I fractionate my cells into cytoplasmic and nuclear and consistently get a higher band in the nuclear fraction as compared to cytoplasmic when probed with my antibody of interest. How do I identify if this is a post translational modification? and if yes, what can I do to verify?
How big is the size difference between the bands? This is important for judging the nature of the modification (if it is really a post-translational modification).
Some modifications can be identified by treating your lysate/sample with enzymes that remove the modification.
For example:
- alkaline phosphatase to test for phosphorylation,
- PNGase F to test for N-glycosylation
If the size difference is lost when you analyse the treated vs. untreated samples you can tell what kind of modification it is that you have.
I hope this helps you.
Best regards.
How big is the size difference between the bands? This is important for judging the nature of the modification (if it is really a post-translational modification).
Some modifications can be identified by treating your lysate/sample with enzymes that remove the modification.
For example:
- alkaline phosphatase to test for phosphorylation,
- PNGase F to test for N-glycosylation
If the size difference is lost when you analyse the treated vs. untreated samples you can tell what kind of modification it is that you have.
I hope this helps you.
Best regards.
First of what is the Mw difference? You could stain with ConA to see if it is glycosylated. If it is PTM you could tell just by Mw difference (Usually PTM adds very little change to Mw). Perhaps you should run low% gel for separation. Or run IEF. There are also antibodies to phospho-ser, tyr, thr.
The classical way to test if an additional band (higher or lower molecular weight than expected) is caused by a post-translational modification is to carry out a pulse-chase analysis (using 35S methionine to label newly synthesized proteins, 1000-fold excess cold methionine for the chase) and detect proteins after immunoprecipitation with protein-A sepharose, washing, running SDS page and exposing on film). If the unexpected band is derived from the band at the expected size, it will be invisible during a short pulse and become noticeable after a longer pulse. During the chase, the unexpected band should become more intense at the expense of the expected band which should get weaker until it disappears. This shows that the unexpected band (post-translational processing product) is actually derived from the expected band (nascent precursor). This technique always works even when you don't know what the modification is. You can do it using entire cell extracts (including the nuclei), so the only variable in the samples is the time of the pulse-labelling, and the time-intervals for the chase. Once you have established that your new band is derived from the expected band as a result of a biological process, you can work out what it is, if it happens in the nucleus, during transport to the nucleus etc...
Run another gel and cut the bands out of the gel. Elute the proteins out of the gel and trypsinized for proteomics.
I concur with Ru-Jeng. First, identify if your bands contain the same peptides, i.e. treat with a protease and compare the fragments. Nowadays this is done by trypsinization followed by mass spectrometric analysis of the fragments. If you don't have access to MS, there is an old technique based on limited proteolysis and analysis by gel electrophoresis that may help; see Cleveland et al., 1977. J. Biol. Chem. 252, 1102-1106.
Make sure that the top band is not due to dimerization, which may occur through intermolecular disulfide formation if your sample buffer is too old (make fresh with good mercaptoethanol). As to glycosylation, there are various kits for assaying the presence of a carbohydrate moiety; alternatively, the PAS (periodic acid Schiff) method works if you have enough material. Further characterization becomes a job for carbohydrate chemists using various tools : lectins, MS etc.
The first step would probably be to mix the two samples together and run another gel. If you still see 2 bands there might be something. Often the different buffers you use to isolate nuclear and cytoplasmic fractions will affect the migration of the proteins. Mixing the samples together will force you to put them in precisely the same buffer. Alternatively, you can just tell us if you see 2 bands when you analyze a whole cell extract.
As others indicated, you can do phosphatase treatment to see if the two bands become one to assess if the modification is phosphorylation. Generally Ser-Pro and Thr-Pro sites cause SDS-PAGE mobility shifts. Other modifications like deamidation can be assessed by leaving samples in high pH buffers for a number of hours. Some proteins have multiple mRNA splice variants and longer forms can be nuclear.
When I run the whole cell extract I still see two bands. None of the splice variants reported have that high a molecular weight (my nuclear band is about 10kDa higher than the cytoplasmic band). I am wondering if it could be sumoylation as there are reports that sumoylation drives some proteins into nucleus and adds approx 8kDa to a protein? in that case, can I check it using immunoprecipitation then western blotting with Sumo antibody?
- Badges
- Science method