PAGE gel DNA band shape - (Apr/24/2007 )

hi all,

I have had a strange problem that when i do a 6% PAGE gel for 308bp PCR product, i load abt 5microliters of the sample along with the loading buffer. the bands take a slightly curved shape at the ends.Actually the ends of the bands look something like a crescent .I use 1XTAE only and run it at 50Vconstant using BIORAD miniprotean III.
Can someone also help me with how i decide upon the voltage and mAmps settings for PAGE gel?

thanks and regards,
nirupa k l

the first hing that comes to mind: how long do you allow your samples to sit in the wells before starting the run? if you let them sit too long then they will diffuse into the "teeth" of the wells and will be slowed relative to the rest of the sample in the well. this will cause an apparent curving at the edges.

another possibility is that you need to flush your wells (better) immediately before applying your samples.

but if i had to bet i would go with the first.

Hi nirupakl,

If you look at most RFLP gels using PAGE in the literature, you'll see this common band shape. I ran many such gels years ago and just lived with it because the problem was not severe enough to affect data interpretation. I think it may be caused by the viscosity of the loading buffer, with some of the loaded sample adhering to the sides of the wells and so entering the gel a bit later than the sample in the middle of the wells. The first band loaded looked exactly like the last band loaded, so I don't think the length of time the sample sits in the well is a factor. I also always cleaned the wells out thoroughly before loading. Then I came across some reference where various different loading buffers were tested, and one where ficoll was used instead of glycerol (I was using glycerol) or sucrose gave nice, flat bands. I tried it, and sure enough, I got nice flat bands. This was many years ago, so I have no idea where I saw this article, but try looking for a loading buffer recipe that uses ficoll, and give it a try.

Hey, it might even be something as easy as running too strong of a current through it. Slow it down, let it run longer and see if there is a difference