To: Cytometry Mailing List <cytometry@flowcyt.cyto.purdue.edu>
Subject: Summary of replies to SP Cells Query
Date sent: Thu, 1 Dec 2005 09:17:23 +1100
Hi, I am having a bit of trouble setting up the Voltages and Compensation for SP Cells. I am acquiring the cells on a BD FACSVantage with DiVa option in Digital Mode. How critical are the filters,I have a Hoechst 33342 450/20 for the blue and a 660/20 for the red but I only have a 525DCSP for the splitter. Are these okay or not specific enough? I also have a problem in the two papers I am referring too. One paper states that the detectors should be in Linear mode while a second paper is in Log. Up till now I have been running in Log so which is correct and I would be very interested to hear anyones experiences. Thank you for your help.
A couple of comments. The blue filter you are using is fine for the Hoechst blue emission; however the red emission is far weaker and the bandpass you are using may be too narrow; I normally would have a long pass (620LP generally). Also the dichroic I use is normally a bit longer - 610LP (or Shortpass if you like, just change the channels - although saying that, it is also good to have a red-sensitive PMT for the Hoechst red emission). There is a degree of flexibility in the choice of filters, you can experiment to see which gives the best separation in your system.
The amplification of both the red and blue signals should be linear and as you are measuring different parts of the emission spectrum of the dye from a single excitation source, you don't need any compensation. There is more information on my website.
Good luck!
Derek Davies
--------------------------------------------------------------------------
Regarding the SP stuff, your 660/20 may not be wide enough to capture all of the red emission. We use a 450/50 to catch the short emission and a 675 LP to catch all the red emission we can get. We use a 505SP to split the signal. I see no reason why your 450/20 and your 525SP wouldn't do the job, but a 660/20 may be cutting off a lot of that far red emission that makes this work.
We use linear for both of the detectors. If I am doing an analysis on the vantage, I try to bump the UV up to 200mW, but I can usually get this to work at lower voltages. Obviously, if I am sorting these populations I try to get away with as little UV as possible...often down to 75-ish mW.
I hope this helps a bit....feel free to email if you have more questions, I might be able to help.
Good luck,
Joel.
--------------------------------------------------------------------------
I use analog Vantage with 100 mW Argon UV. The staining can be highly variable, so I check alignment with beads. Fixed cells stain brighter, so these can also help in set-up.
We use the 675/20 for the red & the 640 LP for dichroic & 424/40 for blue. ONLY use linear scales, NO compensation and a pump-inhibitor (eg, Fumitremorgin C) for your controls. The UV-excited PI fluorescence (dead cells) will be off-scale in the red.
Dennis J. Young
--------------------------------------------------------------------------
One of the original papers on this is Goodell et al J Exp Med 1996 1797-1806 who used 450/20BP, 675 LP and a 610 beam splitter (both measurements, obviously, from the UV line) on a FACSStarplus.
We have used a 424/44BP, 670LP and 610 beam splitter eg Jonkers et al Stem Cells 2005 1059-1065 on a FACSVantageSE (we have also looked since we had the DiVa upgrade and all is much the same, as one might expect).
So I suspect your blue filter and beam splitter should be OK (I can't think that 525 or 610 beam splitter would make that much difference in this assay?) but I would widen things out with a red long pass, 660/20 seems a bit restrictive. Look at both red and blue on linear scales and the blue intensity should be in the ball park of what you would expect for a Hoechst stained sample for DNA analysis (red voltage will probably need to be set a little higher).
Good luck
Ian
--------------------------------------------------------------------------
We do SP cells quite regularly here on our Vantage.
We use similiar sets of filters (perhaps a bit longer on red) but "steal" the 610SP spliter from the FL3/4 division. However, the spliter you have should be fine becuase we have been using a 505 on the LSRII and it looks fine.
We always do the red and blue from the UV in linear.
Dr Adrian Smith
--------------------------------------------------------------------------
Hoechst blue is never a Problem, but Hoechst red is very dim, so may be your 660/20 Filter is to narrow. We are using 670 longpass. On our MoFlo I do not enhance the blue signal, but the voltage on the red PMT is almost maximum.
SP cells you definately have to plot in linear scale, your can see the the diploid, tetraploid or haploid (when staining sperm) cell fractions nicely then.
Start with mouse bone marrow, it works always nicely.
Good luck.
Best regards from Conny/Germany
--------------------------------------------------------------------------
I think the filters should work.
Usually for SP the detectors should be in linear mode, you will not get the same profile in log.
When I set up voltages for SP (on a MoFlo), because Hoechst staining is very bright, the population of cells is often off-scale so I have to bring the voltages down to be able to view it on the display graph.
Caroline
--------------------------------------------------------------------------
I stain and sort SP cells from bone marrow,mcf7 breast cell line, c6 glioma and others .
Stain with Hoechst 33342 at 5ug /ml.PI at 2ug/ml.
37 C for 90 mins.
Leave ON ICE until FACS
Kep cold whilst on cytometer
Excite with a 350-355 MLUV laser lines at 25 or 50mW,
depending how sensitive your optics are.
Split the signal with 610 DMSP or DLP. Hoechst at
450/20 BP and PI at 670/40 or LP, fine tune with
PMT voltages and keep in linear for both parameters.
Since this only ties up the uv pathway you can ues
the FITC for stem cell analysis using SCA 1 and/or CD34.
Gate out the PI positive cells(670nm channel).
You do not need compensation correction. You do not run this in log. but lin. as this is a DNA profile and the SP cells are spewing out of the Go/G1 fraction. Reduce the voltages and you see G1/GO and G2M peaks.
Hope this helps a little
Nigel Miller
--------------------------------------------------------------------------
When we measure the side-population on our FacsVantage we work in linear amplification ( we never tried it in logathitmic mode ) and our optical design is as follows ;
For the " blue hoechst " we use a 424/44 nm BP filter ,
For the "red shifted Hoechst " we use a 600nm Long Pass Filter to obtain as much as possible red fluorescence ,
For the splitting of the different wavelenghts we use a 505 nm Small Pass Filter .
No compensation is required if you only measure the side population, if you further want to fenotype then you need compensation for the identification parameters . If possible I would advice you to change your red BP-filter in a 600 nm Long Pass Filter or another bandpass filter with a broader bandwidth .
Hopefully this hints wiil help you,
Marc
--------------------------------------------------------------------------
It depends on what you want to look at. If you are using Hoechst to check cell cycle then I'd use a linear scale since Log scale compress the peaks. The filters seem OK for what you are looking at but I don't know what red emitting probe you are using? You have to check the products spectral excitation and emission info to set up the right filters
...Ken
--------------------------------------------------------------------------
I think your filters/mirrors are fine.
I do indeed look at both blue and red in a linear mode.
I try to adjust the PMTs so that the majority of the cells are on the diagonal and so that the G1 (easy to make out 1st major blob) are about 2/3rds the way up the scale--this gives max resolution for the (slightly sub-2N) SP cells to show up.
Most importantly, be sure to titrate for both Hoechst concentration and incubation time--I've found that unless both are optimized SP is often missed.
Finally, no compensation is needed! We also add PI to the mix which makes the dead cells much more red than blue and allows for an easy exclusion.
Hope this helps! Any more questions, I'll do my best to help.
Good luck!
Jamie Barber
