> I just finished writing up a note for our customers about controls  
> in flow cytometry and this seems like a good opportunity to get  
> feedback to make sure all I'm declaring is true (see below).  It  
> seems to me that focussing on the prevention of false positives is  
> the best way to go, since in most cases we don't have a good way to  
> identify them in our data.
> Feedback and improvements/additions welcome.  If anybody would like  
> a copy of this doc - email me and I'll send it to you.  I've also  
> attached it to this email - but I'm not sure it will go through.
> Jennifer
> A microliter of prevention…
> Isotype controls are not a particularly good control.  While they  
> are meant to be a control for non-specific binding, they are rarely  
> properly matched to the reagent of interest:
> -the Fluorochrome to Protein (F/P) ratio is often not the same
> -the concentration is likely not the same
> If the isotype control’s F/P ratio and concentration are not the  
> same as the reagent of interest, then it isn't a proper control –  
> where ONE thing is different from the experimental condition.
> There are many reasons you might get false positives in flow  
> cytometry:
> -antibodies binding via Fc receptors
> -true non-specific binding due to protein-protein interactions
> -stickiness of the intracellular matrix
> -dead cells binding antibodies indiscriminately
> -autofluorescence
> -incorrect compensation
> -spreading of populations due to measurement error
> The following chart describes preventative measures that can be  
> taken to prevent false positives; and if that is not possible, at  
> least identify where to place your gates to be confident whether  
> cells are positive or not:
> Avoiding False Positives
> Problem                          Prevention
> --------              	         ----------
> FC receptor binding   	         FC block *
> Non-specific binding 	         Titrate antibodies *
> Intracellular stickiness	 Titrate antibodies *
> Dead cells           	         Viability stains *
> Autofluorescence  	         Use unstained cells as control
> Compensation           	         Proper controls -> use software to  
> calculate
> Spreading           	         FMO controls
> 	                         Bi-Exponential scaling
> 	                         Minimize spreading in important parameters
> In the chart above - you could add "Isotype controls" beside the  
> problems indicated with a star (*).  If isotypes were to work  
> perfectly, they would identify these problems.  However, because  
> isotypes don't work perfectly, preventing the problem is the best  
> method by Fc blocking, viability stains and proper titration.
> Note:  isotype controls and FMO controls have completely different  
> goals.  FMO control came into use at the same time as isotypes  
> started fading from use, so a lot of us viewed FMOs as a proper  
> substitute for isotypes.  However, they DO NOT address the same  
> issues.  FMOs show the spreading due to measurement error and  
> isotypes (attempt to) measure non-specific binding.
> In the end, there is really no good control for non-specific  
> binding, but if you insist on using isotype controls, then you can  
> combine them with the FMO controls to reduce the number of controls.
> -----Original Message-----
>> Of those performing 5 color panels on FC500s, do any of you use  
>> isotype controls-why or why not?
> Jennifer Wilshire
> Memorial Sloan-Kettering Cancer Center
> Zuckerman Building Room Z16-34
> New York, NY
> 10065
> 1-646-888-2381
> wilshirj at mskcc.org
> fccfserver.mskcc.org
> _______________________________________________
> Cytometry mailing list
> Cytometry at lists.purdue.edu
> https://lists.purdue.edu/mailman/listinfo/cytometry

More information about the Cytometry mailing list