Search the CONFOCAL archive at
http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal

Hi Dorothy,

As other replies have pointed out, PE bleaches very quickly, within 
seconds when searching by eye (Hg arc lamp, fluorescein/EGFP long 
pass emission filter). You might be successful focusing the 
microscope with transmitted light by eye, and then acquiring with the 
confocal, ex. 488 excitation with appropriate emission filter. I 
suggest opening the pinhole up, to at least see some staining. Once 
you know you can image, you can always shrink it down towards 1 Airy unit.

If someone at UT has a CompuCyte laser scanning cytometer that you 
can use, give it a shot. It uses a microscope but outputs flow 
cytometry data. The newest version (iCys) also produces decent 
images. The current issue of Cytometry A, 69A n 3 
(http://www3.interscience.wiley.com/cgi-bin/jhome/33945) has many 
articles using LSC.  See also Mittag et al 2005 Cytometry A 65(2): 
103-115 on 8-color LSC.

Another instrument that is a hybrid between flow and microscopy is 
the AMNIS ImageStream (http://www.amnis.com/). You might look into 
getting access to an ImageStream. Both CompuCyte and AMNIS are listed 
as exhibitors at the 2004 ISAC meeting and will likely be at ISAC's 
May meeting in Quebec City.

The fundamental difference between flow and microscopy is that flow 
has one data point per cell for <<<1 second for each many cells, and 
microscopy typically has 100 to 1000 pixels or more imaged over 
second(s) per cell. A typical confocal images each pixel for about 1 
microsecond (i.e. 1e6 pixels per second). In flow, as long as your 
fluorophore signal is above autofluorescence, you'll get a nice 
histogram. In microscopy, Each pixel has zero to a few fluorophores 
plus some autofluorescence, scattering, and detector noise. Signal to 
noise ratio is on a per pixel basis. If the microscope is focused on 
the equator of a spherical cell (i.e. T-cell or leukemia cell line), 
you'd think that you would just see signal from the fluorophores on 
the plasma membrane. Not! Confocal microscopes are excellent at 
producing noise in every pixel, so the inside and outside of the cell 
will always have some noise. (as a bonus, live cells may internalize 
some antibodies).

I had good success with anti-CD20 antibodies labeling human CD20+ 
leukemia cells (the CD20- control cells had essentially no 
background). A chemist in our lab labeled an in-house anti-CD20 mAb 
with the Invitrogen/Molecular Probes Alexa Fluor 647 labeling kit 
(we've labeled both Leu16 and DI-Leu16-IL-2 ... Fcblocking reagent is 
crucial). Imaging was done on City of Hope's Zeiss LSM 510 META NLO 
using 633 nm excitation for the Alexa 647 and 750 nm pulsed 2-photon 
excitation of subsaturating Hoechst 
(http://www.cityofhope.org/SharedResources/LightMicroscopy/InvertedLSM510.htm). 
The META has spectra-appeal but does not produce as good data as the 
standard confocal PMTs (which in turn, are not as sensitive as using 
the non-descanned detectors when using multi-photon excitation in a 
completely dark room - and that in spite of Zeiss' dumb location of 
the NDD detectors).

An advantage of Alexa Fluor 647 is that mammalian cells have 
relatively low autofluorescence in the far-red/near-infrared. The 
labeling kit is pretty simple to use, though you need to make sure 
that the kit you use will not label the buffer. Some of the other 
Alexa Fluor dyes are quite good if the right excitation wavelength is 
available.

Best wishes.


Sincerely,


George



At 01:46 PM 3/9/2006, you wrote:
>Search the CONFOCAL archive at
>http://listserv.acsu.buffalo.edu/cgi-bin/wa?S1=confocal
>
>Hi
>I have recently embarked on flow cytometry and am interested in 
>combining that with looking at intracellular markers by microscopy. 
>Does anyone know why the commonly used flow dye PE 
>(phycoerythrin)  is rarely (to never)  used for immunofluorescence 
>microscopy? I went ahead and stained with a PE labeled antibody and 
>found a confocal that should have allowed me to combine the right 
>ex/em wavelengths to see it and could not (although it was easily 
>seen by flow cytometry). The excitation spectra of PE is very broad 
>and that would make it hard to combine with other dyes but is there 
>something else that makes it unsuitable for immunofluorescence. A 
>flow person suggested that flow cytometry is so much more sensitive 
>that you might not expect to see it. Perhaps someone has direct 
>experience and can advise me.
>
>Thanks
>
>Dorothy
>
>PS   When I stain cell surface labeled cells using my normal 
>immunofluorescence protocol for an intracellular antigen developed 
>for microscopy  I do seem to get a very high background in the flow 
>cytometer with my normal rabbit serum control. By microscopy the 
>difference is like night and day but  by flow there was a huge 
>background so any insight into why or how to reduce that would be 
>nice too (this was using  an anti-rabbit Alexa 488 secondary) Merci!
>____________________________________
>Dorothy I. Mundy, Ph.D.
>Assistant Professor
>Department of Hematology Oncology
>5323 Harry Hines Blvd.
>Dallas, TX 75390-8565
>
>Phone: 214-648-1909
>[log in to unmask]
>
>
>***********************************************
>Dorothy I. Mundy, Ph.D.
>Assistant Professor
>Department of Hematology Oncology, NB7.116
>University of Texas Southwestern Medical School
>5323 Harry Hines Blvd.
>Dallas, TX 75390-8565
>
>Phone: 214-648-1909
>[log in to unmask]




George McNamara, Ph.D.
Division of Cancer Immunotherapeutics and Tumor Immunology
City of Hope National Medical Center
1500 E. Duarte Rd
Duarte, CA 91010
626-359-8111 ext 60035 office
818-547-6909                 home
[log in to unmask]
[log in to unmask]