RE: Auto-fluorescence

From: Webster, Dr Gill (G.Webster@genesis.co.nz)
Date: Sat Jun 08 2002 - 23:48:33 EST


I have alot of experience at looking in transfected tobbacco and Zinnia protoplasts with
green red and cyan fluorescent proteins.  The plots you enclosed are a little unclear
as to what gates are applied to what, if any of the plots.  It looks like to me that
you are excluding most of your protoplasts by thresholding on SSC-your gains appear too
low relative to your thresholding channel.  Regarding your autofluorescence problem-
How are you defining this?-autofluorescence is in part, a feature of cell size-the
bigger the particle analysed the greater the autofluorescence.  you should not be
surpried that you need much less signal amplification when running plant protopalsts
compared to typically smaller mammalian cells.  Autofluorescence is really only a
problem when you can demonstrate that is truly masks your positive signal of interest.
I have not personally ever had a problem in distinguishing between background FL1
and GFP+ fluorescence in protoplasts, once the non viable cells have been excluded
from analytical plot.  You should drop your FL1 gains abit and bring your background
fluorescence down to 1st decade.  You will actually gain more visual sensitivity for
GFP Dim cells if you do this.
I would suggest the following.
1) using SSC Int instead of LOG.
2) I would not threshold on SSC.  In my experience a FSC discriminator is more useful.
In my systems I find my transformants have a range of FSC signals and a lower SSC
signal compared to the non transformants.  When FSC vs SSC plots are displayed on Int
scales it is usually possible to "eyeball" a viable by scatter gate.
3)  Try using FDA to determine a viable scatter gate by backgating on FDA fluorescence.
If you are loking at protoplasts later than 24 hr post transformation, the FDA can
take longer than fresh cells to convert to fluorescent signal.
4)  By looking at your fluorescence vs FSC plots it looks like you do have some GFP+
cells in R5.  I would suggest simply that the transfection efficiency of your system
is very poor-have you checked your cells under a microscope.
5)  If your GFP fluorescence is low and you are concerned about overlapping with
autofluorescence, then display an FL1 vs FL2 plot.  Autofluorescence normally runs on the
diagonal whilst you GFP SP cells will sit underneath this once compensation is applied.


Feel free to contact me directly for more help

Gill Webster PhD
Head of Flow Cytometry
Genesis Research and Development
PO Box 50
Auckland

email: g.webster@genesis.co.nz
direct dial 00 61 9 3735600
fax: 00 61 9 373 2189

This message contains confidential information and is intended only for the named
addressees.  If you are not the named addressee you should not disseminate, distribute
or copy this e-mail.



-----Original Message-----


> ----------
> From:		Ibtissam Abdul-Jabbar
> Sent:		Wednesday, June 5, 2002 10:12 PM
> To:	Cytometry Mailing List
> Subject:	Auto-fluorescence
>
> <<File: autofluorescence.doc>>
> Dear experts,  it is the old and reoccurring problem.  I have to compare
> the expression of GFP protein in protoplasts transfected with plasmids
> carrying the GFP modified gene, In addition to the difficulty in
> standardizing the method I am facing the huge difficulty in distinguishing
> and minimizing the auto fluorescence signal in the FL1 region. cells are
> big, 30ul in diameter at least, and they get bigger after transfection so
> the real negative is transfected with or without the tested plasmid. Using
> the 70 Nozzle and triggering on SSC machine (MoFlo) on SSC using log value
> for the FSC and FL1,
> the question is there any way to treat the cell with crystal violate or
> trypan blue to reduce the auto fluorescence. other option to gated the
> data in differently (please see attach for the gating strategy I am
> using),  to use other filters, or to stain the cell with anti-GFP-FITC to
> enhances the differences. I will deeply appreciate your input in this
> meter.
>  <<autofluorescence.doc>>
> many thanks
> yours
> iaj
> please use the above metioned email address.
>
>



This archive was generated by hypermail 2b29 : Sun Jan 05 2003 - 19:26:12 EST