IGEM:Harvard/2006/DNA nanostructures/Notebook/2006-8-23

Verify folding of design 5 and 6
Ran gel to verify folding from earlier this week. Also comparing new scaffold with old scaffold.



PEG fractionation (from 8-20)
Also including 8% and 10% PEG fractionation for c6.0.A, c6.0.B, and c6.0.C

Ran all three gels 2 hours, 15 minutes at 60 volts. Ran 10 of everything.

Results
So... it seems something strange is going on with the nanostructures. Not only are the design 5 ones very slow to move, it seems that the scaffold is not running consistantly. On gel one, it runs much slower than on gel three. Also on gels one and two a lot of the nanostructures seem to be getting stuck in the wells. Gel 3, although a poorer image quality, seems to be the saving grace - the nanostructures seem to be moving well, as does the scaffold and there are no double bands. This, as opposed to gel one and most of gel two is design 6. In all the gels, the peg fractionation seems to be working well, which is promising. Therefore, I think it's back to folding. Hopefully we can get a better folded batch and compare against these results to determine if the problems are just a result of this one batch or to see if there's some larger folding problem with design 5.

Goals

 * To see that binding with free streptavidin allows less binding with beads afterward (and thus brighter bands in the supernatants that are run on the gel) for outside biotinylated barrels than with bead-treatment for these barrels alone.
 * To see that outside biotinylation allows much higher (or any, perhaps) binding to streptavidin than inside biotinylation.
 * To see that streptavidin actually binds biotinylated oligos.

Expected Results
(Note: In full disclosure, this was written after seeing the gel, but these are the results that would have been expected under the depletion hypothesis. Differences from the gel in actuality are colored red .)
 * Beaded
 * Bright Bands
 * A (unbiotinylated barrel)
 * 9b (biotinylated oligos), much lower down
 * Fb (inside biotinylated barrel)
 * Dim Bands/No Bands
 * Eb (outside biotinylated barrel)
 * Free-Streptavidin
 * Bright Bands
 * All
 * 9b (biotinylated oligos) will have a band farther up, due to binding of streptavidin, which is much bulkier than the oligos are
 * Free, then Beaded
 * Bright Bands
 * All (same as with Free-Streptavidin)
 * Eb band here must be brighter than that in the Beaded Eb lane

Plans, Possible Issues, and Solutions

 * Too much DNA
 * Even if a reasonable amount (say 70% of the biotinylated sites) are binding strep, there wouldn't be a noticeable difference between the bead-treated boxes and non-bead-treated boxes.
 * Lower amount of DNA used - half a reaction? (10uL worked in c3.2, 7.11.06)
 * But the bands aren't that bright as it is, despite use of 5uL of EtBr - possibly due to DNA leeching out during the long (>2hour) electrophoresis? Is that likely at all?  But at the same time, my gels have been coming out fainter and fainter as I run longer each time, and yesterday it wasn't even left in buffer for any amount of time at all.
 * Run at higher voltage (100V), shorter time (1.5hr) - the difficulty in differentiating the folded from the scaffold might be due recently to the switchover to 0.5xTBE. Take a risk with the higher voltage - check on the gel box every 20 minutes to make sure no boiling is occuring, and scrape off the electrode wire each time.
 * Too little free streptavidin
 * Overload with free streptavidin - add from the original stock tube, which is likely safer (in more accurate-for-streptavidin-efficacy) buffer than the dilutions that have been made. This should push the equilibrium towards more complete binding of sites.
 * Biotinylated stuff not binding
 * Check by running a control of biotinylated oligos (untreated, beaded, free streped, free then beaded) - there are large amounts of the original tube oligos, so run 40pmol of biotinylated sites (10uM biotinylated sites in the pre-working stock = 10pmoles/uL), or 5uL (to account for inefficient binding and pipetting error) in one lane and 40uL in another, just to make sure that the binding actually works.
 * If oligos don't show up in the untreated lane, try SYBR gold staining - ssDNA will show up better at low concentrations that way.
 * Suspicion that the magnetic beads may work better because there's no worry of accidentally loading part of the pellet and, thus, bead-bound material.
 * Use magnetic beads, concentrated with the MagnaRack initially, and run high concentrations (ie. low total volumes).

Folded:

 * 4 rxns each of:
 * c5.0.A lidless
 * c5.0.Eb
 * (Already had 4 rxns of c5.0.Fb remaining from Katie's foldings on Monday)
 * Used (instead of 9uL of 44nM p7308) 9.4uL of ~42nM p7308 per 40uL reaction, based on calculation of necessary number of moles of p7308 in the reaction.


 * NB: Noticed that there were errors in the pre-working stock table. See c5.0 Table of Working Stocks.


 * Mixed:
 * c5.0.Eb
 * c5.0.Fb

Test Solutions
TEST SOLUTIONS -- a) c5.0.A lidless (barrel) - 10uL  b) c5.0.9b (biotinylated oligos - inside) - 5uL + 5uL H2O c) c5.0.Eb (outside biotinylated barrels) - 10uL  d) c5.0.Fb (inside biotinylated barrels) - 10uL (folded 8.22.06)

Protocol
TEST CONDITIONS --- 1. Untreated 2. Beaded - pellet 100uL of magnetic streptavidin beads (NEB) and remove initial solution they were packaged in     - mix with each test solution - incubate 5 min - add 1x 30mM folding buffer to 40uL (ie. 30uL) 3. Free-streptavidined - put the 10uL of test solution in tube - mix with 3uL of 1mg/mL stock tube of streptavidin (from NEB) - add 1x 30mM folding buffer to 40uL (ie. 27uL) - incubate 5min 4. Free-strep, then beads - all of (3)'s steps - mix with pelleted beads that had had their initial suspension solutions removed - incubate 5 min

Gel

 * 2% agarose gel, 5uL EtBr, 0.5x TBE, 10mM MgCl2, 2uL of loading dye each.
 * Ran @ 100V in 0.5x TBE, 10mM MgCl2 for 1hr.

Results

 * For some reason, inside biotinylated boxes run SLOWER than outside biotinylated ones AND p7308. And they also seem to have more damage during folding, despite the fact that folding occured at 30mM MgCl2.


 * THE ASSAY WORKED...kind of.
 * The beaded biotinylated boxes (lanes 9 and 10) are much darker than the free-then-beaded biotinylated boxes (lanes 17 and 18). Free-then-beaded biotinylated boxes' bands are actually about as bright as the plain old free-streptavidinated bands (lanes 13 and 14).
 * This indicates that, at least, the outside biotinylated box was bound by the streptavidin on the beads (thus not showing up in the supernatant in lane 9), but the incubation with free streptavidin first made the box immune to binding by the beads later.
 * This is, sadly, also the case with the inside biotinylated box (lane 10), thus...


 * Unfortunately, the protection assay DIDN'T work - both the outside AND inside biotinylated boxes' bands disappeared when beaded (lanes 9 and 10). Only the outside (lane 9) biotinylated box band should have disappeared.
 * The outside biotinylated free-then-streptavidinated box band looks dimmer (lane 18) than the inside free-then-streptavidinated box (lane 17), which would normally imply that the streptavidin beads bound the INSIDE biotinylated box better and thus left less in the supernatant - this would have been a strange result indeed.
 * However, because we have the outside free-streptavidinated box control (lane 14), we can see that both free-streptavidinated outside biotinylated boxes are dimmer - this probably is due to misfolding, implied by the boxes trapped in a smear to the well in the untreated outside biotinylated box lane (lane 6).


 * Since the outside biotinylated box that was free-streptavidinated, then beaded (lane 18) shows up, but wasn't present under beaded conditions (lane 10), this indicates that:
 * 1. The inside biotinylation sites aren't safe from free streptavidin - in this case, probably not because (or not just because) the free streptavidin can access the inside of the barrel, but because the inside is being exposed.
 * 2. The inside biotinylated boxes are definitely being bound by the streptavidin beads.


 * Oligos definitely bind both beads and free streptavidin.

Conclusions & Future Plans

 * Something's going wrong with c5.0.Fb.
 * But what? We know that 9(b) biotinylated oligos (the ones that go into Fb) work fine, because they were used in this assay.
 * It's an issue with its folding - this might be the reason why it's letting the streptavidin bead get to the biotinylated sites inside: the barrel might be broken open, thus also causing it to run slower (large flat sheet vs. a more-compact barrel).


 * What should we do to improve Fb's folding?
 * Take a look at the oligo-scaffold hybridization schematics (and the oligos' sequences) going into Fb (and not going into Eb) and make sure they're the correct ones, and that there was no mistake made in organizing the stocks.
 * Remake the pre-working stocks that go into Fb, and not into anything else.
 * Remake Fb.
 * Refold Fb, with varying concentrations of MgCl2 (try up to 40mM MgCl2, and down to 10mM?).


 * Maybe continue thoughts of eluting off the beads:
 * 0.1%SDS and boiling
 * Higher amounts of proteinase K
 * Fresh trypsin, and look into EDTA being in the buffers of some of these other components/buffers.


 * Mix up the Stains-All and try again with the free streptavidin imaging once we get a correctly-folded Fb and can thus push a good protection assay?