I've got a His-tagged protein in 6M urea, 500 mM imidazole buffer that needs to be quantified before dialysis to ensure there's enough protein worth dialysing. I ran out of my elution buffer which should be used for blanking. I made a fresh buffer with the same contents, it started giving me false results. I could interpret that problem is with the buffer (probably because of change in chemical batch) from the readings I got in BCA and Bradford assays. Hence, I ran a gel to get an idea of presence of protein. Are there ways to estimate protein concentration quantitatively from the gel?
Can concentration of a protein be determined from a gel quantitatively (rough estimation)?
1$\begingroup$ I suppose you could get an estimate if you made a calibration curve with known concentrations of protein, then try to match the intensity of your unknown with the calibration curve. Probably wouldn't be very accurate. I recommend fixing your buffer and using the BCA assay. $\endgroup$– user137Jan 12, 2015 at 6:27
1$\begingroup$ Yup, I am doing that with a BSA standards. BCA assay gave me negative values for all the dilutions. I've tried fixing the buffer, it couldn't help me in any way. Thanks! $\endgroup$– piscean_1993Jan 12, 2015 at 6:32
Yes, you can use SDS-PAGE as a semiquantitative estimate of protein concentration. You need to create a standard curve with a protein of known concentration to compare against. Quantification is done by densitometry. It's a quick and easy process, but keep in mind some limitations:
- Band intensity depends not only on the amount of protein but also on the amino acid composition of the proteins and the dye used
- The unknown protein and standard should be run on the same gel to remove variations in gels, electrophoresis, staining, destaining and imaging
- Densitometry can be quite qualitative depending on how it's done
I'm not sure if you're looking for a protocol, but it's relatively straight forward. I usually load 2, 4, 6, 8 and 10 uL of 100 ng/uL BSA as a standard and 2, 4 and 6 uL of the protein of interest (based on past experience with my proteins and their purification). The gel is stained with Coomassie and the band intensity estimated with a program called ImageJ. A standard curve is created by plotting intensity as a function of mass, from which the concentration of the unknown can be found.
$\begingroup$ Wow! This is new to me. Thanks a lot!! Does a scanned image of gel work on ImageJ? $\endgroup$ Jan 12, 2015 at 8:12
2$\begingroup$ @piscean_1993 Yes, but you'll introduce artifacts into the image during printing and scanning. It's probably best to keep it digital if possible. $\endgroup$ Jan 12, 2015 at 8:14