Smiling bands in Western blot - revisiting a widely asked question - (Nov/18/2014 )

So, I've gone through the search function on Bioforums to see what solutions/discussions there is in the past regarding this topic. Would be great if someone can suggest me some possible causes.

The smiling bands I'm seeing of late in my Western blots are causing the neighboring bands to merge to form a connected series of 'smiles'. I'm only seeing this in my beta tubulin housekeeper protein bands and not my proteins of interest (larger than the beta tubulin).

This observation occurs randomly and sometimes I get really nice bands for b-tubulin. I've been using the same Laemmli buffer frozen stock, same antibody aliquot, same antibody dilutions (both primary and HRP secondary), same buffer compositions, same PAGE tank and same running conditions.

The weirdest thing is that when I ran two gels concurrently in the same tank, one blot turned out perfectly fine and the other had the 'smiles' for b-tubulin. Can I safely say it's a gel problem? I make my gels fresh every time. Both gels were made using the same recipe from the same tube. 

Here are the specifics:

Running buffer: Tris-glycine

Run voltage: 100V (constant)

Sample volume per well: 20ul per 15 well 1.5mm thick gels

Gel composition: 10% resolving (1.5M Tris-HCl pH8.0), 5% stacking (0.5M Tris-HCl pH6.8)

Any advice?

Check your gels for leaks.

Check your gels for leaks.

is the gel with smiles the first one loaded?

the sample will start to diffuse immediately after application so the longer it sits the more sideways diffusion will occur. smaller molecules will diffuse faster than larger ones and more abundant molecules will take precedence over less abundant.