[Cytometry] Dendritic cells and doublet discrimination

Uriel TK utk at urieltk.com
Thu Feb 5 19:09:34 EST 2009

I have been dealing with the subject you mention, although I have no knowledge of literature on that specific subject, i.e. DCs and doublet discrimination. DD relies on the assumption of a round and homogeneous population of cells. Thus when two events occur together you'd get a signal with twice width but the same peak value, whereas a single event that is bigger would show both larger width and peak values. Of course there are variations on this theme depending on the geometry of their alignment vs the laser beam, their size vs the laser spot size, etc. but that is the idea. 

DCs, as they name says, are not uniform. BUT that is true in slides or tissue. I've imaged live DCs in the plainest way by just putting a drop between a slide and a coverslip and letting them settle, and they are round/oval. I have seen DCs in an imagestream and they also appear quite round/oval. Thus it could be said that since in suspension they default to a round shape (as expected since they don't have anchor points), DD should work. BUT there are two things to consider:
1) some DCs still show significant asymmetry even without anchor points, with processes or very elliptical shapes. And that gets mixed together with
2) the forward scatter signal, which is affected by more than just size/shape: granularity and membrane complexity also affect FSC and those two are important feautures of DCs. As a side note the different beam and collection optics also cause differences in the results from different machines. I know because I've actually compared ARIA vs FACSCan vs LSR II to try and get around this same subject. My empirical conclusions/observations are that since DCs have significant membrane complexity, or maybe for some other reasons, DD is problematic in the sense that there are not clear cutoffs like there are for other cell types. In a A vs H plot you can catch the main population and exclude the "obvious" outliers and that's what I do, but the cutoff is a matter of taste regretfully. I've done my best to decide on a measure and stick to it, but you could set your own standard and I wouldn't have a "scientific" basis to say mine is better. Importantly, different treatments sometimes provide slightly but noticeably different distributions, further stressing the point that I'm trying to make. This has functional significance since DC morphology correlates with function and you could be mistakenly excluding important DCs that are just more complex or have more shape variability that confuses the DD process.

So the short answer for me was that I can clean up doublets, but cannot aspire to really get rid of all of them systematically. In the end filtering the sample 60 µm, imaging including post-sorting, and fluorescence measurements and analyses lead me to believe that after gating using my definition of singlets, there are few doublets left. Of course these are hmd-DCs in my hands. Your mileage may vary.

Best regards,


Uriel Trahtemberg, M.D. M.Sc.
PhD student
The Laboratory for Cellular and Molecular Immunology
The Hebrew University - Hadassah Medical Organization
Jerusalem - ISRAEL

"We have always known that heedless self-interest was bad morals; we know now that it is bad economics."
	Franklin D. Roosevelt

> -----Original Message-----
> From: cytometry-bounces at lists.purdue.edu [mailto:cytometry-
> bounces at lists.purdue.edu] On Behalf Of Mar?a Candela Iglesias Chiesa
> Sent: Thursday, February 05, 2009 5:32 PM
> To: Cytometry Mailing List
> Subject: [Cytometry] Dendritic cells and doublet discrimination
> Dear all,
> A basic question on human dendritic cell analysis: do you use a doublet
> discrimination gate (say a diagonal on FSC-H vs FSC-A) when you are looking
> for DCs?  I have heard comments on the doublet discrimination gate not
> working for something bigger than lymphos, and we seem to lose a lot of
> CD123+ and CD11c+ cells when we gate out doublets. On the other hand, how
> can I be sure I don¹t have aggregated cells if I base my first gate only on
> morphology (FSC vs SSC)?.  I¹d appreciate it if someone could point me to
> the right articles or discussion, since I haven¹t been able to find
> anything
> on the Purdue archives so far.
> Thanks for your help!
> Maria Candela Iglesias Chiesa, PhD
> Centro de Investigacion en Enfermedades Infecciosas
> Instituto Nacional de Enfermedades Respiratorias
> Calz. De Tlalpan 4502. Col. Seccion XVI
> Deleg. Tlalpan, Mexico DF, Mexico
> Tel y fax. +52 55 5666.7985
> candela.iglesias at cieni.org.mx
> www.cieni.org.mx
> _______________________________________________
> Cytometry mailing list
> Cytometry at lists.purdue.edu
> https://lists.purdue.edu/mailman/listinfo/cytometry

More information about the Cytometry mailing list