if you use PBMCs from patient blood samples, you should recover a sufficient amount of PBMCs from 10-20ml whole blood (EDTA, citrate or Heparin anticoagulated; PBMC isolation by gradient centrifugation; the rather high sample volume I recommend is thought as a safety margin, since isolation procedures, lysis, protein determination and and western blot techniques might differ. Starting with 20ml you should come up with a decent protein amount to have some tries. If you have implemented the procedure you will likely be able to reduce the amount of blood sample significantly). Do you plan on doing Western Blot from total PBMCs or do you target a specific subset ? (This might necessitate a higher sample volume). What is the source of your PBMCs? Were they frozen or is it fresh blood?
If availability of donor blood is no issue Andreas gave you a very good estimate. Since, however, most vacutainer-like tubes have a volume of ~ 5 mL it may be worth to know that this will still give you a yield of ~ 20 million PBMCs.
In our experience this is more than sufficient to do several Western blots, provided you do not need a specific cell type only, as pointed out above.
We plan on doing western on total PBMCs. The issue we may have based on your answers is we are fairly limited in how many cells we will get. We will only have 1 or 1.5 vacutainer tubes(8ml per tube) devoted to the westerns. So, we might only have enough protein to run a few westerns. We are hoping to look for p38 phosphorylation induced by LPS, in the presence and absence of an inhibitor. So just those conditions would require a lot of cells. There are other things we wanted to try as well, but that may not be possible because the amount of blood we will get is limited. I guess we will have to dial back the amount of experiments we will be able to do, unfortunately...
The advice of Ernst is very helpful. Another idea (what we did when we started to work with PBMCs): Give it a try and work with your own (or a fearless colleague's PBMCs) to get a feeling what is feasible and what should be adapted