Sequencing GC-Rich DNA
marcusj at helix.mgh.harvard.edu
Tue Oct 29 03:30:57 EST 1996
<Pine.GSO.3.95-960729.961025104728.25836A-100000 at altair.dur.ac.uk>, Malia
Fullerton <S.M.Fullerton at durham.ac.uk> wrote:
: We are having considerable difficulty obtaining readable sequence from
: several different human nuclear genes. I suspect that GC-richness and/or
: secondary structure may be influencing sequencing efficiency, as primers
: placed at different locations give the same poor-quality sequence, and we
: are using (what has been with other loci) ample quantities of
: single-stranded template.
: What are the best (or at least most commonly used) methods for addressing
: this sort of sequencing problem? References appreciated.
Glad you asked. I've been having a lot of trouble with this lately, and
have tried several methods that have worked with varying degrees of
success. Here's a synopsis:
1. My original compressions and stops was a reult of using Sequenase 2.0
to sequence very GC-rich DNA.
2. Inosine reactions using the Sequenase kit didn't seem to work. In
fact, I've never been able to get them to work. :-/
3. I then tried the Fidelity kit (Oncor), which I saw advertised in
BioTechniques. Their claim is that, through the use of small DNA-binding
proteins in the reaction mix, their enzyme will read through structures
that typically cause stops and compressions. The Fidelity protocol is
longer and requires more steps than Sequenase, but I figured it was worth
a shot. Indeed, I got better resolution of many but not all of my stops
and compressions (70-80% of the problems were eliminated), and the
protocol was only mildly more tedious than Sequenase's. However, another
problem erupted: I found that band spacing was not as even as with
Sequenase, thus raising as many new questions as there were others
resolved. Other researchers in my lab have said this kit is fantastic,
and that they'll never go back to Sequenase; I thought it was simply a
little better than Sequenase. It's pretty certain that you'll get
improved sequence with this kit, though.
4. By chance I then stumbled across my real salvation: an article in
BioTechniques 17(2), pages 286-7. This article describes Klenow
co-sequencing without the extra step of adding the Klenow separately - you
just add the Klenow to the main reaction mix (along with Sequenase). This
method worked wonders for me; I got really beautiful sequence, with no
stops or compressions!
Briefly, here's how you do it -
Proceed as normal with the annealing step, then add 5.5 ul of the
following mix to the annealed template & primer:
(mix recipe for 3 reactions, with a little extra)
0.1M DTT 4.0 ul
undiluted label mix 1.6 ul
35S-dATP 2.0 ul
Klenow 1.2 ul
H2O 5.2 ul
Sequenase mix 8.0 ul
The Sequenase mix contains:
1.0 ul Sequenase
0.5 ul pyrophosphatase (included in kit)
6.5 ul Sequenase dilution buffer
As usual, let the reaction go for 5 minutes at RT, and then proceed with
the termination reactions at 37 degrees for 30 to 45 minutes (I know that
sounds like a long time, but believe me, it works!).
And yes, there is more glycerol than usual in your reactions due to the
Klenow and the pyrophosphatase, but don't worry, you probably won't have
to use glycerol-tolerant conditions. You _may_ need a glycerol-tolerant
gel for runs that are longer than 4 hours; I can certainly attest to the
fact that a four hour run looks just fine with a normal TBE gel.
Hope this helps! Let me know if you figure out anything else that works better.
More information about the Methods