Ponceau S and Coomasie do not correlate to b actin signal
Posted 15 May 2010 - 06:06 AM
I run lately two western blots from some in vivo saples (retina) and I came accross an odd thing.
The samples were lysed in the usual buffer I use for in vitro purposes and of course for my protein of interest and then the samples were sonicated as usual. The supernatant was kept and the protein assay was performed as usual. The triplicates used for the protein assay from each sample give more or less the same values in the spectrofotometer througout the same sample which tells me that my pipetting error is very small. After running the gel I stained the blot with Ponso as usual and the transfer was fine in all lanes. The loading was also equal based on the coomasie stain of the gel after transfer.
Then I did the primary and the secondary Ab and then developed the blot.
The odd thing is that the b actin signal or b tubulin signal (I have done both) does not correlate at all with the pattern of the Ponso. And it is not not just 20% off but a lot. I have for instance in one lane a very robust b actin signal and in the next lane I have a very faint b actin signal, as if there is no b actin over there. This pattern goes on thoughtout the blot having very robust and very faint signal next to each other.
Now, when I am handling in vitro samples everything is fine, Ponso correlates perfectly to the b actin signal or b tubulin and my loading is equal.
I have not changed anything in the way I prepare or perform my western blot and I have no idea why this is happening. As if the b actin is not coming from the same blot... I suspected at first that my samples were not homogeneous, but I can hardly believe that since both Ponso and Coomasie give a very good pattern of loading and transfering.
Do you have any idea why this is happening? Has anyone out there encountered this before?
Thank you all and have a nice day.
Posted 16 May 2010 - 04:38 PM
Posted 17 May 2010 - 05:11 AM
No actually, everything was done under the same conditions. I collect the retina as a whole and I lyse it. I am not collecting different parts of it.
Posted 17 May 2010 - 05:34 PM
Posted 17 May 2010 - 08:09 PM
Certainly, I upload the Ponso stain of the blot and the b tubulin film. Both go from left to right as you see them. I hope this helps.
Posted 18 May 2010 - 04:16 PM
Posted 19 May 2010 - 05:32 AM
Thank you very much for your answer which raises a further question from my side.
Whenever I am running protein from in vitro samples I never have this problem. The Ponso reflects the b actin signal and my loading is equal based on the b actin or b tubulin signal.
Do you think that this is a loading problem and only? In this case what should I do in order to sovle this? My protein assay is the same, so I do not know where to start now in order to fix this.
Posted 24 May 2010 - 05:22 PM
Posted 25 May 2010 - 11:54 AM
so what do you suggest? I have protease inhibitors in my lysis buffer. How long should I let the retinas sit in the lysis buffer before I sonicate them?
My assumption was that my original samples are not homogeous for some reason...
Posted 07 June 2010 - 01:15 PM
Posted 08 June 2010 - 05:33 AM
Thank you for your answer.
I have been asking around a lot lately and it seems that I am not the only one who has these issues. Maybe not to that extent but alla the fellows that I have asked and they are dealing with tissue samples seem to have this kind of problem.
Now, the odd thing is that I am handling all my samples (retinas) in the same manner and while I have huge discrepancied in the b actin signal, my protein of interest is not beeing affected that much.
Another issue is the fact that each lane is the sample from a retina from a different animal, so there is no way that the relative amount of b actin is the same between different animals. That would have been the case if I was supposed to take the contralateral eye and run a gel.
But even in that case I would have trouble between that different animal groups.
One thing I could do is to run a gel and see the differences and then adjust my loading according to the b actin signal i.e. load from each sample more or less according to the signal I get. I have already tried that in some older experiements and it gave me some results. Still I am not quite sure if this is the right way to go, since I will be also manipulating my samples according to the b actin.