this is just a guess and you should ask a FACs experet if it might work.  if you were to stain with the hoechst stain mentioned above, and then run a plot with the color as "A" (area) on one axis and "H" (Height) on the other, you might be able to distinguish because a binucleated cell would have the stain distributed as two lobes (where A and H would not be the same) whereas a tetraploid cell would (at least should) have the stain be distributed more as a sphere  (unless your cells are like polymorphonuclear cells with multilopbed nucleii).  thus the tetraplid cells would fall on a line and the binucleated ones should fall off the line.  it's similar to how people distinguish single cells from doublets.

another possibility might be find something that stains the nuclear membrane.  do you know if the nuclei in the two types of cells are similar in size?  would the binuclear cells have a greater amount of nuclear membrane than the tetraploid cells?

i'm curious to know why you want to do this.

Hi Ankur,

yes, go for the FACS, that's fast and easy. Does it matter if the cells are dead or do you want to harvest live cells? 

I wanted to harvest live cells.

Then use Hoechst 33342 staining. Compared to PI and DAPI, Hoechst can penetrate the plasma membranes of living cells and stains DNA. You will need a FACS equipped with an UV laser. 

here's a method paper: http://onlinelibrary.wiley.com/doi/10.1002/cyto.10173/pdf

Thanks Olivier. But I dont want to separate cells in different phases of cell cycle. Instead , I am trying to sort Bi nucleated and Tetraploid cells. I hope you get my point.

Sorry that I misunderstood your point. You're right, that approach will lead you nowhere. If your binucleated cells are bigger than tetraploid cells, then you could gate them using the SSC and FSC on the FACS. But I doubt, that this is a straightforward approach. If your tetraploid cells are in G2 because of mitosis, you could distinguish them by adding EdU or BrdU, but that only works with fixed cells.

Maybe somebody else has an idea.... all the best! 

 this is just a guess and you should ask a FACs experet if it might work.  if you were to stain with the hoechst stain mentioned above, and then run a plot with the color as "A" (area) on one axis and "H" (Height) on the other, you might be able to distinguish because a binucleated cell would have the stain distributed as two lobes (where A and H would not be the same) whereas a tetraploid cell would (at least should) have the stain be distributed more as a sphere  (unless your cells are like polymorphonuclear cells with multilopbed nucleii).  thus the tetraplid cells would fall on a line and the binucleated ones should fall off the line.  it's similar to how people distinguish single cells from doublets.

another possibility might be find something that stains the nuclear membrane.  do you know if the nuclei in the two types of cells are similar in size?  would the binuclear cells have a greater amount of nuclear membrane than the tetraploid cells?

i'm curious to know why you want to do this.

Transfect plasmid vectors containing the code of GFP-fused protein of nuclear membrane (ex. lamins).

Incubate.

Stain with Hoechst33342.

Make respective gates with the levels of GFP and Hoecst33342 as follows and sort them.

●Hoechst33342-high and GFP-high cells are binucleated cells.

●Hoechst33342-high and GFP-low (similar levels to normal MNCs) are tetraploid cells.

●Hoechst33342-low and GFP-low are MNCs.

……My idea might be difficult because transfected cells might show different behavior from intact cells.

Hello

I agree with all , For more information see these links 

Good Luck

https://www.noble.org/Global/medicagohandbook/pdf/FlowCytometry.pdf

http://www.einstein.yu.edu/research/facilities/facs/page.aspx?id=22633

https://home.ccr.cancer.gov/flowcore/Telford_NUVLD_proof.pdf

As Polly Matzinger has already mentioned in her reply, it depends on what do you want to do with the cells after sorting. But transfecting the cells with plasmid vectors is not really helpful to use the cells for long term functional responses especially after sorting them. And in this case they are also not so flexible even to study short term signalling events.

I would rather try Polly Matzinger's idea.

As an alternative dye you might want to consider DRAQ5.

My idea in above is completely useless in the first place, I realized after answered.

In this method distinction of GFP-high and -low cells may not be impossible or not be clear because disorderly produced GFP-fused proteins which are not recruited into nuclear membrane may also bright.

Please forget or pretend you didn't see my comment!!

Following from Olivier's answer, is it correct that you mean tetraploid cells that are in normal G2/M while the binucleate (and multinucleate if present) cells would be arrested? 

If so, there are live cell-cycle markers like the FUCCI markers that might be suitable, but again it involves transfection.  

A nuclear membrane marker does sound a possibility - yes there ought to be more membrane (by about 26%) if the same volume of chromatin is subdivided into 2 equal spheres, but that might be quite a small difference to detect, and again only if you are prepared to transfect (or electroporate or something).

As Polly above explains: PI, Hoechst,Dapi or any other dna dye will work.you need to gate out doublet's,  clumps etc. Since you are sorting then sort each population if you are unsure where the cells are and check on microscope. It will work a bit better with fixed cells.

Hi Ankur,

We are sharing the same interest in identifying what we call "polyploid cells" (mean more than 4N DNA content) in a diploid population and we are also using FACS. We have recently published using this methods (Jungas et al, JCB 2016, attached). A similar strategy was also used by Di Cunto F, neuron, 2000.

Here is a rapid description.  We do the same with primary cells (eg fresh murine cortical neurons) and cell line (eg HeLa, U251).

We dissociate our cells to have them as singlets (mechanically or enzymatically), we fix using 70% ethanol and we stain with propidium idodie solution containing Rnase. Using other dye should also work (we have teste 7-AAD).

For Facs acquisition: On FSC/SSC dot plot we exclude debris. On Fl2-H we set our first peak (corresponding to G1) at 200 fluo units. On Fl2-A/Fl2-W we exclude doublets and more (they have a longer Width) and then we go back on Fl2-H to analyse cell cycle distribution.

G1(2N) peak  will be at 200 units

S will be at 200 to 400 units

G2 +M (4N + 2x2N) peak will be at 400 units

We consider that cells that have a DNA content over 4N and that are not doublets or triplets  are polyploids, meaning more than a diploid content in the same cell. In a model of cytokinesis failure we have observe peak of 4N, 8N and 16N. ... Unfortunately with this analysis you cannot conclude on endoreplication or binucleation of those cells cause FACS couldn't dicriminate if DNA is segregated in a unique or in two nuclear enveloppe. To decipher this point we go back to immunofluorescence. At it could help you to quantify the proportion of polpyploid cells in your population and will encourage you to go spend time counting under microsocpe...

Feel free to contact me if you need more detail procedure or discuss around any point. And of course if you have a method to discriminate bi-nucleated cells versus endoreplicated, i'm really interested!

Thomas

Article Eph-mediated tyrosine phosphorylation of citron kinase contr...

HI All,

I just read all your comments, it seems like most of them suggest for Hoechst stain.

Earlier I used Hoechst stain to sort polyploid cells but problem is after sorting cell does not grow. Seem to me cell does not like growing with Hoechst stain.

Do you have any other alternative to sort live polyploid cells to do further experiments on them