Bleed through of membrane label DiI into green channel

[Mel Symeonides]

Hello,

I am using a DeltaVision microscope with stock filters to image fixed cells labeled with the lipophilic tracer DiI and the cytoplasmic label Cell Tracker Green CMFDA. There are two co-cultured cell populations previously labeled with either dye and I am looking for double-labeled cells as a result of exchange of membrane or cytoplasm between cells.

I believe I am experiencing bleed through of DiI into the FITC channel and want to see whether anyone else has experienced this or could predict whether there would be bleed through based on its spectral profile.

The DeltaVision FITC/green filter has the following properties:
Excitation: 490/20nm (laser at 488nm)
Emission: 528/38nm

If you load DiI up on Invitrogen's Fluorescence SpectraViewer and plug in the filter spectra above, you should notice two things: DiI will excite to around 20%, and around 5-10% of DiI emission will be collected by this filter. I think I am interpreting this correctly.


What I actually experience during imaging is that in locations of very high levels of DiI fluorescence, there is low level signal in the FITC channel where there shouldn't be any of the green CMFDA dye. This signal precisely co-localizes with the strongest areas of DiI fluorescence, which leads me to believe that it is bleed-through. Moreover, I have quantified the level of the bled-through signal to around 0.5-1% of the corresponding level in the T-RED channel, which I use to excite and collect DiI fluorescence. This proportion is in line with what I would expect if the FITC channel excites DiI by 20% and collects 5% of that, if Fluorescence SpectraViewer is accurate. (20% * 5% = 1%)

At 1%, it should not be an issue to simply scale it out, however there are two compounding issues: Firstly, DiI is an extremely strong label, which increases the level of bleed through, and secondly, Cell Tracker Green labels very weakly, to the extent that the DiI bleed through is sometimes stronger than the real green fluorescence! I am in the process of optimizing the green label in order to boost the signal.


All this makes sense to me, but I wanted to see if I am getting this right. I have been offered the alternative explanation that the areas of "bleed through" are actually autofluorescence due to the cells taking up material from the environment, or the cells being unhappy due to manipulation. However, the co-localization of regions of intense DiI and non-specific green signal, as well as the spectral profile of DiI, suggest bleed-through to me.


Thanks for reading!
Mel

[Tamara Howard]

I may have missed something, but have you actually looked
at cells labeled ONLY with DiI and checked the other
channel under the imaging conditions you are using with
the CMFDA? That would tell you right away if you are
seeing bleed-through...it should be standard practice to
always run a set of single labels when trying to do
doubles/co-localizations - that way you'll know if you
have any issues, instead of just guessing/assuming!

Tamara

On Mon, 26 Apr 2010 15:10:39 -0700
  Mel Symeonides <[hidden email]> wrote:

> Hello,
>
> I am using a DeltaVision microscope with stock filters
>to image fixed cells
> labeled with the lipophilic tracer DiI and the
>cytoplasmic label Cell
> Tracker Green CMFDA. There are two co-cultured cell
>populations previously
> labeled with either dye and I am looking for
>double-labeled cells as a
> result of exchange of membrane or cytoplasm between
>cells.
>
> I believe I am experiencing bleed through of DiI into
>the FITC channel and
> want to see whether anyone else has experienced this or
>could predict
> whether there would be bleed through based on its
>spectral profile.
>
> The DeltaVision FITC/green filter has the following
>properties:
> Excitation: 490/20nm (laser at 488nm)
> Emission: 528/38nm
>
> If you load DiI up on Invitrogen's Fluorescence
>SpectraViewer and plug in
> the filter spectra above, you should notice two things:
>DiI will excite to
> around 20%, and around 5-10% of DiI emission will be
>collected by this
> filter. I think I am interpreting this correctly.
>
>
> What I actually experience during imaging is that in
>locations of very high
> levels of DiI fluorescence, there is low level signal in
>the FITC channel
> where there shouldn't be any of the green CMFDA dye.
>This signal precisely
> co-localizes with the strongest areas of DiI
>fluorescence, which leads me to
> believe that it is bleed-through. Moreover, I have
>quantified the level of
> the bled-through signal to around 0.5-1% of the
>corresponding level in the
> T-RED channel, which I use to excite and collect DiI
>fluorescence. This
> proportion is in line with what I would expect if the
>FITC channel excites
> DiI by 20% and collects 5% of that, if Fluorescence
>SpectraViewer is
> accurate. (20% * 5% = 1%)
>
> At 1%, it should not be an issue to simply scale it out,
>however there are
> two compounding issues: Firstly, DiI is an extremely
>strong label, which
> increases the level of bleed through, and secondly, Cell
>Tracker Green
> labels very weakly, to the extent that the DiI bleed
>through is sometimes
> stronger than the real green fluorescence! I am in the
>process of optimizing
> the green label in order to boost the signal.
>
>
> All this makes sense to me, but I wanted to see if I am
>getting this right.
> I have been offered the alternative explanation that the
>areas of "bleed
> through" are actually autofluorescence due to the cells
>taking up material
> from the environment, or the cells being unhappy due to
>manipulation.
> However, the co-localization of regions of intense DiI
>and non-specific
> green signal, as well as the spectral profile of DiI,
>suggest bleed-through
> to me.
>
>
> Thanks for reading!
> Mel
> --
> View this message in context:
>http://confocal-microscopy-list.588098.n2.nabble.com/Bleed-through-of-membrane-label-DiI-into-green-channel-tp4965362p4965362.html
> Sent from the Confocal Microscopy List mailing list
>archive at Nabble.com.

***************************
Tamara Howard
Cell Biology & Physiology
UNM-HSC
Albuquerque, NM
***************************

[John Oreopoulos]

Based on the original comments about quantifying bleed-through signals, it sounds like Mel has made some attempt to look at singly labeled control samples of Cell Tracker Green and DiI separately. To further Tamara's suggestions, if you want to check for autofluorescence exclusively, try imaging a completely unlabelled sample with the same acquisition settings for both channels (preferably on the same day with identical preparation procedures minus the labelling step for the unlabelled sample).

One way you could verify/confirm if the cause is really signal bleed-through into the opposite channel due to DiI fluorescence would be to reverse the colour of the probes by using spectral variants of the same labels. For DiI, you could switch to the green emitting variant DiO, and I believe there is a probe called Cell Tracker Red (no commercial interest for wherever you get these probes). Check and see if your colocalization coefficients change under these conditions. There are also longer wavelength versions of DiI, DiD and DiR, which might be compatible with your current Texas Red filters.

If signal bleed through is the culprit, short of purchasing better filters optimized for these two dyes or switching probes, you could try linear unmixing or spectral imaging if your department has access to a confocal microscope with spectrally resolved detection. Alternatively, your current microscope system might have a shorter wavelength laser which can still sufficiently excite Cell Tracker Green with lesser excitation of DiI.

John Oreopoulos, BSc,

PhD Candidate

University of Toronto

Institute For Biomaterials and Biomedical Engineering

Centre For Studies in Molecular Imaging

On 2010-04-26, at 6:54 PM, Tamara A Howard wrote:

I may have missed something, but have you actually looked
at cells labeled ONLY with DiI and checked the other
channel under the imaging conditions you are using with
the CMFDA? That would tell you right away if you are
seeing bleed-through...it should be standard practice to
always run a set of single labels when trying to do
doubles/co-localizations - that way you'll know if you
have any issues, instead of just guessing/assuming!

Tamara

On Mon, 26 Apr 2010 15:10:39 -0700
 Mel Symeonides <[hidden email]> wrote:

Hello,

I am using a DeltaVision microscope with stock filters

to image fixed cells

labeled with the lipophilic tracer DiI and the

cytoplasmic label Cell

Tracker Green CMFDA. There are two co-cultured cell

populations previously

labeled with either dye and I am looking for

double-labeled cells as a

result of exchange of membrane or cytoplasm between

cells.

I believe I am experiencing bleed through of DiI into

the FITC channel and

want to see whether anyone else has experienced this or

could predict

whether there would be bleed through based on its

spectral profile.

The DeltaVision FITC/green filter has the following

properties:

Excitation: 490/20nm (laser at 488nm)

Emission: 528/38nm

If you load DiI up on Invitrogen's Fluorescence

SpectraViewer and plug in

the filter spectra above, you should notice two things:

DiI will excite to

around 20%, and around 5-10% of DiI emission will be

collected by this

filter. I think I am interpreting this correctly.

What I actually experience during imaging is that in

locations of very high

levels of DiI fluorescence, there is low level signal in

the FITC channel

where there shouldn't be any of the green CMFDA dye.

This signal precisely

co-localizes with the strongest areas of DiI

fluorescence, which leads me to

believe that it is bleed-through. Moreover, I have

quantified the level of

the bled-through signal to around 0.5-1% of the

corresponding level in the

T-RED channel, which I use to excite and collect DiI

fluorescence. This

proportion is in line with what I would expect if the

FITC channel excites

DiI by 20% and collects 5% of that, if Fluorescence

SpectraViewer is

accurate. (20% * 5% = 1%)

At 1%, it should not be an issue to simply scale it out,

however there are

two compounding issues: Firstly, DiI is an extremely

strong label, which

increases the level of bleed through, and secondly, Cell

Tracker Green

labels very weakly, to the extent that the DiI bleed

through is sometimes

stronger than the real green fluorescence! I am in the

process of optimizing

the green label in order to boost the signal.

All this makes sense to me, but I wanted to see if I am

getting this right.

I have been offered the alternative explanation that the

areas of "bleed

through" are actually autofluorescence due to the cells

taking up material

from the environment, or the cells being unhappy due to

manipulation.

However, the co-localization of regions of intense DiI

and non-specific

green signal, as well as the spectral profile of DiI,

suggest bleed-through

to me.

Thanks for reading!

Mel

--

View this message in context:

http://confocal-microscopy-list.588098.n2.nabble.com/Bleed-through-of-membrane-label-DiI-into-green-channel-tp4965362p4965362.html

Sent from the Confocal Microscopy List mailing list

archive at Nabble.com.

***************************
Tamara Howard
Cell Biology & Physiology
UNM-HSC
Albuquerque, NM
***************************

[Mel Symeonides]

Thanks Tamara & John. I have not imaged samples where there are no Cell Tracker Green-labeled cells around, but the phenomenon I am looking for occurs at extremely low frequency so in fact the vast majority of cells are singly labeled anyway. Also the cells are at very low density so that they are well spaced apart unless they are actually interacting as part of the phenomenon. I think at this point however I should really image singly labeled cells as well as unlabeled cells to look for autofluorescence, just to be 100% certain of what I am seeing. If there is no way around the issue I will have to look into the alternative labels you suggested.

Just to clarify though - does my interpretation of the spectral profiles make sense? i.e. should I in fact expect about 1% bleed-through into the green channel? I am fairly new to this so I want to be sure I am thinking along the right tracks.

Mel

[Daniel James White]

Hi Mel,

Begin forwarded message:

>
> Date:    Mon, 26 Apr 2010 20:25:38 -0700
> From:    Mel Symeonides <[hidden email]>
> Subject: Re: Bleed through of membrane label DiI into green channel
>
> Thanks Tamara & John. I have not imaged samples where there are no Cell
> Tracker Green-labeled cells around, but the phenomenon I am looking for
> occurs at extremely low frequency so in fact the vast majority of cells are
> singly labeled anyway. Also the cells are at very low density so that they
> are well spaced apart unless they are actually interacting as part of the
> phenomenon. I think at this point however I should really image singly
> labeled cells as well as unlabeled cells to look for autofluorescence, just
> to be 100% certain of what I am seeing.

thats exactly what you should do.
You must always do the single labelled control to check for bleed through,
especially when you expect to get it by looking at the spectra.
You have interpreted those correctly.

> If there is no way around the issue
> I will have to look into the alternative labels you suggested.

this is by far the best choice.

You can keep the Dil dye and use a a red or even far red dye (far red might be harder for your camera to detect)
so as to minimise the possibility of wrong excitation and wrong emission. (crosstalk and bleed through...)
Or do it the other way around.

Watch out for "blue" dyes leaking into green detection,
and "red" emitting dyes being excited by "blue" excitation light.

Always check the spectra first.

You can use image processing methods to separate these signals,
but thats tricky, and error prone unless you really know what you are doing in a well characterized system.
But why bother with all that?
Its better to get spectrally separated data from the get go by using dyes that are better separated.

>
> Just to clarify though - does my interpretation of the spectral profiles
> make sense? i.e. should I in fact expect about 1% bleed-through into the
> green channel?

Yes and Yes.

> I am fairly new to this so I want to be sure I am thinking
> along the right tracks.

you are already doing much better than many others,
by even recognizing the problem.

I've seen nature and cell papers where this problem is clearly there but ignored.

cheers

Dan



Dr. Daniel James White BSc. (Hons.) PhD
Senior Microscopist / Image Visualisation, Processing and Analysis
Light Microscopy and Image Processing Facilities
Max Planck Institute of Molecular Cell Biology and Genetics
Pfotenhauerstrasse 108
01307 DRESDEN
Germany

[phone] (German Mobile)
[phone] (Work phone at MPI-CBG)
[phone] (Fax MPI-CBG LMF)

http://www.bioimagexd.net  BioImageXD
http://pacific.mpi-cbg.de                Fiji -  is just ImageJ (Batteries Included)
http://www.chalkie.org.uk                Dan's Homepages
https://ifn.mpi-cbg.de  Dresden Imaging Facility Network
dan (at) chalkie.org.uk
( white (at) mpi-cbg.de )

[George McNamara]

Hi Mel,

Yes, DiI excites nicely at 488 nm on our confocal scopes (see Li ... Wen et al 2008 Nature Protocols for our major use).

Suggestions:

a) switch to DiD or DiR (or DiA)?

b) DeltaVision is quantitative - bleedthrough should be linear, subtract it.

***

Instead of estimating from a web graph, the spectra data for most fluorophores is at the pubspectra zip file at http://www.sylvester.org/AICF/pubspectra/

The UA graphing site organized by Carl Boswell, http://www.mcb.arizona.edu/ipc/fret/index.html  has a calculations option (for those fluorophores with full details).

If anyone cannot find their favorite fluorophore spectra or filter, etc, in the Pubspectra.xlsx, I am happy to add more data but am about done with digitizing graphs - please send me data files or web links to data.

George



At 06:10 PM 4/26/2010, you wrote:

Hello,

I am using a DeltaVision microscope with stock filters to image fixed cells
labeled with the lipophilic tracer DiI and the cytoplasmic label Cell
Tracker Green CMFDA. There are two co-cultured cell populations previously
labeled with either dye and I am looking for double-labeled cells as a
result of exchange of membrane or cytoplasm between cells.

I believe I am experiencing bleed through of DiI into the FITC channel and
want to see whether anyone else has experienced this or could predict
whether there would be bleed through based on its spectral profile.

The DeltaVision FITC/green filter has the following properties:
Excitation: 490/20nm (laser at 488nm)
Emission: 528/38nm

If you load DiI up on Invitrogen's Fluorescence SpectraViewer and plug in
the filter spectra above, you should notice two things: DiI will excite to
around 20%, and around 5-10% of DiI emission will be collected by this
filter. I think I am interpreting this correctly.


What I actually experience during imaging is that in locations of very high
levels of DiI fluorescence, there is low level signal in the FITC channel
where there shouldn't be any of the green CMFDA dye. This signal precisely
co-localizes with the strongest areas of DiI fluorescence, which leads me to
believe that it is bleed-through. Moreover, I have quantified the level of
the bled-through signal to around 0.5-1% of the corresponding level in the
T-RED channel, which I use to excite and collect DiI fluorescence. This
proportion is in line with what I would expect if the FITC channel excites
DiI by 20% and collects 5% of that, if Fluorescence SpectraViewer is
accurate. (20% * 5% = 1%)

At 1%, it should not be an issue to simply scale it out, however there are
two compounding issues: Firstly, DiI is an extremely strong label, which
increases the level of bleed through, and secondly, Cell Tracker Green
labels very weakly, to the extent that the DiI bleed through is sometimes
stronger than the real green fluorescence! I am in the process of optimizing
the green label in order to boost the signal.


All this makes sense to me, but I wanted to see if I am getting this right.
I have been offered the alternative explanation that the areas of "bleed
through" are actually autofluorescence due to the cells taking up material
from the environment, or the cells being unhappy due to manipulation.
However, the co-localization of regions of intense DiI and non-specific
green signal, as well as the spectral profile of DiI, suggest bleed-through
to me.


Thanks for reading!
Mel
--
View this message in context: http://confocal-microscopy-list.588098.n2.nabble.com/Bleed-through-of-membrane-label-DiI-into-green-channel-tp4965362p4965362.html
Sent from the Confocal Microscopy List mailing list archive at Nabble.com.

George McNamara, Ph.D.
Image Core Manager
Analytical Imaging Core Facility
University of Miami, Miller School of Medicine
Miami, FL 33136
[hidden email]
[hidden email]
305-243-8436 office
http://www.sylvester.org/AICF (Analytical Imaging Core Facility)
http://www.sylvester.org/AICF/pubspectra.zip (the entire 2000+ spectra .xlsx file is in the zip file)
http://home.earthlink.net/~geomcnamara