Autofluorescence Hela

[B. Prabhakar Pandian]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Hello All,

                          We are trying to do some transfection studies with Hela cells. However, we are noticing that even our control cells are giving signal at 488 (FITC).

Is there a way to reduce this auto fluorescence.

 
Thanks,

 
-Prabhakar

[Alfred Bahnson]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

Prabhakar,

We found a way around autofluorescence to pull out the GFP signal.
http://www.nature.com/leu/journal/v24/n1/full/leu2009191a.html

Basically, you acquire two images at every time-point for each location,
one at your FITC excitation (488nm) and another at a higher wavelength
(e.g. 555nm, PE).   In our case, the autofluorescence was similar in both
channels, but the GFP only emitted with the shorter wavelength (488nm)
excitation.  Since motion was negligible between the two images,
subtracting the 555nm image from the 488nm image left only the GFP signal.
 This process also did a nice job of flattening the dome-shaped background.


We included an adjustment factor to first equalize the average intensities
based on a control well with autofluorescing cells but not GFP expressing
cells; in our case the 555nm image had a higher average intensity, so we
multiplied these images by an adjustment factor = (average control
intensity at 488/ average control intensity at 555) before subtracting.

Interestingly, this experiment continued for almost 8 weeks.  Over this
time, our intensity adjustment factor remained practically constant.   But
the average intensity of both the 488 and the 555nm signal changed a lot as
the lamp output (an Xcite metal halide bulb) changed.  So we normalized the
net signal intensity across the entire experiment as well, based on the
average intensities from the control well at each time point.  The metal
halide lamp is more stable than the conventional Hg-arc lamp over time
frames relevant to the difference in time between the control well and the
experimental wells, so this whole process yielded a very satisfactory
result.

Al Bahnson
www.kairosinstruments.com



On Fri, Jul 11, 2014 at 3:50 PM, Prabhakar Pandian <[hidden email]> wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Hello All,
>
>                           We are trying to do some transfection studies
> with Hela cells. However, we are noticing that even our control cells are
> giving signal at 488 (FITC).
>
> Is there a way to reduce this auto fluorescence.
>
>
> Thanks,
>
>
> -Prabhakar
>
>
>

[George McNamara]

*****
To join, leave or search the confocal microscopy listserv, go to:
http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
Post images on http://www.imgur.com and include the link in your posting.
*****

1.  Grow your cells in Evrogen's DMEMgfp or Marker Gene Tech's Opti-Klear.
2. acquire at an "autofluorescence only" wavelength, and use that to
ratio or difference out the autofluorescence. See another reply, which
recommended PE channel, or my recommendation of "use the CFP cube or
channel" (440 or 457 nm laser, <500 nm emission if confocal).
3. Get Adam Hoppe et al's 3DFSR (it's not only for FRET) and use that
(61 lines of MatLab code),
http://www.ncbi.nlm.nih.gov/pubmed/18339754
... we are close to submitting to Intel's IPCC a proposal (open source
output is a condition: a great feature of the IPCC's) to parallelize and
otherwise optimize this for Xeon CPUs, Xeon Phi cards -- hopefully
including TACC Stampede https://www.tacc.utexas.edu/stampede/ (free
access to UT staff, part of an NSF program, so many researchers have
access, held steady at #7 in the recent Top500), and I am especially
psyched about Xeon&Phi optimizations being instantly portable to the
Knights Landing CPU (3 Teraflops, double precision) in mid to late 2015
-- both local KNL(s) and I figure TACC Stampede will get a rack or aisle
of these. 3DFSR should also be optimizable for GPU's, so if you are good
at MatLab parallel programming (or can find someone who is, or maybe
just need their paprallel library and options), you could port 3DFSR to
TITAN Z card(s) (8 Teraflops single precision, 4 Teraflops double
precision, available now and of course having "Z" in the name is great
for 3D and multi-D microscopy).

sincerely,

George
p.s. more on KNL at
http://vr-zone.com/articles/intel-unveils-knights-landing/79686.html
http://vr-zone.com/articles/xeon-phi-knights-series-continues-landing-2015/64112.html 

I just came back from the TACC hosted IXPug 2014 meeting and many of the
presentations are already online at
https://www.ixpug.org/meetings/2014-user-meeting
one not online is Carson Brownlee's ray tracing presentation - abstract at
https://www.ixpug.org/TACC-talks
which with iPython Notebook     http://ipython.org/notebook.html   , R,
etc, running on TACC and similar supercomputers, and sending just the
visual output to our desktops, could revolutionize how we view our data.

On 7/11/2014 2:50 PM, Prabhakar Pandian wrote:

> *****
> To join, leave or search the confocal microscopy listserv, go to:
> http://lists.umn.edu/cgi-bin/wa?A0=confocalmicroscopy
> Post images on http://www.imgur.com and include the link in your posting.
> *****
>
> Hello All,
>
>                            We are trying to do some transfection studies with Hela cells. However, we are noticing that even our control cells are giving signal at 488 (FITC).
>
> Is there a way to reduce this auto fluorescence.
>
>  
> Thanks,
>
>  
> -Prabhakar
>
>
>
>    

--



George McNamara, Ph.D.
Single Cells Analyst
L.J.N. Cooper Lab
University of Texas M.D. Anderson Cancer Center
Houston, TX 77054
Tattletales http://works.bepress.com/gmcnamara/42