Upside-down smiley occurs when running SDS PAGE - (Mar/12/2012 )

I have been getting the same problem with my SDS PAGE's a couple times in a row now and am getting very frustrated .

When running the gel (is a 10% gel) at 110 V, an "upside-down" smiley effect occurs, where the protein in wells 1-3 and 8-10 travel faster than wells 4-7 (see attached photo).

Does anyone have an idea of what could be causing this? I use 1X SDS Running Buffer that is FRESH, and also use fresh SDS Loading Dye.

I had added 100 uL SDS loading dye to my pellets, sonicated, centrifuged, boiled, centrifuged, and loaded 15 uL of each sample.

I talked with our gel specialist, his first reaction is that it looks like there is an issue with your current. The current that is running through your gel is facilitated by the running buffer so you need to check:
<*>Is your electrode assembly leaking buffer? If the wells aren't totally covered with buffer then you will get that smiling effect. Typically if you are experiencing a leak the outside portions of the gel will be the last parts to leak and thus your samples will run faster on the sides. It looks like you are using a bio-rad gel box, make sure the gaskets are seated properly.
<*>Are you using the correct buffer for the type of gel you are using (bis-tris, tris-glycine, etc.)?

It also looks like leaking around the edges to me, however, other possibilities include:
1) bad gels - are these precast? it looks like the invitrogen precast style chamber to me. Precast and self-made gels that have been stored do go off, and tend to show funny running patterns.
2) Incorrect buffering - check your buffer concentration and components, especially if it is a communal stock.
3)Uneven heating of the gel causes these sorts of things, this is usually related to having the worng buffer or running at too high a current, which shouldn't be a problem in your case.

I have seen that precast gels that are past their expiration date, or sometimes very close, do this, as the poster above said. If you don't have leaking around the edges, that's my guess. If for some reason you think you might have uneven heating, you can place your whole gel assembly on ice. I have seen this work also, especially if you are running the gel at a really high current.