gribskov@FCRFV1.NCIFCRF.GOV ("Gribskov, Michael") (09/06/90)
Jim Gordon writes: >Does anyone have any experience with proteins that migrate significantly slower >by SDS-PAGE than their DNA sequence predicts? I have a 22.3 Kd ORF that yields >a 35 Kd polypeptide on gels. I would like to know if someone has similar >experience or some possible explanation(s). I have personally worked with several proteins that show anomalous migration on SDS gels. The first group of proteins were the histones and HMG proteins. I don't recall the direction of the anomaly but it gave molecular weights that were quite incorrect (at least as a fraction), since these are rather small proteins. More recently I have worked with the sigma subunit of E. coli RNA polymerase. The actual molecular weight of the protein is about 70K but it runs as about 90K on SDS gels. These proteins are all highly charged, the histones are very basic, HMG proteins highly basic and acidic (~50% overall), and sigma is very acidic (net charge -45). In the case of sigma, it is known from examination of proteolytic fragments that the anomalous migration is primarily due to one segment of the protein that contains 18 acidic residues out of 22. My prediction would be that your ORF protein is either acidic, or contains a highly acidic segment. The anomalous migration thus appears to be due to high charge on the protein. The hand-waving explanation is that this affects SDS binding, and therefore compromises the basic assumption of SDS-PAGE: that all proteins have an equal charge density due to complete coating of the molecule with SDS. If there are significant regions with low SDS density, the protein may not adopt the extended conformation thought to be typical of proteins in SDS. I don't know that anyone has really studied the problem in detail, though. One additional anomaly I have encountered: If you habitually overlay the running gel with isobutanol to produce a smooth interface, hydrophobic proteins may run abberantly. In the case I experienced, the protein would not even enter the running gel if there had been any overlayering, presumably due to traces of isobutanol remaining in the top of the gel. Michael Gribskov GRIBSKOV @ NCIFCRF.GOV