Koch Lab:Protocols/Unzipping constructs/17mer/Adapter duplex

Overview
The purpose of the adapter duplex (aka "insert duplex") is:
 * Provide the final essential components for unzipping: an internal nick (missing covalent bond) and a biotin label.
 * Provide another sticky end for ligation of downstream unzipping segment

There are a variety of ways this can be achieved. The specific implementation here seemed to work well.

Legacy
We switched to this method after having some issues with a "fork unzipping construct". The method is adapted from Heslot, Essevaz-Roulet, Bockelmann 1997 method. One difference from that method is the biotin is internal to the duplex, as opposed to a dangling end of DNA. We chose to do this so the T4 DNA ligase would be "happy", as the DNA can form a nice hybrid and the biotin is far enough away to not affect ligase. We don't really know whether this would have been an issue, but the design ended up having a nice feature: unzipping commences with an initial moderate-force shearing event. This serves as a very nice landmark in the force-extension data.

Also, when using downstream unzipping segment to study site-specific DNA binding proteins, you should probably design the adapter insert to not have those binding sites.

Alternatives
In this protocol, we are using the adapter duplex as just a way of connecting the anchoring segment to the downstream unzipping segment, while providing a nick and a biotin label. So, we don't care much about the DNA content of the duplex. However, there are a lot of opportunities for putting the science directly into the adapter insert, and forgetting about the downstream segment altogether. One could do this with even an adapter hairpin (so that unzipping doesn't break, but results in a completely single-stranded reversible strand). In doing so, one would want to add length to the duplex (probably) and binding sites of interest. Constructing something like this would be a lot easier than the full 17-mer unzipping construct we're describing in this protocol. (But 17-mer repetitive DNA has a lot of advantages.)

Oligonucleotides
Pay special attention to presence or absence of 5' phosphate on these oligos


 * 17mer Adapter I -- overhang
 * 5’-P-GCTGTCTGAATTCTAATGTAGTATAGTAATCCGCTCATCG
 * Modifications: 5' phosphate
 * Overhangs (sticky ends) are in bold above
 * 17mer Adapter II -- biotin underhang
 * 5'-GAGCGGAT(6)ACTATACTACATTAGAATTCAGAC
 * (6) = biotin-dT modification
 * NO 5' phosphate

Steve recommends Alpha DNA for the oligos. See Koch_Lab:Protocols/Unzipping_constructs/17mer/Anchoring_segment.

Oligo resuspension buffer
Many different buffers will work. Try 0.1x TE buffer.

Oligo annealing buffer
The final annealing buffer should be approximately 10 mM Tris pH 8.0, 50 mM NaCl, 1 mM EDTA. You can either mix up a 2x or 5x or whatever of this buffer, or you can just add the 10x components individually to an annealing mix. The annealing buffer isn't too much of a big deal.

Resuspend Oligos
Resuspend oligos in oligo resuspension buffer (such as 0.1x TE buffer). Use an appropriate volume so that based on the certificate of analysis sent with oligos you will end up with about 100 micromolar of oligo.

Assess actual concentration using a spectrophotometer and look at the whole scan from UV through 500 or so nm. If there are impurities in the oligo preparation (as we have seen with some companies, and particularly dig-modified oligos, this may show up in the absorption scan). Calculate the actual concentration, and use this in subsequent calculations. This isn't an exact measure of concentration -- a better and easier method would be good, as we would like to have equimolar concentration of oligos in the annealing step.

Anneal duplexes
Should have about 100 micromolar starting oligos. Mix them to a final concentration of 10 micromolar (each) in annealing buffer. For example:

100 microliter total reaction

10 microliters 17mer Adapter I -- overhang 10 microliters 7mer Adapter II -- biotin underhang 10 microliters 100 mM Tris pH 8.0 10 microliters 500 mM NaCl 10 microliters 10 mM EDTA 50 microliters water

Mix this reaction in a PCR tube (doesn't need to be thin walled), and program thermal cycler to heat to 95C, and then cool down to room temperature as slowly as possible over an hour or two (this is probably overkill, but easy enough to wait). The cheaper way to do this is to boil about 250 ml of water, float the tube in it, and let it cool on the benchtop.

Assessing stoichiometry
You can try non-denaturing polyacrylamide gels to assess product formation, but it's not that easy. You may want to just hope for the best. (Final ligation product will let you know it worked on a regular agarose gel.)

Storage and handling
Store in 0.5 ml annealing tube. Freeze thaw cycles should not be an issue, but try not to bring too far above 4C when thawing.