Cell Signaling tends to be the quality leader, these seem to be their preferred choices:

https://www.cellsignal.com/products/4276.html

https://www.cellsignal.com/products/9197.html

It might make sense, however, to use a rabbit antibody for phospho and then a mouse antibody with total, so that you can kill the anti-rabbit HRP enzyme with sodium azide and just reprobe with the mouse antibody, in which case maybe this total CREB antibody would be good:

https://www.cellsignal.com/products/9104.html

The antibodies above are affordable if you want to perform the experiment a couple of times, but other companies sell antibodies at a much lower price per microgram if you want to do more of the QC yourself and possibly have a lifetime supply of antibody:

http://www.genetex.com/CREB-antibody-GTX112846.html

http://www.genetex.com/CREB-phospho-Ser133-antibody-GTX82716.html

http://www.enogene.com/search.php?searchsubmit=club&searchtxt=ser133&x=0&y=0

thank you a lot Mark. But do you have any personal experience with these particular antibodies? I have to agree that Cell Signaling is one of the best antibody producers, however, some phospho antibodies might be tricky. some of the mouse Cell Signaling antibodies against signaling transduction proteins do not give sufficient signal even after longer exposition and higher protein concentration per lane particularly when reprobing the membrane.

I haven't used any CREB antibodies for a long time and the lot #s have changed for every antibody I have used. Lot-to-lot variability is something the people focused on reproducibility tend to ignore.

thats very true. Lot to lot differences are the reason of a poor reproducibility.