Biomod/2011/Caltech/DeoxyriboNucleicAwesome/Protocols

Wednesday, September 24, 2014

Home

Members

Project

Protocols

Progress

Discussion

References

Protocols

Contents

Things that need to be written

• How to run a gel
• How to do AFM (when we learn this)
• How we ran experiments...

Quantitating Newly Ordered DNA Strands

• Provided by Damien Woods

This is a 2-step protocol where we first try to set the DNA concentration to be very roughly 100 μM, and then adjust to bring it close to 50 μM. For a faster 1-step protocol you could easily adjust the calculations and go straight for 50 μM.

The protocol assumes we will dissolve in 1x TE, some people use Milli Q H20 instead.

1. Switch on biophotometer. Prepare bench, lay out the things you'll need (for consistency, do this the same way every time you do the protocol).

2. Spin down lyophilized (dried) DNA to make sure there are no flakes on/near lid.

3. First, shoot for 200 μM: If IDT claim to give us x nanomoles, we add 5x μL 1x TE, to the dried DNA, to get roughly 200μM, explained as follows:

$\frac{x \times 10^{-9}}{5x \times 10^{-6}} \quad \frac{moles}{litre} \quad =\quad \frac{2x \times 10^{-9}}{x \times 10^{-5}} \quad \frac{moles}{litre} \quad =\quad 2\times 10^{-4} \quad \frac{moles}{litre}$

$\quad=\quad 200 \times 10^{-6} \quad \frac{moles}{litre} \quad =\quad 200 \quad \frac{micro moles}{litre} \quad = \quad 200 \, \mu$M

4. Vortex (mix) for a few minutes. Spin down using microcentrifuge to remove any droplets form the lid/top of test tube. Vortex again. Spin. It is important to get the DNA completely into solution, so mix well. Finish by spinning down.

5. Put 98 μL of solvent (1X TE buffer) into a plastic cuvette. Place in biophotometer with (upwards pointing) triangle facing you. Push down in a specific pattern (top-left-bottom-right, then bottom-left-top-right, then top left). (Insert the cuvette in exactly same direction, and using the same technique, every time.) Set large pipettor to one side (do not let the tip touch anything, I leave it over the side of the bench).

6. Hit "blank".

7. Add 2 μL of DNA. Mix with small pipettor (suck up and down). Mix with large pipettor (suck up and down very carefully). Try not to introduce bubbles, nor splash on side of pipettor.

8. Insert cuvette, using the same technique as before. Hit "sample", and record the value.

9. Mix with large pipettor (stir, while in biophotometer), record value.

10. Mix with large pipettor (remove from biophotometer, suck up and down), record value. Discard cuvette.

11. If the 3 values are close to each other (within a few percent), then record the first one. If not, try again.

12. Make a note of the fact you have removed 2 μL from the DNA.

13. Repeat at least 3 times, by going back to step 5 on each iteration. Usually I do 3 to 5 samples. If the results are not consistent, re-mix, and do more. Let A260 be the average of all samples. (If there is one outlier I ignore it.)

14. Use the following formula to find the concentration: []μM = dilution-factor $\times \frac{A_{260}}{\epsilon} \times 10^6$

Here dilution-factor = 50 (i.e. you put $2 \, \mu$L of DNA into $98 \, \mu$L 1x TE), A260 is the absorbance value, and ε is the extinction coefficient of your oligo usually given in /M/cm (supplied by IDT with your oligo, or else use the in-house DNA group code for double stranded complexes). This comes from: $A_{260} = C \times L \times \epsilon$ where C is the concentration, L is the path length (1 cm for the Biophotometer). And this in turn comes from the Beer-Lambert Law. See IDT's page: http://www.idtdna.com/analyzer/applications/Instructions/Default.aspx?AnalyzerDefinitions=true#ExtinctionCoefficient

15. Shoot for 100 μM. We will calculate how much 1x TE to add in order to get as close as possible to 100 μM. To get the current volume of DNA in solution, subtract $2y \, \mu$L from the initial volume ($5x \, \mu$L), where y is the number of DNA samples you took. Then apply the following formula to find out how much 1x TE to add to get the desired concentration (100 μM in our case):

$\textrm{volume\_to\_add} \, (\mu L) = \frac{\textrm{current\_concentration} \, (\mu M) \times \textrm{current\_volume} \, (\mu L) }{\textrm{desired\_concentration} \, (\mu M)} \,\, - \,\, \textrm{current\_volume} \, (\mu L)$


16. Add $\textrm{volume\_to\_add} \, (\mu \textrm{L})$ to the solution. Mix, spin, mix spin.

17. Do steps 4-14 with the new sample. If everything went well, you should get a value close to 100 μM.

Gel Purification

• 1. Anneal the complex (Q-function)

For 300ul sample, add 50ul BB_ND dye 170ul of sample will be loaded per lane. Each sample will take 2 lanes, and one blank lane should be placed between different samples.

• 2. Wash the big plates. Dry the plates for 45 minutes

• 3. Make a gel. 12% ND gel for purification

Lulu Qian’s recipe for 40ml of 12% ND gel which works for one gel - 12ml 40% Acrylamide - 4ml 10x TAE/Mg++ - Fill up to 40ul MQ water (add 24 ml) - 240ul 10% APS, slow mix - 24ul TEMED, slow mix

   APS should be no more than 10 days old

• 4. Run the gel under 150V for roughly 4.5 hours ~ 6 hours (when the dye nearly runs off the gel) depending on the size of the sample. For example, if the sample is 122nt~176nt, run for 6 hours. Change the buffer every 1.5 hours.

SPEX protocol

• 1. Turn on the lamp (back side). Wait for 30 min

Bottom big on/off button is for the machine. It is always on. Top small one is for the lamp. If there’s somebody who will use the SPEX tomorrow, then do not turn off the lamp after the experiment.

• 2. Turn on the computer if it’s not on.
• 3. Software: Instrument control center

Default: Click OK

• 4. Temperature set up

Click the third square icon called “Visual setup” Click the middle square Temperature control icon: Turn on and Set to 25 degreeC Choose T bath (internal). NOT Probe (external).

• 5. Wash the cuvettes.

Be careful! Hold only top or bottom. DW (10 times) 70% ethanol (twice) DW (5 times) 70% ethanol (once) While you wash, you can touch the side. Hold it in a stable way. Let it dry (upside down)

• 6. Dry the cuvettes using lens paper.
• 7. Click second square icon called “real time display”.

High voltage: HV on, S 950, T 950 Bandpass : Set 2 for everything

• 8. Click the last square icon called “run CWA”.

Open a file called “ROX_CWA” from e:\SPEX DATA\biomod Optional mode: kinetics, Integration time: 10s Define: Repeat acquisition: 1 min (minimum) for total of 180 min (3 hours)

• 9. Mixing: Hold top. 350ul of 1000ul pipette all the way in. Mix it for 30 times.
• 10. Put the cuvettes. Logo faces out!
• 11. Start acquisition. Run sample.

Range of the real data should not exceed 1 million (1000,000) Number we see is S/R  will be (100 million)

• 12. Add more samples in the middle of SPEX experiment

When the status is “waiting”, click “stop”. Don’t click anything else. Add and mix. After you put the cuvettes inside the box, click “no”.

• 13. Save the data

When the status is “waiting”, click “stop”. Click “yes” Save data in two forms (DWA datafile, multi~ test) Check if the data is properly saved: Check the last data.

• If you want to see the reference, go to options: view option and display raw R.

Origami Purification

• Provided by Nadine Dabby and Sungwook Woo.

We suggest using the Millipore Microcon YM-100 (blue) spin filters. ( http://www.millipore.com/catalogue/item/42413 ). The times below are estimates for how long to spin your sample (to err on the safe side I would check on the sample after 30 minutes and then estimate how much longer one needs to spin it so as to preserve a small amount of liquid on top of the filter but being VERY careful not to allow it to dry out completely.

this protocol should drastically reduce the presence of excess strands. (the more rounds of adding buffer and filtering you do, the cleaner your sample, but the lower the yield.

Admittedly the yield will be quite low (we haven't figured out exactly how low) -- supposedly each round of filtration should preserve 95% of the origami according to the filter spec but one problem here is dilution (since I have almost always ended up with a sample that is 3X more dilute than I started with). The dilution can be fixed by vaccuum centrifuging at 30 degrees (this does not hurt the origami at all according to Sungwook, but I haven't tried this with our substrate-laden samples.

1. Mix 100uL of your sample with 300uL of 1X buffer of your choice.

2. Vortex mixture and pipette it into a YM-100 spin filter.

3. Spin at 100g for ~30 minutes at 4°C. (Spinning at this low speed prevents the origami from being ripped apart)

4. Remove sample for centrifuge, leaving the machine at 4°C because we’ll be using it again in a moment. The spin filter should have only a few tens of microliters of your sample left in it, just enough to cover the filter. Remove the filter and get rid of any liquid in the tube below. Now reinsert your spin filter and add 200uL of 1X buffer, then pipette up and down just over the filter (five to six times) to recover any bound origami on the filter. Be very careful not to puncture the filter during this step. Finally, add another 200uL of buffer to your sample to bring the final volume to ~400uL.

5. Spin at 100g for ~30 minutes at 4°C.

6. Remove sample for centrifuge, leaving the machine at 4°C because we’ll be using it again in a moment. The spin filter should have only a few tens of microliters of your sample left in it, just enough to cover the filter. Remove the filter and get rid of any liquid in the tube below. Now reinsert your spin filter and add 200uL of 1X buffer, then pipette up and down just over the filter (five to six times) to recover any bound origami on the filter. Be very careful not to puncture the filter during this step. Finally, add another 200uL of buffer to your sample to bring the final volume to ~400uL.

7. Spin at 100g for ~30 minutes at 4°C.

8. Remove sample for centrifuge, leaving the machine at 4°C because we’ll be using it again in a moment. The spin filter should have only a few tens of microliters of your sample left in it, just enough to cover the filter. Remove the filter and get rid of any liquid in the tube below. Now reinsert your spin filter and add 200uL of 1X buffer, then pipette up and down just over the filter (five to six times) to recover any bound origami on the filter. Be very careful not to puncture the filter during this step. Finally, add another 200uL of buffer to your sample to bring the final volume to ~400uL.

9. Spin at 100g for ~20 minutes at 4°C. This should leave you with ~70-100uL of sample retained in the spin filter. Perform the above pipetting step again to recover as many origami as possible from the spin filter.

10. Remove the spin filter, and place it upside down in a fresh tube (being careful not to dump your sample on the ground!). You can now spin your sample in a microcentrifuge for 12 minutes to collect everything.

A similar spin protocol appears necessary for binding DNA origami to sticky patches on silicon. Without it, the staples seem to interfere in binding. This protocol was originally taken from Rob Barish's protocol for origami/ribbon constructs. I think that some parts of the protocol may not be necessary for regular origami. Notably, I'm not sure that pipetting up and down is necessary, except perhaps at the last step to maximize recovery. Similarly the use of a 4C centrifuge seems unnecessary, the centrifugation may go much faster at room temperature. If you use a different centrifuge or temperature, calibrate the passage of buffer through the filter before you try purifying any origami to get an idea of how much buffer is retained in the filter after a spin. Fewer spins will be required if a smaller excess of staples is used. You should be able to get away with 10X fewer staples than we typically use according to my standard origami protocol, easily. That is, a 5-10X stoichiometric excess of staples over scaffold is probably fine and will result in cleaner origami faster.

How to Open SPEX Data in MATLAB

• Open up MATLAB and use File->Import Data
• Navigate to the raw-text (.txt) data file and open it
• The table on the right should show eight rows, where the odd ones are the times associated with the signal from the row under them. The top set is sample 1; the next is 2, the next is 3, the bottom set is sample 4. Press "next."
• Optionally right click on "data" and rename it to whatever you want. This isn't necessary unless you're afraid of your data getting overwritten eventually when you import another set.
• Click "finish."
• Your data should be visible in the "workspace" window with the name "data" or whatever you might have renamed it to. Right click on it -> "open selection" to view it.
• To plot a single sample's data, use this, where "data" is the name of your data variable:
plot(data(1, :), data(2, :))
• To plot all four samples' data on the same plot, use this:
plot(data(1, :), data(2, :), data(3, :), data(4, :), data(5, :), data(6, :), data(7, :), data(8, :))

How to normalize SPEX Data in MATLAB

• Upon opening the data, use the following code, replacing "start" with the beginning of the experiment (for origami experiments, this will usually be immediately after the walker trigger is added), and replacing "finish" with the data point corresponding to the reaction reaching its completion level.
• Your data should be visible in the "workspace" window with the name "data" or whatever you might have renamed it to. Right click on it -> "open selection" to view it.
• The data is normalized such that the start of the reaction is at 1, and the completion level is at 0, by translating the completion level down, and then scaling the data.
time1 = data(1, start:finish);
time2 = data(3, start:finish);
time3 = data(5, start:finish);
time4 = data(7, start:finish);
data1 = data(2, start:finish);
data2 = data(4, start:finish);
data3 = data(6, start:finish);
data4 = data(8, start:finish);
time1 = time1 - time1(1);
time2 = time2 - time2(1);
time3 = time3 - time3(1);
time4 = time4 - time4(1);
ndata1 = data1 - data1(finish - start + 1);
ndata2 = data2 - data2(finish - start + 1);
ndata3 = data3 - data3(finish - start + 1);
ndata4 = data4 - data2(finish - start + 1);
ndata1 = ndata1/ndata1(1);
ndata2 = ndata2/ndata2(1);
ndata3 = ndata3/ndata3(1);
ndata4 = ndata4/ndata4(1);
plot(time1, ndata1, time2, ndata2, time3, ndata3, time4, ndata4);