supercoiled DNA migration in agrose [TBE vs. TAE or TPE gels]

[afc at gnv.ifas.ufl.edu]

In article <1hn8b6INN5ee at usenet.INS.CWRU.Edu>, djt2 at po.CWRU.Edu (Dennis J. Templeton) writes:
> 
> In a previous article, vjongene at isrec-sun1.unil.ch (Victor Jongeneel) says:
> 
>>chai_z at wehi.edu.au wrote:
>>
>>: Hi,
>>
>>: 	Several batches of M13 RF DNA of mine have been cut by EcoRI or SmaI,
>>: etc. and then electrophoresized in agrose gel. Uncut DNA (supercoiled) of some
>>: batches migrated faster than the cut one (linear), while the others slower.
>>: Most likely, some batches of my DNA were nicked (open circular). Could any one
>>: tell me the correct agrose pattern of supercoiled, open circular and linear DNA?
>>: or tell me where to refer to?
>>
>>: Thanks in advance.
>>
>>: Zhonglin Chai 
>>
>>
>>In most agarose gels (but _not_ always), supercoiled circular DNA
>>migrates fastest, followed by linear and nicked circles.  Linear does
>>not always separate from supercoiled, nicked circles usually do.  This
>>does of course not exclude the presence of miniphage (see other
>>response in thread).  The best way to find out is to start
>>with freshly purified (CsCl/EtBr) supercoiled DNA and treat it so as
>>to make linears and nicked circles.
>>
>>Cheers,
>>
> 
> To clarify Victors remark, the buffer used in Agarose gels affects the 
> migration of superhelical plasmid DNA, and the other forms to a lesser 
> extent. 
> 
> In TBE gels, form II (open circles) migrate slowest, and form I (CCC or
> superhelical or supercoiled) migrates slightly faster. Form III migrates
> fastest of all in TBE (that's full length linear DNA)
> 
> In tris acetate, or in the Tris phosphate buffer we prefer, form II
> migrates slowest, with form III running just slightly ahead, and form I
> (CCC) running far ahead of that.  The exact migration varies depending on
> the size of the plasmid and the agarose concentration; in our 0.7% gels in
> TPE, a 3 kb plasmid (supercoil, CCC) migrates near the 1.6 kb band from the
> 1 kb ladder.
> 
> We have never compared the migration in TBE and TPE directly, as it's hard
> to run adjacent lanes of a gel in different buffers!
> 
> An added note, I would recommend saving your TBE for the acrylamide gels;
> TPE can be made up at 50x and allows your gel pieces to dissolve in NaI for
> the glass powder elution protocol.
> 
> good luck
> 
> dennis

To clarify the clarification:

The relative mobilities of the three different forms is affected by just
about everything: buffer, agarose concentration, voltage, and probably
moon phase.  In my hands, running TBE gels a la Maniatis, the following
generalities hold.

Open circles run slowest.  A 15 kb mitochondrial molecule runs behind even
uncut genomic DNA.

Supercoiled plasmids run at about the same mobility as linear DNA of 1/2
the length.  I assume that this is because they are, in fact, linear 
molecules of 1/2 the length of a linearized plasmid.  This is highly 
dependent on the concentration of ethidium bromide in the gel.

There are lots of papers that discuss this issue rigorously.  I have one
by Beverly (Nucleic Acids Res 16:925-939, 1988) that talks about pulse
field gels, but also references the earlier literature.

Andrew Cockburn
USDA