3-colour labeling for the single argon-ion laser? - DNA + 2 antibodies (Nov/19/2009 )

High,
Does anyone have an experience of a 3-colour labeling for flow cytometer with only one excitation argon-ion 488 nm laser? I need to stain DNA and two antigens by indirect immunofluorescence (or, in other experiments, DNA, 1 antigen and CFDA-SE). There is no problem with FL1 (CY2, AlexaFluor488, DyLight 488-conjugated antibody, or CFDA-SE), in my case they need minimal or no compensation in FL2. But it appeared difficult to use simultaneously a red DNA-binding stain and a red antibody. PI is not good at all, as it gives fluorescence in FL2, FL3 and FL4. I tried to use PE-conjugated antibody (FL2) and 7-aminoactinomycin D (7-AAD) for DNA (in FL4). However, in reality 7-AAD yelded in low resolution when used in 2 mkg/ml, and even in this relatively low concentration gave some FL2 fluorescence. Moreovere, it was hard to compensate for it since the concentration of 7-AAD was unsaturating and the positions of 7-AAD peaks varied from sample to sample. If anyone worked with 7-AAD, what concentration have you used? Do you feel it is a good pair indeed?
On the other hand, longer-wavelength stain LDS 751 (Molecular Probes) exists, but I am not sure if it is good for quantitative staining and if it can help to overcome these problems (we are thinking of buying it if it is really better than 7AAD).
As an alternative I thought about exchanging the chanels: to buy PerCP-conjugated secondary antibody (Jackson ImmunoResearch) to view them in FL4, and to use some DNA stain for FL2, but I do not know any good FL2 DNA stain with no fluorescence in FL4 (do you know?)
Do you know anything about these or other alternative pairs?
Thank you!

I have 6 channels for 488 exitation on my LSR-II. Everything is OK.

Did CFSE- CDxxPE- CDxxx-PE-TexRed-7AAD staining many times even using old stuff like EpicsXL. Was working great for PBMCs

However if your cytometer does not have a full compenstion matrix (like Calibur or Elite) such an analysis could be impossible.