How can we directly sequence PCR-amplified DNA from agarose.

Hank Seifert h-seifert at
Fri Apr 29 09:38:34 EST 1994

In article <Pine.3.87.9404282351.D13603-0100000 at cmu.chiangmai> asmsi002 at CMU.CHIANGMAI.AC.TH (Wasun Chantratita) writes:
>On 27 Apr 1994, David Schuster wrote:
>> Dear Wansun,
>> Why aren't you using Gibco BRL's GlassMAX cartridges?  I've talked to a number a
>> customers that were successful with direct sequencing of 5' RACE products.
>> Bands were excised from agarose gels, purified over GlassMAX, and then sequenced
>> by cycle sequencing.  Its very important to use an internal primer for the
>> sequencing reaction.  Primers used for PCR amplification don't seem to work.
>Hi Dave
>Long time, no see.
>  I glad to meet you again though just from text mode thru internet
>world. I also heard people saying that Its a must to use an internal
>primer for the sequencing reaction.  Primers used for PCR amplification
>don't seem to work. Do you know the reason, why?
>	According to purification of PCR-amplified DNAs using
>GlassMax, would it be possible to use with PCR-amplified DNAs which their
>sizes less than to 200 bp?
>	Yesterday I failed to obtain any PCR-amplified DNA (540 bp) from low
>melt agarose using Glassmax. Does glassmax will work only with ordinary
>       Would it be possible to utilize a PCR-amplified DNA obtained from
>amplification reaction mixture using Glassmax for cycle sequencing?
>Best regards,
Dear Wasun:
We have used amplification primers sucessfully many times 
although internal primers usually work better (i.e., cleaner 
sequence).  To use the same primers you must get rid of the 
unlabeled PCR primers.  First make sure you are using a minimal 
amount of primers in the PCR to produce a product.  Although 
many protocols recommend 50 pm, we often find that 10-20 pm 
works well.  We purify our product by spin dialysis through 
Sepherose CL6B which removes the primers, enzyme etc, but any 
purificcation scheme should work.  We then do cycle sequencing 
with end labeled primers. This works almost every time after 
each invididual worker has done it three times.  I don't really 
understand this learning curve, but it is very consistent.
I hope this helps.
Hank Seifert
Northwestern University

More information about the Methods mailing list