I ordered recombinant hemagglutinin (HA0 ectodomain) with c-terminal His tags. However I'm getting a lot of streaking in my native gel using Tris-Glycine pH 8.6 and Tris-Acetate 3-8% NuPAGE gel. The recombinant protein was originally lyophilised in PBS pH 7.4, 5% trehalose and 5% mannitol. The pI for H1 is 6.8 while for H5 is 8.4. I know why H5 didn't migrate much due to its pI being close to the pH of the buffer but then the streaking is again present. I'm not sure how to stop non-covalent aggregation without disrupting any disulphide bonds. Maybe using small amounts of non-ionic detergent or low amounts of DTT or BME?
Attached figure - (Coomassie R-250) stained reference bands are shown). BSA is on the left. HA is on the right
Hi Robert. In my experience the most common causes of streaking in native PAGE are 1) aggregation 2) glycosylation and 3) high salt/buffer ion conc.. I don't see any staining in the bottom of your wells, so it is probably not aggregation. I'm guessing this is a bacterial recombinant, so there is no glycosylation.
Did you add diH20 to the lyophilized protein to bring it up to 1X PBS?
The problem with the recombinant proteins is that it can have heterogeneity of sialic acids in the glycans or carbohydrates of the protein, then the migration form a smear. This result depends on cellular system used to produce the protein. This glycosylation deffect also shows in mammalian cells (Heterogeneity of Recombinant Human Antithrombin III Expressed in Baby Hamster Kidney Cells, JBC 1993; Characterization of Recombinant Human Antithrombin III Synthesized in Chinese Hamster Ovary Cells, JBC 1989). Unfortunately, I think that you cannot perform native-PAGE, because it cannot resolve with neuraminidase digestion. It improve the migration, but is difficult to understand the results.
Hi Robert. I would recommend you to try using low concentrations of detergents as you point out. For example dodecylmaltoside or Triton-X100 at 0.1-0.2 % may help to avoid aggregation without affecting disulphide bonds. Good luck.
I agree with Patricio, try to play with mild detergents (assuming that you have your sample in 1% PBS). Besides DDM and Triton-X100, you can use digitonin, octylglucoside (40 mM) or saponin. In general, the working concentrations are 0.5%-1.5%, but you never know which one is ideal for your sample. I would make several dilutions and see which one is better.
I forgot to mention that 5ug of recombinant HA0 was deglycosylated O/N with 3U of PNGase-F prior to sample loading. I'm using recombinant HA (expressed in human cells) from Sino Bio which has it lyophilised from PBS. When it arrived, I resuspended it in 1/2 recommended vol (so probably 2X PBS).
In attempt to get rid of streaking, I did 2 treatments: desalt using centrifuge spin column and ~1.2mM of DTT to the sample buffer. The streaking is still present in both treatments. My last option now is to treat sample with non-ionic detergent. But since streaking is issue is whether the aggregation is occuring within the gel rather than in the sample buffer. Also I noted that my loading volumes are 26-30uL which may be a bit too high. Previous native gels in my lab were done successfully with no streaking using 20uL with same protocol so I'm increasingly suspicious of Sino Bio's quality control.
This native gel troubleshoot below is for H1 HA only. Note for the desalting lane, I took 2xamount of HA as I was advised by another member of my lab, to assume I would lose 50% protein from my spin column. It does look slightly better but maybe because I started off with more protein.
Have you run these samples on a regular SDS-PAGE to see if they are heterogenous, either from the supplier or after the PNGase treatment.
OK, I'm officially stumped, sorry.
The only thing I can think of is to use higher % gels. The 3-8% gel is augmenting all structural, charge, conformational, etc. differences. Also, BSA is probably not a good marker for these experiments since it is not glycosylated.
It turns out I may not be hitting the HA with enough PNGase-F. It appeared as a very faint band under bright light and not in the scanner image. Ignore the pellet data, that was something to see if the addition of PNGase-F buffer with the HA buffer may have caused precipitation of the protein. The S/N data shows that there might be a lower band just below the non-deglyc HA so I will use 3x as much.
Ok, I've added 6U instead of 2U of PNGaseF and it has succesfully deglycosylated by the looks of it. That is for the 1st gel scan.
I also tested protein aggregation in gel and found out that the streaking in the native gel is 2-component problem. De-salting removes the | | pattern in presence of 0.4% Triton-X (I tested no Triton and the | | pattern remains) but band streaking remains. I am now testing 6U PNGaseF effect on Native Gel alone and with Triton & De-salting
- Badges
- Science method