plasmid preps & inverted repeats
Scott Bruce Mulrooney
: Hi folks:
: I would like to get some suggestions on a particularly untractable
: sub-cloning project. I have a 1.5 kb insert in pBlueScriptII which is a
: perfect 750 bp inveretd repeat. I want to complete its sequence, which
: has proven quite a chore. It won't sequence at all using cycle
: sequencing and only works with vector primers using Sequenase. Internal
: primers will not work. I presume this is because of the secondary
: structure formed when denaturing the plasmid i.e. the inverted repeats on
: each strand snap back to form a huge hairpin and prevent primer binding
: and Taq pol procession. Making ssDNA has not been useful. I can readily
: subclone portions of each segment of the repeat and have got most of the
: clone sequenced doing this. However, about 50 bp at the i.r. junction
: eludes me. Restriction mapping indicates the junction is a perfect
: palindrome or contains only a very small unique sequence. It has been
: impossible to obtain sub-clone of this region. I have put this down to
: the potential for cruciform structure formation in the parent plasmid
: during the alkaline lysis step in the plasmid prep. This any enzymes
: cutting in this region will a single stranded molecule which is foled
: back on itself and thus has only one end available for cloning.
: What I need is a non-denaturing plasmid prep to avoid this problem. This
Ten or 15 years ago, I spent most of my time constructing cruciform DNA of
various lengths for use in experiments to measure the rate of branch
migration of Holliday junctions. We were working only with linear DNA, but
some of the things I found may be of a little use to you.
We found that the inverted repeats had what we called zero-time renaturation,
that is if you put some of these hairpins in boiling water and immediately
quenched on ice, they still renatured into the hairpin form - no single
stranded DNA was found.
I don't think that using a non-denaturing method of plasmid preparation (if
there was one) would help you here. In supercoiled circular DNA, the
supercoiling will force any inverted repeats to be extended in a
hairpin and not part of the circle. I do know that people have used plasmids
containing inverted repeats for studying resolution of Holliday junctions and
other things - so they must have sequenced the things. I'm not in the DNA
business these days, but if your interested I might be able to dig up some
references where people have worked with these hairpins in plasmids - but
I don't know if they will say anything about how they were sequenced.
The fact that you can use a restriction enzyme and cut off a smaller hairpin
piece might still work. My brain is a little fuzzy on this, but there may
be some cloning vectors that contain short inverted repeats in their
multiple cloning sites - M13mp7 I think is one. Also, you might try to
sequence linearized DNA - this may help to destabilize the hairpin.