ief gels: samples jumping lanes?

Phil Hardy philhardy at
Fri Feb 11 16:37:16 EST 2000

Will, the symptoms sound like you have minor spillage from one lane to the
next during gel loading.  I've not used a higher urea content in the
running buffer than in the gel, but for a while I had the same problem on
RNA urea gels.  If there is too much buffer/sample mixing during loading, a
volume of buffer/sample equal to the amount of sample you add spills out of
the top of the well and may sink into the next. This normally happens, but
if there is no sample in the runoff it is of no consequence.  A small
amount of protein leaking over won't be seen with coomassie, but will light
up in  a Western.
    This is how I fixed my problem, maybe you could adapt your protocol
using this info:
Prior to loading, I washed the wells out by squirting running buffer in
them (cleaned out unpolymerized acrylamide and gel bits).  Then waited
about 5-10 minutes before loading.  This allowed urea from the gel to leach
into the buffer in the well.  When I loaded the samples (slowly with gel
loading tip) the sample would sink directly to the bottom of the well - I
think the presence of the high urea conc stopped the sample from mixing
with the buffer.

good luck
phrom phil hardy

Will Fuller wrote:

> Dear all,
> I wonder if anyone can throw any light on a problem I'm having
> interpreting Western Blots from ief gels.
> Briefly, my gels (BioRad mini-Protean II system) are composed of 9.2M
> urea, 2% (each) of TX-100, nonidet P40 and CHAPS, 2% ampholytes, and 4%
> acrylamide. Gel loading buffer is the same except 9.5M urea and no
> acrylamide. I add 50ul of sample (homogenate) to 200ul loading buffer.
> Transfer is semi-dry in 0.7% acetic acid for 2.5 hours at 100mA per
> mini-gel. I can stain my gels after transfer and my PVDF membranes with
> Coomassie after probing.
> Generally there is some protein left in the gels following transfer, and
> it usually looks good, that is to say it runs evenly and there is
> protein in the lanes I've loaded and none in the lanes I haven't. The
> same is true when I stain my pvdf membrane after probing. However, the
> images I get following probing the blots are very confusing. It appears
> that my samples are able to jump from lane to lane. That is, where I
> load a "blank" lane (that is, my loading buffer only) I get a signal.
> This signal seems to be related to the signals from the lanes either
> side of the blank, and is of comparable magnitude. I use a fresh loading
> tip for each well when loading. The "blank" loading buffer is not
> contaminted with protein (I have made it up fresh many times), and of
> course there is no sign of any protein in these blank lanes on my
> stained membranes and transferred gels. What's more I don't believe this
> is a case of lateral diffusion in the gel during running because the
> signals from the blank lanes are always "discrete", that is there is
> always a gap between the signal from the blank lane and that from the
> lanes alongside it.
> I am not loading a large volume of sample for the size of the well (5ul
> in a well that will hold 15ul), and pay particular attention to ensuring
> that when the gels are loaded they are not moved or disturbed in a way
> that might allow sample to "jump" from one lane to the next. When I put
> the film down to get an image from the membranes after probing I am
> certain there is no lateral movement of the film that might create some
> sort of "ghost" image.... I do quite long exposures (about 1-5 minutes)
> in a film cassette.
> I can see no explanation for this phenomenon.... There is no protein
> there according to Coomassie stains, but there is a similar amount of
> the protein I am blotting for in blank and sample lanes according to
> Western analysis. I routinely run SDS PAGE gels and blot them without
> this sort of problem, so it seems to be specific to ief gels.
> If anyone has any tips, experience, and so on I'd be delighted to hear
> from you. This has been going on for some time now and I'm heartily sick
> of it!
> Cheers
> Will

More information about the Methods mailing list