PBMC from Cyno blood--Summary

From: Mario Roederer (roederer@drmr.com)
Date: Tue Oct 09 2001 - 14:05:41 EST


I want to thank those people who responded to my query about getting
PBMC from cyno blood.  There were a number of different suggested
methodologies (and lots of commiseration about the difficulties!).
We are going to try a few of the methods suggested below and compare
them; anyone who is interested in results should email me in no less
than about 6-8 weeks.

The responses are listed below, with editing only to remove the names
of the innocent.

(PS, I wish to reiterate the suggestion that all procedures involving
centrifugation be provided in terms of force (number of g's); RPM is
only relevant if you are using exactly the same bucket and
centrifuge--and so if you don't provide that information, becomes
almost meaningless!)

mr



This is what I have learned from others: They dilute the ficoll gradient
with RPMI (80 - 90 % Ficoll and 20 - 10% RPMI). Sometimes, it even seams to
vary between animals. If I would have to do that I would make a few diluted
ficoll gradients with a density between 1.077 and 1.070 (density of ficoll
is 1.077) and would try them \ compare them.

We have been isolating cyno PBMC  for many years using Lymphoprep
gradient from Nycomed. It works quite well. First we dilute the
heparinized blood 1:3 with media or PBS. We then layer it over the
Lymphoprep at a ratio of 1 part Lymphoprep to 2 parts diluted blood.
Spin for 30 minutes, room temp, no brake at a speed of about 1800
RPM. Harvest the interface, which is sometimes a little more diffuse
than you would see in a human prep but easily retrievable. We wash
twice and lyse red blood cells if necessary and the cells are ready
to use. We routinely get about 10 million cells from 5 ml of blood.
They are very viable and are essentially all mononuclear cells. We
spin one of the washes at about 1200 RPM. This helps to get rid of
most of the platelets that sneak in when we harvest our interface.
Some animals will have more red cell contamination of the interface
than others. It seem that the more you bleed them, the more red cell
contamination you see. Naive animals usually have very clean
interfaces. It is not really a problem for us because we either just
lyse the RBC using standard non-fixative red lysis buffer or we leave
them. Often times we are freezing the cells down for later use or
using them in culture conditions were a  few RBC are ok. This
Lymphoprep method is easy and not very time consuming. We can harvest
PBMC from a dozen or more monkeys in a morning and that includes
bleeding them! Hope this helps.

The only thing I've found that works consistently for cyno blood is a
53% Percoll gradient.  I have had bad results with Ficoll gradients
as well as the BD CPT tubes.

Having worked with baboon, rhesus, and cynomolgous blood I have to
say that for some reason cyno blood alone out of the three is very
difficult to get good pbmc out of.  For some reasons cynos just don't
work well with ficoll separation.  we've tried altering ficoll spin
time and speed to no avail.  Nothing that we tried worked
consistently, most of the times the layers are just plain nasty.  But
for the record we dilute 15 mo of blood in 15 ml of RPMI 1640, then
overlay over 15ml of room temp Ficoll-paque (Pharmacia).  Centrifuge
for 30 minutes at 1450, room temp and brake off.  Pull of PBMC layer
(usually down lower than you would expect) into a 50ml centrifuge
tube and fill tube with RPPMI 1640.  Centrifuge at 1500 for 10
minutes, decant supernatant and resuspent pellet in 10 ml RPMI.  Due
to the usually poor preps we then lyse the red blood cells in the
pellet by addign 20 or so ml of ammonium chloride (stem cell
technologies), incubate on ice for 10 minutes, fill tube with RPMi
and centrifuge again.  Decant supernatant and resuspend clean pellet
and perform cell count.  One of my colleagues claimed that she got
better results by spinning the ficoll at 2000 instead of 1450, but
again this was not consistent from one cyno to another.

We are using the standard ficoll procedure but first we dilute the
ficoll to 90% with water. Using this technique we get good separation
and good yields.

I'm afraid there may not be a good solution for you.  During my work,
I had dozens of cyno
samples every few days to monitor the immune status of the transplant
recipients.  I guess my question to you would be what are you doing with the
samples?  I found that a simple RBC lysis (ammonium chloride) was sufficient
for the samples that only needed routine flow analysis.  But for everything
else, sorting antigen reactive B cells, culture assays and the such, I HAD to
use a gradient (and buy another centrifufge!).  In the beginning I tried ALL
of the separation medias out there.  LSM produced by Mediatech
  http://www.cellgro.com/catalog2/results2.cfm?ID=75 ) worked significantly
better than ANY other product.  Trust me on this one, for cyno blood its the
best.

We still use Ficol gradients to isolate PBMC from ACD blood.  We routinely
recover 2 million cells per mL blood.  We have used this procedure monthly for
the past 7 years with cynomolgus macaques.  It is time consuming and
not without
problems (resulting in loss of sample).

We routinely isolate PBMC's from cyno's as follows:
Dilute heparinized blood approx 1:4 with PBS
Layer onto 90% Ficoll-Paque (diluted with PBS)
Centrifuge 30minutes at 1400rpm (~400g) without the brake.
>From this point, it's fairly routine and the same as human PBMC's.
We've also had some luck using the CPT vacutainer tubes from BD.  Fresh
blood works best with this method.

Monkey red blood cells are not easily eliminated when using regular
Ficoll-Paque however,
this problem is circumvented by adding an anti-red blood cell MAb to the blood
before centrifugation on Ficoll.  The PBMC were usually cleared from RBC
although not all the times.  The buffy coat from 5 mL of blood is
diluted to a final volume of 2 mL with
RPMI.  An aliquot of 50 µL of Red-Out (Robbins Scientific Corp.,
Sunnyvale, CA),
a murine monoclonal antibody to human red blood cells, is added and the
suspension is incubated for 5 min.  The suspension is diluted to 8 mL
final with
RPMI and overlayed on 5 mL of Ficoll-Paque (Pharmacia).  The
preparation is spun
at 2200 rpm (Sorvall GLC-2B, Dupont)  for 30 min. without brake.  The
PBMC layer
is harvested and washed twice in 10 mL of RPMI at 300g for 10 min.



This archive was generated by hypermail 2b29 : Sun Jan 05 2003 - 19:01:33 EST