Band is at the wrong size, but same for all antibodies - (May/23/2012 )

I've been consistently getting bands on every western at ~60kDa. I get these bands in every well with protein loaded in it. Sometimes I'll also get a band where I'm expecting one, but sometimes the 60kDa band is the only one. Both bands will be faint, but the 60kDa band is always stronger.

I thought I might be having problems with protein aggregation, so I tried cooking at 70 degrees for 10 minutes rather than boiling my samples, but that didn't help.

I've been using cells collected from mouse or rat spleen, thymus, and uterus. I extract protein with a 1% Tx100 buffer with 150mM NaCl and 50mM Tris, pH 8, with protease inhibitors. Vortex, shake at 4 degrees for one hour, and centrifuge to remove debris. My protein concentrations this way have been in the 1-5mg/mL range (BioRad DC or RC DC assay), and I've been loading 10-50ug per lane. I use precast gels from BioRad.

I've probed for actin, tubulin, Bcl-2 (25kDa), Bcl-xl (also 25kDa, but this one didn't show any bands at all), and a few untested cox-11 antibodies (31kDa; these could be hitting the wrong thing, since they're new and untested, but they shouldn't all be hitting 60kDa). Of these, only actin and Bcl-2 have been hitting the correct size, but it's always a fainter band than the one at 60kDa.

Any idea what's going on?

the 60kDa band is probably an artifact (most likely keratins from dust) which appears more and more prominently as the reducing agent in your sample buffer ages.

try to repeat your western but use a fresh lot of reducing agent (usually 2-mercaptoethanol or dtt).

in fact, for our westerns (and silver stained gels) we were able to omit reducing agent altogether so that we wouldn't see the artifact.

Also, test to see if the band comes from your secondary antibody alone. This is obvious, but mouse spleen should contain mouse immunoglobulins, so any anti-mouse secondary antibody will likely pick up the heavy chain at 50-60kDa.

It could be that the reducing agent is old. Lots of our reagents are old (though I've bought fresh stocks of almost every reagent used in blotting, just not the BME). Problem is, our lab has tons of (old) bottles of BME, so I don't know that my lab manager will allow me to buy new stuff. I'll try it without a reducer (should be fine for actin and other smaller proteins).

I'm pretty sure it isn't native Igs (or any effect of the 2y), since I see the same effect in human cells and rat uterus, and with some primaries, the 60kDa band doesn't show at all (I haven't gotten any bands with Bcl-xl blots yet).

Alright, the reducing agent was part of the problem, but not the whole thing. My bands are a lot sharper and more consistent now, but that's about it.

I've attached an image to show what I'm looking at.

The blots are arranged as mirrors around the ladder in the middle (I was running low on ladder and had to get creative). I'm mostly interested in the two lanes closest to the ladder, where you see the big fat bands. I assume the other samples had low protein extraction efficiency, which I'll work out separately.

3D18 and 4N2 are supposed to bind to a 30kDa protein (you can see it on the 4N2 blot, one lane appears to have a PTM of ~8kDa (possibly ubiquitin)). The 4O18 antibody binds to a 105kDa protein. Yet all proteins, including actin, show big fat bands where they're not supposed to be. In this blot, actin barely shows up at all.

Edit: those two lanes are from mouse thymus, fyi.

what are your conditions for performing the western?

in what buffer(s) are the antibodies prepared?

block?

species of secondary?

we routinely block with bsa + normal serum from the species of the secondary (no tween in the block)

and

we add that serum to our antibody solutions (both primary and secondary)

do you add tween-20 to the buffers? how much? we use 0.2% to help reduce non-specific binding of the antibodies.

Samples are prepared as I describe in my first post. I add 1:1 2x laemmli buffer (4% SDS, 20% glycerol, .004% bromphenol, 125mM Tris-HCl, pH 6.8, BME added to 10% before running), boil for 5 minutes (heating to 70 degrees for 10 minutes does not change the result), and run.

Blocking is with 5% milk in PBST, with .05% Tween. Washes are done in PBST, secondary is goat anti-mouse from SCBT. I can add more Tween to the buffers, and remove it from the blocking, but I'm not sure it would help. Since I'm not seeing a strong actin band, I think something's going wrong before the staining.