Bradford Query - (Feb/03/2010 )

Hi,

I just have a query about the bradford assay. I have done a couple of WB always doing a bradford on my samples beforehand and equalising the samples before loading them onto the gel. However, one of my supervisors said yesterday that he has never used a bradford assay, instead has just pelleted the cells resuspended them in RIPA buffer then loaded about 20ul onto the gel. He has stained for b-actin to check for equal loading and has never had variations between the samples.

Although, I understand his reasoning, if you have plated down the same number of cells in each well, there should be the same level of protein in each sample. Obviously though if you are treating the cells with a compund that is killing the cells, there won't be the same number of cells in each group. Basically my question is to bradford or not to bradford? What are peoples opinions on this?

Thanks,

smr86

Hello

It's easier not to Bradford...

I used to do Bradford assays during my PhD until a post-doc told me that he just uses 20 million cells/lane!

Are you able to count your cells which have been treated with the compound? If so, why don't you just do that and see what happens?

Clare

Is it really shorter to count 20 million cells/lane?! Not in my opinion. A bradford assay do not take more than 30 minutes, and you have a really precise protein concentration. Using constant volume of lysate really do not guarantee that you'll have constant amount of proteins loaded onto your gel.

Your thought is right, smr86. Different treatments could have different results on cell survival/concentation. Thus, you should always quantify your assays. I use adipocytes that need to differentiate for 6 days, and I take samples at day 0, 2, 4 and 6. Protein concentrations vary from 1 µg/µl to 4.5 µg/µl.. so not to Bradford is a really bad idea..

On the other hand, if you see that your samples vary from let's say 1,3 to 1,5 µg/µl, and that this has been true for the last 6 months.. you could always give up the Bradford...

it sounds like what your supervisor does is "normalize" the results based on the actin concentration, not that (s)he is loading the same amount of protein for each sample.

I use cell counts per lane as I am frequently looking at viral effects that cause a lot of cell debris which cause erroneous values for protein content. That is kind of a special case though.

Having said that I do usually count my cells at the end of an experimental timepoint, as I like to know the live/dead proportion when I am working with proteins that affect cell viability.

IMO yes - how long does it take to count cells? Just a couple of mins.
C

madrius1 on Feb 3 2010, 04:27 PM said: