Sequencing mRNA from variable regions

Robert Horton horton at
Thu Jan 11 15:59:21 EST 1996

Peggy L. Bishop (pbishop at wrote:

: I am attempting to sequence the mRNA of the T cell receptor 
: variable region.  The samples are isolated from PBMC 
: either from patients or normals.  In brief, my method is to 
: make the cDNA, perform RT_PCR using a 32-P end-labeled 
: primer.  I can visualize the PCR products on a sequencing 
: gel.  Each primer pair yields up to 14 differently sized 
: PCR products.  Normals pretty much universally lead to 
: many bands for a given V primer that appear Gaussain in 
: distribution.  Patients diverge from that Gaussain pattern. 

These ladders are due to N-region diversity among rearranged TCR
sequences. If you measure carefully, you'll find they are 3bp apart,
reflecting selection for the right reading frame. If a patient gives a
single band, you know that the TCR usage is not as diverse as normal,
though it could still represent multiple different sequences of the
same size. But maybe you can make a statistical argument based solely
on the number of bands? You could even come up with a P-value testing
the hypothesis that this difference between normals and controls is

:  In quite a number of the patients, there are V primers 
: that yield only one or two bands.  Sometimes these single 
: or double bands are accompanied by a faint background 
: ladder. In the samples that yield one band with a 
: background ladder, it seems that the prominent band is 50% 
: or more of the signal strength.  This is an estimation done 
: by eye of the autorad.

: We want to find out if the V primers that yield single 
: bands are comprised of a unique DNA (RNA) sequence.[...]

Actually, any of the conventional methods to determine heterogeneity
("find mutations") among PCR products might work, like denaturing
gradient gel electrophoresis, etc. 

But do you really care if there are three different V-beta
14-containing sequences vs. one, as long as you know its not 35, or
whatever the normal control has? 


--- Robert M. Horton                        Have a :) day

More information about the Methods mailing list