Protocol Online logo
Top : Forum Archives: : SDS-PAGE and Western Blotting

Experimental design question for Western blot - (Oct/06/2008 )

Hi folks,
I'm planning to do so much western blotting for HO-1 (hemeoxygenase-1) that inter-gel and inter-experimental comparisons are unavoidable.
So my question is: how best should I handle it in order to increase comparability of multiple experiments?
Unlike ELISA in which a standard curve can care for everything, it seems that merely an actin loading control cannot solve every potential problem. For example, I even suspect that calculated fold change could vary if you develop films with the same gel for different time durations.
Thanks, Hanhan


The short answer is... you can't do quantitative comparisons. The best you can do is subjective estimations.

The long answer is... you would need to standardise everything, that means absolutely everything. For example for a gel you will need to pour exactly the same volume gel, with the same volume stacker each time, then run them in fresh batches of standardised buffer each time for the same duration at the same voltage/amperage each time, then transfer onto the membrane in fresh buffer using the same voltage, temperature, and time. Fresh block and primary and secondary each time (preferably from frozen aliquots of antibody, as it degrades in the fridge)...

Can you see where I am going with this?

I guess that it is maybe possible, but it would need to be validated so as to show that inter-gel comparison is valid.


Yeah hanhan, I am with bob1, it's practically not possible to get 100% comparability between different gels. Do the best you can with actin or maybe even a second control, but WB will always stay a semi-quantitative method. I have done this too and done densitometric measurements. You can see in these data also great variation. Best is, you keep most parameters in your gels and blots as stable as possible and get a certain routine in handling things. Prepare and run as much samples as possible at one time, otherwise time-dependant variation will add also upon your other variations. And then hope the best smile.gif That is bio-science wink.gif not as stable as math or physics


Thanks, guys!
So, this is a disadvantage of western compared to ELISA. Maybe I should consider ELISA, which seems quite expensive for HO-1.
Yeah, I face a challenge of doing math for biology biggrin.gif . I'm a toxicologist and I have to compare toxicity of different toxicants in order to rank them based on dose-response curves with certain biological endpoints as response metrics.