BrdU and GFP

From: Rob Wechsler-Reya (rw.reya@duke.edu)
Date: Thu Feb 14 2002 - 16:41:30 EST


Hi. We've been doing some staining for BrdU incorporation in
GFP-expressing cells, and have been getting some strange results.
I'm hoping someone out there might have a solution, or at least an
explanation.

The assay goes like this:
- Plate cells (neuronal precursors) in a 24-well plate and infect
with retroviruses carrying Gene X-IRES-GFP (with or without mitogens)
- At 48 hours, pulse with BrdU
- The next morning, harvest the cells (they're adherent, but a brief
papain treatment gets them off)
- Fix with 1-4% paraformaldehyde (2% seems to work the best, but
we've tried them all)
- Permeabilize with 0.2% Triton X-100 (or Tween-20)
- Treat with DNase to expose BrdU epitopes (can't use HCl, because it
kills GFP)
- Stain with anti-BrdU (PE-conjugated) and anti-GFP (FITC-conjugated)
- Analyze by FACS for BrdU and GFP expression

Regardless of whether the virus is empty or contains something that
should induce cell cycle progression, we get the same results:  some
cells that are GFP-positive/BrdU-negative, some cells that are
GFP-negative/BrdU-positive, but none that are double-positive. The
fact that we're getting both types of cells in a given sample
suggests that the BrdU staining is working, and that the GFP hasn't
leaked out of the cell.  Another possibility is that the viruses (or
GFP) are toxic to the cells, and any cell that has GFP is unable to
proliferate.  However, when I do essentially the same experiment
using immunofluorescent (IF) staining of cells on coverslips, I get
plenty of double-labelled cells.  (The results of that assay are very
clear, but much more difficult to quantitate, which is why we've been
trying to do it by FACS.)  The only differences I can come up with
between the assays are: (1) for IF, the cells are growing on
coverslips instead of on plastic; (2) for IF, cells don't get
papained off before staining; (3) for IF, we use biotin-anti-BrdU +
streptavidin-TRITC (rhodamine) instead of a PE-coupled antibody.
None of these seem likely explanations for the dichotomy.  Has anyone
had this experience?  Anyone want to take a stab at an explanation?

Thanks very much for your help,
Rob

--
Robert Wechsler-Reya, Ph.D.
Department of Pharmacology and Cancer Biology
Box 3813, Duke University Medical Center
LSRC Building, Room C303
Durham, NC 27710
Phone: 919-613-8754
Fax:     919-681-8461



This archive was generated by hypermail 2b29 : Wed Apr 03 2002 - 11:59:24 EST