Preparation of peripheral blood mononuclear cells - How to have my cake and eat it? (Jan/18/2010 )

There are zillions of methods for the preparation of human PBMCs from whole blood and buffy coat but I have yet to come across a single method that allows me to also recover at least 95% of the undiluted plasma and doesn't use expensive magnetic beads.

Most of the methods start off: "dilute the blood 1:1 with PBS and layer on top of an equal volume of Lymphoprep/Histopaque 1077...", which means I can't use the plasma. Some say you can layer whole undiluted blood on Lymphoprep, but I want to avoid any contamination of the sample with Lymphoprep, having sensitive assays to do on the red cells and plasma that might be affected.

If I centrifuge the blood sample first (1200 x g for 15 min), the packed leukocyte layer (buffy coat) can be removed and resuspended in PBS-EDTA and layered onto Lymphoprep. However, after centrifugation, the cells form clumps instead of a nice homogeneous band at the density interface, and this is probably due to platelet activation during the initial centrifugation. I could add platelet activation inhibitors to the blood, but I prefer not to be adding things whose impact on the other cells might be undesirable.

A milder initial centrifugation (200 x g) is the established way of producing buffy coat that isn't activated, but then I lose a lot of plasma.

Surely there must be other people out there who are trying to prepare plasma and cells from limited samples and with similar constraints? Am I missing some simple method kept secret by hematologists, or is this really such a difficult problem to solve? If anyone has a simple (and preferably inexpensive) way of solving this problem, I would be very grateful.

The simplest way would be to ask for a separate blood tube just for plasma. A green top vaccutainer filled at the same time the blood draw occurs.

We get units of blood to separate out the PBMCs and still have left over serum for other things.

To just get the serum and cells separated, we just use a Ficoll gradient and centrifuge for 30 mins. The result isn't the cleanest prep of PBMCs, but it isn't all that bad.

After that, we use the expensive magnetic beads to ensure a clean population.

Not sure if that entirely answers your question, but hopefully it helps!

You should be starting with a spin of the whole blood without any ficol. The RBCs will be at the bottom, then the buffy coat where your PBMCs are, then plasma on top. Remove plasma with a pipette; this can be clotted with CaCl if you want serum. Skim off the buffy coat and do ficol on that.