image analysis-distribution of protein

[Anchall ............]

Hello all,
           This query is regarding image analysis.I am lookin for some way I can quantify the following data.
I have images with a protein A usually localised in ER and Golgi.But when I overexpress another protein B which kind of solubilises PROTEIN A then I see diffuse expression of protein A all over the cell basically in cytoplasm.Now what I want to find out is whether the protein is still present in ER & Golgi but it is not so distinguisable as now there is lots in cytoplasm too.Is there any way I can quantify this.. using IMAGE J as i mostly use only that...
I hope some help in this regard..
Thanks
Anchal

[Chris Tully]

Anchal,

Regardless of which software you are using, you need some way of isolating the ER and Golgi from the rest of the cell.  You may want to consider fixing the cells and using a membrane stain for isolate the ER and Golgi based on shape, or you will need to express a third FP in one of the proteins that populates the ER and Golgi membranes.  With a seperate channel it should be relatively easy to derive ROIs from the membrane channel for use in measuring the Protein A distribution in the main image.  Now, this technique is only partially accurate.  Since both the ER and Golgi are relatively small bodies that do not fill the entire vertical space of the cell you really need to look at this problem in 3D.  You will need a lot of slices because you want to get as close to a cubic voxel as possible.

I am sure there are some 3D tools for Image J but I would recommend that you also check out commercial software such as 3D Constructor (for Image-Pro Plus; http://www.mediacy.com), Volocity (http://www.improvision.com/products/volocity/), and Imaris (http://www.bitplane.com/).

Chris Tully

On Thu, Nov 13, 2008 at 10:37 PM, Anchall ............ <[hidden email]> wrote:

Hello all,
           This query is regarding image analysis.I am lookin for some way I can quantify the following data.
I have images with a protein A usually localised in ER and Golgi.But when I overexpress another protein B which kind of solubilises PROTEIN A then I see diffuse expression of protein A all over the cell basically in cytoplasm.Now what I want to find out is whether the protein is still present in ER & Golgi but it is not so distinguisable as now there is lots in cytoplasm too.Is there any way I can quantify this.. using IMAGE J as i mostly use only that...
I hope some help in this regard..
Thanks
Anchal

--
Chris Tully
Microscopy and Image Analysis Expert
[hidden email]
240-888-1021
http://www.linkedin.com/in/christully

[Gudrun Ihrke]

Dear Anchal,
You may want to consider to wash out the cytoplasmic pool of the  
protein by permeabilizing the cells with saponin before fixation. You  
can then compare the membrane-bound pools under your different  
experimental conditions more easily. This method worked for us very  
well [see Savage et al (2002). A digenic disorder of human insulin  
action. Nature Genetics 31: 379-384]. We observed such stark  
differences in our case that we did not need to quantify, but it  
would have been technically feasible.

Gudrun


On Nov 13, 2008, at 10:37 PM, Anchall ............ wrote:

> Hello all,
>             This query is regarding image analysis.I am lookin for  
> some way I can quantify the following data.
>  I have images with a protein A usually localised in ER and  
> Golgi.But when I overexpress another protein B which kind of  
> solubilises PROTEIN A then I see diffuse expression of protein A  
> all over the cell basically in cytoplasm.Now what I want to find  
> out is whether the protein is still present in ER & Golgi but it is  
> not so distinguisable as now there is lots in cytoplasm too.Is  
> there any way I can quantify this.. using IMAGE J as i mostly use  
> only that...
>  I hope some help in this regard..
>  Thanks
>  Anchal
>

____________________________________________
Gudrun Ihrke, Ph.D., CIV
Research Assistant Professor
Department of Pharmacology (C2025)
Uniformed Services University School of Medicine
4301 Jones Bridge Road
Bethesda, MD  20814
[phone]
[hidden email]

[xavier Sanjuan]

Another option wolud be the use of digitonin instead of saponin. It
permeabilizes cholesterol-rich membranes (= plasma membrane but not
intracellular organelle membranes) when used at appropriate concentrations.
See Lorenz et al. Fluorescence protease protection of GFP chimeras to reveal
protein topology and subcellular localization. Nature Methods - 3, 205 - 210
(2006).

Best regards,

Xavi.

___________________________________

Xavier Sanjuan
Servei de Microscòpia Confocal
Departament de Ciències Experimentals i de la Salut
Universitat Pompeu Fabra
Parc de Recerca Biomèdica de Barcelona
Doctor Aiguader, 88 - Sala 309
08003 Barcelona - Spain

Tel.:  + 34 93 316 08 64
Fax: + 34 93 316 09 01
E-mail: [hidden email]
Web: http://www.upf.edu/cexs/sct

-----Mensaje original-----
De: Confocal Microscopy List [mailto:[hidden email]] En
nombre de Gudrun Ihrke
Enviado el: sábado, 15 de noviembre de 2008 2:04
Para: [hidden email]
Asunto: Re: image analysis-distribution of protein

Dear Anchal,
You may want to consider to wash out the cytoplasmic pool of the  
protein by permeabilizing the cells with saponin before fixation. You  
can then compare the membrane-bound pools under your different  
experimental conditions more easily. This method worked for us very  
well [see Savage et al (2002). A digenic disorder of human insulin  
action. Nature Genetics 31: 379-384]. We observed such stark  
differences in our case that we did not need to quantify, but it  
would have been technically feasible.

Gudrun


On Nov 13, 2008, at 10:37 PM, Anchall ............ wrote:

> Hello all,
>             This query is regarding image analysis.I am lookin for  
> some way I can quantify the following data.
>  I have images with a protein A usually localised in ER and  
> Golgi.But when I overexpress another protein B which kind of  
> solubilises PROTEIN A then I see diffuse expression of protein A  
> all over the cell basically in cytoplasm.Now what I want to find  
> out is whether the protein is still present in ER & Golgi but it is  
> not so distinguisable as now there is lots in cytoplasm too.Is  
> there any way I can quantify this.. using IMAGE J as i mostly use  
> only that...
>  I hope some help in this regard..
>  Thanks
>  Anchal
>

____________________________________________
Gudrun Ihrke, Ph.D., CIV
Research Assistant Professor
Department of Pharmacology (C2025)
Uniformed Services University School of Medicine
4301 Jones Bridge Road
Bethesda, MD  20814
[phone]
[hidden email]

[mahogny]

Chris Tully wrote:

> Anchal,
>
> Regardless of which software you are using, you need some way of
> isolating the ER and Golgi from the rest of the cell.  You may want to
> consider fixing the cells and using a membrane stain for isolate the
> ER and Golgi based on shape, or you will need to express a third FP in
> one of the proteins that populates the ER and Golgi membranes.  With a
> seperate channel it should be relatively easy to derive ROIs from the
> membrane channel for use in measuring the Protein A distribution in
> the main image.  Now, this technique is only partially accurate.
> Since both the ER and Golgi are relatively small bodies that do not
> fill the entire vertical space of the cell you really need to look at
> this problem in 3D.  You will need a lot of slices because you want to
> get as close to a cubic voxel as possible.
>
> I am sure there are some 3D tools for Image J but I would recommend
> that you also check out commercial software such as 3D Constructor
> (for Image-Pro Plus; http://www.mediacy.com <http://www.mediacy.com>),
> Volocity (http://www.improvision.com/products/volocity/), and Imaris
> (http://www.bitplane.com/).
>

or the newer open source packages, Endrov, OME and Bioimagexd

/Johan

--
--
------------------------------------------------
Johan Henriksson
MSc Engineering
PhD student, Karolinska Institutet
http://mahogny.areta.org http://www.endrov.net

[David Basiji]

***Commercial response***

Johan, et al.,

We do something similar to what you describe in terms of labeling
the structure of interest and then using that image to determine an ROI
for assessing localization or another label to that structure. I
agree that the absolute error in doing this with a 2D projection of a
single cell may be high due to orientation and focus considerations.
However, it is not necessary to form a 3D image if you have the ability
to measure enough individual cells in 2D, whether using our system or
any other 2D imaging platform. Rather than relying on the
localization measurement of a single cell, if you assess localization by
measuring the distribution of measurements over many cells the
assessment
will be very robust. The error in each individual measurement will tend
to
broaden the distribution but since the odds of all the cumulative errors
being
in the "same direction" will be very low, the mean of the distribution
will
be very consistent.

Best Regards,
David


David Basiji, Ph.D.
President and CEO, Amnis Corporation
2505 Third Ave., Suite 210
Seattle, WA 98121
[phone] direct
[phone] mobile
[phone] fax
www.amnis.com
 
This email and any attachment contain information which is intended for
the addressees only. If you have received this email in error, please
notify the sender.
 

-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]]
On Behalf Of Johan Henriksson
Sent: Monday, November 17, 2008 3:44 AM
To: [hidden email]
Subject: Re: image analysis-distribution of protein

Chris Tully wrote:
> Anchal,
>
> Regardless of which software you are using, you need some way of
> isolating the ER and Golgi from the rest of the cell.  You may want to

> consider fixing the cells and using a membrane stain for isolate the
> ER and Golgi based on shape, or you will need to express a third FP in

> one of the proteins that populates the ER and Golgi membranes.  With a

> seperate channel it should be relatively easy to derive ROIs from the
> membrane channel for use in measuring the Protein A distribution in
> the main image.  Now, this technique is only partially accurate.
> Since both the ER and Golgi are relatively small bodies that do not
> fill the entire vertical space of the cell you really need to look at
> this problem in 3D.  You will need a lot of slices because you want to

> get as close to a cubic voxel as possible.
>
> I am sure there are some 3D tools for Image J but I would recommend
> that you also check out commercial software such as 3D Constructor
> (for Image-Pro Plus; http://www.mediacy.com <http://www.mediacy.com>),

> Volocity (http://www.improvision.com/products/volocity/), and Imaris
> (http://www.bitplane.com/).
>
or the newer open source packages, Endrov, OME and Bioimagexd

/Johan

--
--
------------------------------------------------
Johan Henriksson
MSc Engineering
PhD student, Karolinska Institutet
http://mahogny.areta.org http://www.endrov.net
###########################################

This email and any attachment may contain information which is private
and confidential and is intended for the addressee only.
If you have received this email in error, please destroy it
and notify the sender by return email.