Beautiful bands in wrong place! (Western)

[Graeme]

Hi. I'm blotting for a membrane channel, approx 150kDa and trying to confirm species cross-reactivity (human --> ferret). I'm getting a lovely, reproducible & concentration-dependent band in all my species, including mouse that has confirmed cross-reactivity. The only trouble is it's about 40kDa too low, with no band at the expected position! It vanishes when I pre-incubate as a negative control, so I'm pretty sure it's the primary rather than the secondary that's reacting.

Is it possible that my denaturing (5 mins, 95C in 1/20 Betamercap) or freezing (6 months @ -80, 4 weeks @ -20) could have given me a cleaved fragment? Or could my pre-stained marker (Kalaidescope) be giving me a misleading migration?

I'm using 12% SDS, overnight wet transfer @30V, and overnight non-specific and primary blots until I can get it working. My prestained marker seems to transfer over fine, and ponceu looks like there should be a reasonable amount of protein all the way up the lane. I'm waiting for a different ladder to arrive to see if that helps, but any other suggestions would be gratefully received!

[Neurotrophin]

Just a thought. Are there confirmed/predicted glycosylation sites in the human primary amino acid sequence? Are these sites conserved in the ferret sequence? It's possible this protein is not processed the same in ferret cells.
----
Instead of buying a new ladder, it may have been easier to just blot for any protein that runs around 150 kD and see if the ladder looks right.
-----
Have you cloned the ferret cDNA? If so, how about epitope tagging it and expressing it in a human cell line (293 for example) and see where it runs both with your antibody and an antibody against the tag. I would do the same experiment in a ferret cell line if you have one.

Good Luck.

[Compton]

I saved this piece of text from somewhere. Don't remember if I'm quoting another forum user or if it was qiagen support pages but anyway, here goes:

SDS-PAGE is not a reliable method for accurately determining the molecular weight of proteins. This is why the term apparent molecular weight is often used. The molecular weight estimated by SDS-PAGE can be very different from the actual molecular weight. Matagne et al. reported 55 kDa for a 29-kDa b-lactamase (1), whereas the apparent weight of a 114-kDa protein was measured as 180 kDa (2). Separation of proteins by SDS-PAGE depends on the uniform binding of negatively charged SDS to the protein to give a constant charge-to-mass ratio, but in reality the amount of SDS bound can vary for different proteins. Many factors can affect mobility on SDS polyacrylamide gels, such as hydrophobicity, charge, glycosylation (for proteins expressed in eukaryotic expression systems), and phosphorylation (3). In some proteins, even substitution of single amino acids has been reported to result in altered mobility (4). Also glycolysation and phosphorylation e.g. in insect cells can cause altered mobility
References
1. Matagne, A., Joris, B., and Frere, J.-M. (1991) Anomalous behaviour of a protein during SDS/PAGE corrected chemical modifcation of carboxylic groups. Biochem. J. 280, 553.
2. Casaregola, S., Jacq, A., Laoudj, J., McGurk, G., Margarson, S., Tempete, M., Norris, V., and Holland, Cloning and analysis of the entire Escherichia coli ams gene. J. Mol. Biol. 228, 30.
3. Hames, B.D. (1990) One-dimensional polyacrylamide gel electrophoresis. In: Hames, B.D. and Rickwood,
Gel electrophoresis of proteins: a practical approach. Oxford: Oxford University Press, p 1.
4. Fasano, O., Alfdrich, T., Tamanoi, F., Taparowsky, E., Furth, M., and Wigler, M. (1984) Analysis of the potential of the human H-ras gene by random mutagenesis. Proc. Natl. Acad. Sci. USA 81, 4008.

[kmunson779]

From a Google search, this text comes from the Qiagen support pages.

[marcin]

Could be regulatory splice variant. Try RT-PCR for smaller mRNA.

[MolBioGangsta]

Is your antibody monoclonal?

[sarpeyer]

Your membrane protein may not be fully denaturing. Porin for instance needs to be boiled in phenol before it runs at its correct molecular weight! Membrane proteins can readily aggregate due to the hydrophobicity of their transmembrane regions and so assume a compact aggregated form when run on a gel - which runs faster than the fully denatured protein (hence low apparent molecular weight). You could do worse than follow the same denaturation steps as you would for something like Porin and see what happens. Sorry, I don't have the reference to hand.

[mwbradbury]

One thing to know is if the band is in the correct place for other species. If it is, one possible source of the problem is cleavage of an Asp-Pro sequence. This has been reported to happen when denaturing proteins using the conditions you mention, but is more likely if the pH is a bit too low. I have seen this with a GFP fusion protein that had the Asp-Pro sequence linking it to my protein due to the cloning vector I used. An antibody against GFP gave me a band fro the fusion protein and one just larger than standard GFP in the same lane, so some was most likely cleaved and some remained intact. If your antibody does not recognize the second fragment of about 40 kD, you will only see the one band. You can try preparing more sample with a strict attention to pH before and during denaturation, and you can use a lower temperature. I saw a recommendation that 65 C is as good as 95 C for denaturation, and have tried it. It seems to work, the gels look similar, and it probably lessens the chance of cleavage, though I have not tried it on a sample that I know has the Asp-Pro sequence.

[yeaster]

If I were you, I'll not do any denature steps for a membrane protein. Add Sample Buffer only and run the gel.

[tyronegenade]

Hi. I'm blotting for a membrane channel... I'm using 12% SDS...

1. Repeated freeze thawing can cleave proteins; and the loading dye, if it contains beta-mercaptoethanol, can break disulfide bridges which means that is the 150 kDa protein is a complex of two different proteins held together by a disulfide bridge then yes, boiling can break them apart. But I feel this is unlikely to be the case.

2. If this is a membrane protein then your choice of extraction buffer can be the problem. For membrane bound proteins I use 0.1 m TRIS, 1% Nonidet P-40 substitute (Sigma, 74385), 0.01% SDS, 1 µg/mL Aprotinin (Roche, 10236624001) and 0.1 µm PMSF. Using a standard RIPA buffer with Triton X-100 and Na-deoxycholate I do not get good extraction of lipid based proteins. Whether this is because the high detergent concentration simply prevents micelle formation or solubilizes so much more protein that the target protein is too dilute I do not know. So, what extraction buffer are you using?

3. 12% acrylamide is too high a concentration for such a high molecular weight protein. You should be using a 7.5% gel. See below for the correct SDS concentration for protein resolution:
5%: 60--220 kDa
7.5%: 45--120 kDa
10%: 25--75 kDa
12%: 14.4--65 kDa
15%: 6.5--45 kDa
17.5%: 5.5--30 kDa
You can resolve a 150 kDa band with a 12% gel but in blotting very little of the protein will move out of the gel onto the blot. If you don't believe me (and you shouldn't) stain the blot with ponceau S and the gel with coomassie blue after transfer and compare the transfer efficiency.

I suspect that your protein was either not extracted properly or that it never left the gel. In any event, when you repeat with a different extraction buffer or low concentration gel you may get your 150 kDa band as well as the 110 kDa band. Using protein sequencing you would then need to prove that this is fragment of the target protein and not simply cross-reactivity which would invalidate your antibody.

Good luck

[AMH]

Proteins often migrate differently from their predicted molecular mass, but 40 kDa is a big shift. If I were you, I would first consult the data sheet that came with the antibody. You should be able to find it online. How big do they say it is? Do they say if it migrates at 110 kda or 150 kDa? See if the company provides an image of a sample Western blot. How does the proteim migrate there? Is it an endogenous protein or recombinant? If it is a tagged recombinant protein, its mobility will be different from the natural version. If you come up empty here, search the literature for Western blots of your protein. How big is it on those blots? If your protein is migrating differently from theirs, see what lysis buffers they use and consider changing your detergent. Also, 12% acrylamide really is too high for a 150 kda protein.