detecting pcr-products by hplc

Peter Doris P.Doris at
Thu Mar 9 15:41:56 EST 1995

>I'm trying to detect pcr-products by a hplc-method.  I'm working with 
>fluorescein-marked primers and a fluorescence detector. My interests concern 
>the kind of hplc-columns and buffers to use with this tecnique.
>Has anyone made experiences with this subject?

>e-mail to bohne at

HPLC separation and detection of PCR products has taken a quantum leap
forward as a result of the work of Peter Oefner and colleagues who have
developed reversed phase/ion pair methods for the rapid and highly
efficient separation of double stranded DNA. Some of this work has been
described in BioTechniques 16:898 1994.

If you are interested in learning more about the methodology I think
you should contact Dr. Peter Oefner, Dept. of Biochemistry, Stanford
University (oefner at You may also find some of his
other publications on the topic via a literature search.

We have been working with one of Dr. Oefner's columns for several
months. Frankly, outside of sequencing gels, I dont there are very many
circumstances in my use where PAGE would be preferable to HPLC. System
requirements include a gradient HPLC system and a column heater (we run
at 50C). We are very easily able to detect PCR reaction products using
UV absorbance and we typically inject 10microliters of PCR reaction
product. Run times are very short. Ours are about 8 minutes but we have
stuck with an elution system which will resolve easily from 120 - 600+
bp, if we tailor the elution to reaction products in a narrower size
range the profile can be developed faster and run times of 4-5 minutes
are possible. Our reaction products are typically 100bp different in
size, but resolving 30 bp will be no problem.

The solvent system is a gradient of acetonitrile in triethylammonium
acetate. Elution times are very reliable and there is no
re-equilibration interval between runs (2 minutes maximum to switch
back to starting conditions at the end of the gradient). As you can
imagine, this makes a very efficient system and if you have an
autosampler you will be able to analyze a lot of samples very quickly
and with great precision. There is no additional processing of samples
from PCR reaction to HPLC analysis so there is a lot less work and less
opportunity for error because of reduced sample handling. An on-line
integration system will give you peak areas which can be used to
calculate gene expression level directly.

In our quantitative RT-PCR reaction titrations we have obtained
correlation coefficients and slope values both very close to
theoretical ideals. Much better than we have ever seen reported using
electrophoresis and densitometry or phophorimaging.

Dr. Oefner's column technology will soon be available commercially (I
think.... better ask him yourself). I have no commercial interest in
this column and my opinions reflect only the very satisfying work we
have accomplished with this technique.

Peter A. Doris
Associate Professor of Cell Biology and Biochemistry
Texas Tech University School of Medicine
P.Doris at
(806) 743 3046             

More information about the Methods mailing list