7 replies to this topic

[science noob]

I am measuring what seems like a rarely expressed protein of interest (140kDa) on western blot.  My solution to this is to load more protein (30-50ug per lane) in order to visualise the band through chemiluminescence.

Problem now is that yes, the protein of interest can now be seen on the blot; BUT the housekeeping protein control - beta-tubulin (55kDa) is now overly bright after exposing chemiluminescence.  This has caused bands from neighbouring lanes to merge horizontally because of the overloading.

Question: Is there a way where I can avoid merging bands and still see my protein of interest (140kDa) and not over-exposing/causing band merging in the beta-tubulin (55kDa).

My post-doc suggested running two sets of time-points in the same gel.  One set in alternate lanes to be ran first for a while followed by another set in the unfilled alternate lanes so that we get a staggered effect and protein bands do not end up side-by-side which causes the merging problem. Has anyone done this before?

[bob1]

That is one solution - you could just take a shorter exposure for the tubulin, and use that as a reference, so long as you keep the exposure times consistent between gels.

[science noob]

bob1, on 21 February 2012 - 04:44 PM, said:

That is one solution - you could just take a shorter exposure for the tubulin, and use that as a reference, so long as you keep the exposure times consistent between gels.

I've tried using very short exposure times for tubulin - still see merged bands across horizontally.  Going shorter than that will not produce a clear image.  I think the problem lies with the amount of protein in the blot.  Maybe use less primary AND/OR secondary?

I've also tried incubating membranes for shorter time in ECL before exposure.

[bob1]

Less primary or secondary might work.  I have also seen people using stacking pieces of film over bright bands, such that the pieces closest to the membrane act as a filter for your actual exposure.

Could it be that your tubulin bands have actually diffused a bit in the gel?  If so, it probably occurred during the incubation between running the gel and setting up the blotting stack.  I try to keep this incubation down to less than 5 min.

[mdfenko]

stacking the film may work with radioactive samples but the film is opaque until processed so the backing film shouldn't get exposed by chemiluminescence (unless the film is processed so that it is translucent and is being used as a density filter).

[science noob]

mdfenko, on 24 February 2012 - 01:01 PM, said:

stacking the film may work with radioactive samples but the film is opaque until processed so the backing film shouldn't get exposed by chemiluminescence (unless the film is processed so that it is translucent and is being used as a density filter).

I don't use films to generate my bands.  I'm using the ECL on blot visualisation (HRP conjugate-substrate reaction).

[mdfenko]

does your instrument allow you to add density filters? or can you adjust the gain? is exposure time the only parameter that can be manipulated?

you can try diluting the ecl reagent.

[Wost]

So your protein of interest (POI) is rarely expressed but you use ECL for detection? Why not simply take a substrate with higher sensitivity and reduce the protein load again?

On a seperate note: If your POI is low abundance and your HKP is (typically) high abundance you might run into a problem with the linear response range of your blotting experiment. In consequence your HKP signal will not reveal any valid information. So why not using total protein load to validate/normalize your western blot result? Some papers describe this approach

The use of total protein stains as loading controls: an alternative to high-abundance single-protein controls in semi-quantitative immunoblotting.

Georgina M Aldridge, David M Podrebarac, William T Greenough, Ivan Jeanne Weiler

Journal of Neuroscience Methods (2008)
Volume: 172, Issue: 2, Publisher: Elsevier, Pages: 250-254