 |
 |
 |
 |
Below are a collection of internet messages which related to this subject back
in October.
In response to bulletin board messages and personal mail, I thought
I'd give a summary of the information I have gleaned from the
internet and perhaps stimulate more replies.
My interest centres on the potential use of GFP in tagging CD34+
cells using a retroviral vector. For a number of years I have been
using the lacZ gene - which works but has a number of problems
including these 2:
the only substrate sensitive enough for detection is FDG, which
requires long cold incubations which isn't ideal if you have to
reculture the cells (fatal for some types). In my experience the
other fluorogenic substrates are no good for flow (unless you're
working with sperm!)
the gene is large (~3kb) - not the best thing for a retroviral vector
- especially when you want to stick other genes in too
These problems are overcome using GFP -its small and intrinsically
fluorescent -but would it work in this application? The
requirements here are stringent: single integrated copy in rapidly
dividing human primary cells .
Here are some responses old and new ( I hope the sender doesn't mind
me doing this). Here goes
From: Mario Roederer <ROEDERER@Darwin.Stanford.EDU>
People in our lab have measured GFP in cells from single-copy vectors...
you need very good promoters to get separation from background.
We have also expressed the brighter mutant GFP... and yes, it is indeed
5x brighter than the other version! We made the gene by PCR-directed
mutagenesis of the wt version.
---------------------------------------------------------------------------
unfortunately he doesn't say what type of cells and if rapidly
dividing . Mutants now available from Clontech but....
From: PATW@clontech.com
Subject: RSGFP-technical enquiry -Reply
I have spoken with our scientists and as far as they are aware, there is
still not a definitive case of stable GFP expression from a single
integrated copy. They also do not know if the RSGFP can give the level
of sensitivity that detecting expression from a single copy would require.
-----------------------------------------
and from MR again:
We have been pursuing the GFP as a reporter gene, and can compare it to
our extensive experience with lacZ/FDG.
There is no way that the GFP will EVER be as sensitive as lacZ/FDG. We
were able to run the FACS-Gal assay down to 5 molecules of lacZ per
cell (!). Typically, at least 20-30 are needed for minimal detection
in most cell lines; by the time you get to 100 per cell, you have a
reasonably bright system!
100 GFP's are unlikely to be detectable in a single cell.
That said, GFP is, however, quite useful. Especially the more recently
published mutants which are considerably "brighter"... with a
reasonable strong promoter, expression can be easily detected with
positively-expressing cells 1-2 orders of magnitude above
autofluorescence.... the catalytic activity of lacZ, which allows
single enzyme molecules to generate thousands of fluors, gives that
system an amplification advantage for sensitivity that will not be
overcome by GFP.
------------------------------------------
-I guess the level of expression Mario refers to is from a transient
burst?
----------------------------------------------
CONCLUSION
My guess is GFP will not be helpful to me since I am only just
comfortable with the sensitivity of lacZ. I think the other piece
of evidence worth quoting is that if it could have been done, we
would have seen it by now!
Good hunting
################################################################
From: Dr Richard Darley
LRF Preleukaemia Unit
Department of Haematology
UWCM
Cardiff, CF4 4XN
U.K.
tel (+44) 12 22 74 45 24
Fax (+44) 12 22 74 45 23
e-mail: darley@cardiff.ac.uk
We have been using this same gfp mutant with some success recently.
We started this work on a FACS Vantage with a standard FITC setup
(488nm excitation; 530/30 bp filter) and have seen good fluorescence
from transiently transfected murine cells and from cells infected with
a mutant gfp-encoding retrovirus. We are now using an improved
version of this setup, in which the 530/30 bp filter is replaced with
a 510/20 (Omega Optical). This has the dual advantage of cutting down
background from autofluorescence and increasing the signal from the
gfp. With either setup, the mutant gfp is better/brighter than the
original excited by UV in our hands. If you were able to detect
expression of the original gfp, I don't think you'll have any problems
with the mutant. If you do, let me know.
_______________________________________________________________________________
Subject: Anybody using Green Fl. Protein?
From: owner-cyto-sendout@flowcyt.cyto.purdue.edu at Internet_Gateway
Date: 3/26/96 2:10 PM
We are interested in using it with Flow. any thoughts out there on this?
does it work as advertised? etc.
thanks,
Chris
|
|
|
|