RNA from T-cells

[Anton Scott Goustin]

vblasch at gwdg.de (Volker Blaschke) wrote:
>Dear fellow researchers,
>
>I need to make RNA from Buffy-coat (native) T-cells. However, with all 
the protocolls I've used so far (e.g. GuSCN/Ultracentrifugation, 
Chomczynski/Sacchi and Qiagen RNeasy kit), yield was incredibly low 
(around 5 ug total RNA from 20 million cells).
>Does anyone know if this yield is in fact what one gets from native 
T-cells or of another methods giving higher yields?
Qiagen states in their protocoll that their yield is around 5 ug total 
RNA from 1 million T-cells, but what they don't say (and can't say on 
the phone) is if that's native cells or stimulated cells or a cell line.
                                        ^^^^^^^^^^

Volker--
You hit on a main point: stimulated versus resting.  Resting T cells 
have only a thin layer of cytoplasm around their big nuclei.  The low 
yield of RNA is thus a natural consequence of the resting state.  I 
normally use direct capture of poly(A+) RNA as a first approach to any 
cell or tissue, a method first used by Manfred Schwab in 1982 as a 
postdoc in Mike Bishop's lab in San Francisco.  My experience here is 
dismal:  I got almost nothing!  It turns out that not only do resting T 
cells have little RNA, they have low ratios of poly(A+) RNA to total RNA 
and shorter poly(A) tails.  The solution is to use hellish numbers of 
cells...and, dare I say:  RT/PCR!  Northerns are an uphill battle.  Can 
you stimulate your T cells with PHA, cytokines?  That will help, if 
stimulated is OK.  But, you probably have the same problem as I do:  you 
want to show that the native cells also express the mRNA....

If you are working with humans, as I have, you can always scale up so 
several hundred mL of blood (~1 million T cells per mL).  Try 
leukophoresis units.

That's where my experience ends!