Biomod/2011/Harvard/HarvarDNAnos:Protocols

Preparing/Running an Agarose Gel

 * Notes: Should add magnesium after microwaving since it reacts with the agarose when heated. Good to cool off the gel mixture slightly before pouring so that it does not warp the plastic. Good to run gel in ice water bath to prevent overheating.
 * Running buffer (11 mM MgCl2 in 0.5X TBE): add 0.5X TBE, keep track of how much you added since you will need to add the proportional amount of 1M MgCl2 (11 mL per 1 L of 0.5X TBE); or use the pre-prepared buffer (with Mg already added)
 * Gel (2% agarose, large MgCl2 gel): 2.4 g agarose in 600 mL beaker, 0.5X TBE up to 120 g, microwave 2 minutes, swirl. Add 1.3 mL of 1 M MgCl2 and 6 uL of 10 ug/mL EtBr for final concentration of 0.5 ug/mL. Or use 8 uL SYBR-Safe (Invitrogen) in place of the EtBr. Pour and let harden in cold room with comb.
 * Loading buffer:
 * add to 10 uL of sample: 3 uL 5X LA loading medium, 2 uL H2O
 * add to 5 uL of sample: 2 uL 6X AGLB lite, 4 uL H20
 * pipette up and down once and then load into gel lane
 * DNA ladder (1kb)
 * Distilled water - 400 μl
 * 6X Blue Loading Dye (600 uL glycerol, 150 uL 1% wt/vol orange G, 220 uL dH2O, 30 uL 1% wt/volume xylene cyanol) - 100 μl
 * DNA Ladder from NEB - 100 μl
 * Total volume - 600 μl
 * Add about 15 uL to ladder lane
 * Run at 80V for ~ 2 hours before imaging using the gel imager
 * Excision from the gel: use a fresh razor blade on top of plastic wrap on top of safe imager, first excise lane, then band, both vertically and in-plane if possible, giving yourself enough room; take a picture first
 * Filtration and recovery:
 * Crush gel slice w/ pestle in a 1.5 mL eppendorf tube
 * Spin for 1 minute at 4000 rcf (for 3 tubes, positions 1, 9 and 17 to balance the centrifuge) at 4C
 * cut off tube bottom a few mm above the 100 uL mark, invert it into spin column (actually put the cut off bottom in there upside down) [Bio-Rad 732-6165]
 * You can use the tube cutter from USA Scientific
 * Spin for 3 minutes at same settings
 * Take out the filter, trash it, and close the tube!
 * Want around 2 nM concentration, so may need to dilute in 1X folding buffer (10X folding buffer diluted 1:10 in water)

Preparing/Running a Polyacrylamide Gel

 * Use 10% TBE (no Urea) gel - take off tape on bottom, and take out the comb on top
 * Put into gel box, printed words facing outside of box (wells facing in)
 * Put in blank on other side
 * Fill with 0.5x TBE buffer to cover both of the electrodes
 * Middle section up to top
 * Outer section 1/3 the way up
 * Clear lanes of gel fluid by using squeeze pipette to displace the gel fluid with buffer
 * Load lanes with samples - if possible leave a spacer lane between ladder and samples
 * Use 0.5 uL ladder, 20 uL water, and 2 uL loading dye for the ladder lanes; pre-mix and load 11 uL of this solution into each ladder lane
 * Use 10 uL sample + 2 uL (of 6X) loading dye for sample lanes
 * Attach box head, insert box head wires into electrical source, and run the gel at specified voltage and for specified amount of time. usually, when the yellow dye band (from green loading dye) has reached the bottom of the gel, it should be done.
 * After electricity is turned on, you should see small bubbles flowing from the bottom of the gel to the top due to electrolysis of the buffer

Using the Typhoon to Image a Gel

 * Use gloves with the Typhoon; don't use gloves with the computer
 * 1) Pour staining liquid out of staining tub; fill the tub with just enough water to keep the gel wet
 * 2) Pull out the Typhoon scanning plate
 * 3) Using two hands, gently pull the bottom of the gel against a wall of the tub and slide it up the wall until you can lift the gel (with both hands) out of the tub
 * 4) Place the gel on the Typhoon scanning plate squarely, and make note of the region on which it is sitting; insert the scanning plate back into the machine
 * 5) On the computer, select the folder in which to save the images; give the image a file name (be sure not to overwrite previously scanned files!)
 * 6) Select the region on which the gel is sitting; select the staining agent that was used; adjust the PMT voltage (higher will give darker bands); choose a pixel size
 * 7) Press "Scan"; as the image scans, you can adjust the levels
 * 8) After the image is done scanning, pressing "Return" on the scanning console will automatically save the image
 * 9) Remove the gel and discard; clean the scanning plate with ethanol, water, and Kimwipes

Gel Purification

 * Excise desired band from the gel
 * Use a fresh razor blade on top of plastic wrap on top of safe imager
 * First excise lane, then band, both vertically and in-plane if possible, giving yourself enough room
 * Take a picture first
 * Filtration and recovery:
 * Crush gel slice w/ pestle in a 1.5 mL eppendorf tube
 * Spin for 1 minute at 4000 rcf (for 3 tubes, positions 1, 9 and 17 to balance the centrifuge) at 4C
 * Cut off tube bottom a few mm above the 100 uL mark, invert it into spin column (actually put the cut off bottom in there upside down) [Bio-Rad 732-6165]
 * Alternatively, use a pipette tip to scoop crushed gel out of eppendorf tube
 * Spin for 3 minutes at same settings
 * Take out the filter, trash it, and close the tube!
 * Want around 2 nM concentration, so may need to dilute in 1X folding buffer (10X folding buffer diluted 1:10 in water)
 * 5 uM Tris
 * 1 mM EDTA
 * 8 mM MgCl2

Using the NanoDrop to Analyze Concentrations

 * The NanoDrop is a "Micro-Volume UV-Vis Spectrophotometer for Nucleic Acid and Protein Quantitation"
 * For 5 nm gold nanoparticles:
 * Open the Nanodrop 2000 program
 * Click on the "UV-Vis" button
 * Adjust wavelength to 520 nm and the pathlength to 1 mm; uncheck "Use cuvette"
 * To blank: load 1.5 uL phosphine buffer directly onto the pedestal and lower the arm; press the "Blank" button
 * Clean the pedestal with water and a kimwipe
 * Load 1.5 uL of your sample onto the pedestal and lower the arm; press "Measure"
 * Multiply the A520 measurement by a factor of 10 to get your concentration in uM
 * For oligos:
 * Open the Nanodrop 2000 program
 * Click on the "Nucleic Acid" button
 * Choose the "DNA" option from the dropdown menu in the right panel
 * Set the baseline correction to 340 nm (see this reference for why), and uncheck "Use cuvette"
 * Proceed as you would for nanoparticles, using water or TE buffer (or whatever you used to resuspend the oligos) to blank
 * The measurement that you get is in units of ng/uL; to get concentration in uM, multiply this number by (1000 * moles/g) or by (nmoles/mg * 1/1000) which is molecular weight data you can get off the oligo spec sheet

Using the DLS to Determine Size and Aggregation

 * 1) Load cuvette with room temperature sample to appropriate height after mixing, or your data may convey that your sample contains extremely large aggregates (the cuvettes available at the Wyss Institute either require 1mL or 40 uL of sample)
 * 2) Open the Zetasizer program on the computer, and open a new measurement file to record data [File, New, Measurement File]
 * 3) Create a new Standard Operating Procedure (SOP) if needed to specify cuvette size and run conditions
 * 4) Run the appropriate SOP [Measurement, Start SOP] (at the Wyss Institute: "Colloidal Gold 1 mL Cuvette.sop" for 1mL cuvetter or "Gold Particle Project.sop" for 40 uL cuvette)
 * Forgetting to mix your samples can lead to misrepresented data; here is a comparison of the same sample before mixing and after mixing
 * A poly-dispersity index (PDI) of less than or equal to 0.1 means that a given sample has a uniform peak (low aggregation)
 * Intensity (first-pass measurement) is independent of Refractivity Index and Absorption (properties of the material) and is the most reliable measurement for the size of particles because it makes the least assumptions/inferences about the collected data
 * Second-pass measurement can translate intensity to relative volume of the particles of different sizes, but is less accurate
 * Each measurement will run between 10 and 30 trials
 * Use this website to convert .xps files to PDFs (to get the graphs off the program)

Imaging with the AFM

 * Puck
 * Use scotch tape to remove a thin layer of mica from an AFM puck
 * Apply 5 uL of 5x buffer and then 5 uL of your sample to the mica; let deposit
 * Place puck into magnetic receptacle in AFM
 * Carefully add an additional 30 uL of buffer to the mica, making sure not to get liquid onto the AFM
 * To "turn on" the AFM, flip switch from STM to AFM
 * Fluid Cell and Wafer
 * Using tweezers, grab the bottom of a SNL-10 wafer (each has 4 cantilevers with a tip each) and place it into the curvy trough receptacle of a fluid cell
 * Secure the wafer by pushing up on the spring, turning the lever, and relaxing the spring, so that it holds the wafer against the fluid cell
 * Using tweezers, pick up the fluid cell by a hole on the side, and invert it into the AFM opening; let it settle into the grooves
 * Using knob on back of AFM, depress the electrodes onto the cell
 * Open the Nanoscope program; use "ScanAsyst in Fluid"
 * Turn on Veeco light
 * Adjust laser using knobs on the top so that it hits cantilever, and adjust the mirror slider on the back, maximizing the SUM signal to at least 5
 * Adjust the detector using the knobs on the top so that VERT and HORZ are near 0 (+/- 1 is okay)
 * Press "Engage" until the tip finds the surface; then "Withdraw" and re-"Engage" a couple of times
 * Allow the image to scan
 * Add more fluid if the sample is getting dry (water needs to extend from the puck to the fluid cell)
 * Generally, 512 samples/line, 0.977 Hz scan rate, 3-30 nm height scale bars, and 1-5 uL scan size is a good place to start

TEM Sample Prep

 * 2% aqueous uranyl formate stain solution (3 mL H20, 0.06 g uranyl formate):
 * Uranyl formate (EMS, powder, store with parafilm over lid because seems to degrade in air)
 * Weigh out 0.06 g uranyl formate in 10 mL beaker, using the balance in the hood
 * Put in magnetic stir bar
 * Put on magnetic stirrer under large inverted foil-topped beaker (don't stir it yet)
 * Using a pipette, measure out 3 g H20 + 40 uL H20 = 3 mL H20 + 40 uL H20 (nuclease free, Ambion) in another 10 mL beaker
 * Boil this water on a hot plate (NOT the same stirrer used for the uranyl formate) and shut off heat right when it boils
 * Can carry over to the formate solution on the stirrer with gloves even though the water beaker is hot
 * Pour hot water into formate beaker and turn on magnetic stirring; cover with large inverted foil-topped beaker


 * Glow discharging the grids
 * Formvar/Carbon coated grids (SPI # 3440C)
 * Grab glass slide (e.g., VWR micro slides), gently wrap a piece of parafilm (e.g., 1.5 squares long, 1 sq wide) around a section of the glass slide
 * Using tiny tweezers, grab grids from the edge and transfer to the parafilm wrapping on the slide
 * Tweezers are Dumont N4AC from EMS. Buy your own tweezers; they break easily. Have one tweezer per grid if doing multiple grid preps in parallel.
 * Put in EMS glow discharger. Settings: 25 mA, 45 s, 0.1 mBar, negative HT polarity. Press "start".


 * Putting sample on grid and staining
 * Materials: 15 mL falcon tube (wrap in foil), filter (Acrodisk, 0.2 um) that mounts on 5 mL syringe tip, 5 mL syringe (BD)
 * Draw stain solution into the syringe, screw on filter, discharge through filter into fresh 15 mL tube
 * Add 1 mL of uranyl formate solution into a fresh 2 mL eppendorf tube using a pipette
 * Add 5 uL of 5N NaOH (JT Baker - this is DANGEROUS - DO NOT GET INTO EYES) into the 1 mL of stain solution in the eppendorf
 * Vortex; the solution should get slightly darker
 * Use a grid mat in a petri dish
 * Line up tweezers, 1 tweezer per grid
 * Grab grid edge and let rest on tweezers, suspended over air
 * Pipette 3.5 uL of sample onto the grid for 4 min
 * Take a piece of whatman paper, bend it in the middle
 * Wick off sample by bringing whatman paper in contact with the grid edge from the side
 * Immediately add 3.5 uL of stain solution onto the grid and let sit for 1 min
 * Wick off the stain as before
 * Leave for a minute or two to dry
 * Transfer grid to mat at an angle: As grid nears mat surface, open tweezers slightly--the grid will still stick to the tweezers, and drag up to a ridge on the mat, at which point the grid will detach from the tweezers and rest up against the ridge on the mat
 * It is crucial not to bend the grid or exert any forces on it
 * Put stuff that contacted uranyl formate in the radioactive waste bin
 * TEM signup link: ???

Resuspending Oligos in Solution

 * 1) Download ResuspendingDNA.xlsx
 * 2) Input requested data from order sheets contained with packing info into Columns B & D (note: for desired concentration in Column D, indicate a concentration that is double your "final desired concentration" as to minimize error and the possibility for over-dilution)
 * 3) Add specified volume of DD/DI water to samples as specified in Column F
 * 4) Vortex your samples for 15 seconds and microcentrifuge for 90 seconds
 * 5) Verify the concentration of your sample using the NanoDrop by selecting Home-->Nucleic Acid-->ssDNA, then blanking with DD or DI water, then beginning to measure samples (taking two measurements for each sample)
 * 6) Input the above data into Columns G & H, and then again add the specified volume of DD/DI water, as specified in Column Q, to your sample.
 * 7) Vortex your samples for 15 seconds and microcentrifuge for 90 seconds
 * 8) Re-nanodrop samples and input data to verify final concentrations

Basic Folding Protocol

 * Assume starting with:
 * 100 nM scaffold stock
 * Applied Biosystems Buffer Kit including:
 * 0.5 M EDTA pH 8
 * 1 M MgCl2
 * 1 M Tris pH 8.0
 * DEPC treated water
 * Staples at 100 uM concentration
 * 1) Make 10X folding buffer (50 mL) -- 50 mM Tris, 10 mM EDTA, 80 mM MgCl2
 * 2.5 mL 1 M Tris stock
 * 1 mL 0.5 M EDTA
 * 4 mL 1 M MgCl2 stock
 * Add DI water to 50 mL
 * 2) Make oligo pre-stocks (corresponding to different staple pools)
 * First spin down plates from manufacturer for 30s at 1000g
 * There are i = 1 to N_groups pre-stocks (e.g., here Ngroups = 2)
 * For each i, #i is # of staples in group i
 * Take 10 uL of each oligo from group i and put in pre-stock i
 * 3) Make working stock (mixture of staples from the different pre-stocks)
 * For each i, combine the following in a single tube: #i uL from pre-stock i
 * Each oligo now has concentration 100 uM / (total # of oligos) within the working stock
 * 4) Make folding mixture (here for a 50 uL reaction). Combine in a PCR tube:
 * 5 uL 10X folding buffer
 * (0.075 uL)*(total # oligos) from the staple working stock
 * 15 uL scaffold stock
 * Nuclease free water up to 50 uL total volume
 * These should give 150 nM concentration of each staple (5 fold excess), and 30 nM scaffold concentration, assuming staples start at 100 uM concentration from the synthesis company and that the scaffold stock is at 100 nM.
 * 5) Run the annealing rxn (here for a 6hb nanotube). Run rxn mix in PCR machine with following annealing program:
 * 1: 80C for 5 min
 * 2: 80C for X min
 * -1C per cycle
 * Goto 2 60 times
 * End
 * For the 6hb nanotube (or 2D structure?), X = 2 minutes will work


 * Making 100 nM scaffold stock
 * Measuring scaffold concentration on the nanodrop:
 * 1.5 uL on the nanodrop, nucleic acid mode, custom extinction coeff = 37
 * Nanodrop outputs a value in ng/uL. Convert from ng/uL to nM using spreadsheet that takes into account exact molecular weight of scaffold (see 2010-08-18 folder which has spreadsheet with this value for many of our scaffolds). Nanodrop reading varies over time (due to evaporation?). For concentrated samples, may be wise to consider the actual concentration value as ~ 10% lower than the nanodrop reading converted to nM.

Making 2D Origami

 * 10x origami folding buffer (50 mL) -- 50 mM Tris, 10 mM EDTA, 12.5 mM MgCl2
 * 2.5 mL 1 M Tris stock
 * 1 mL 0.5 M EDTA
 * 6.25 mL 1 M MgCl2 stock
 * Add DI water to 50 mL
 * Folding mixtures for origami:
 * There are ~ 224 staples in the tall rectangle, depending on removal of edge staples for de-stacking etc. Using a multichannel pipette, take 5 uL from each plate well on your plates of staples (we're assuming you ordered your plates hydrated at 100 uM concentration) and combine in a 2 mL eppendorf tube. This is the staple stock.
 * Thus for a 50 uL reaction we can use:
 * 5 uL 10X folding buffer
 * (0.15 uL)*224 = 33 uL from the staple working stock
 * 10 uL scaffold stock
 * 2 uL nuclease free water (optional, since the exact concentrations don't matter that much)
 * These should give 300 nM concentration of each staple (15 fold excess), and ~ 20 nM scaffold concentration, assuming staples start at 100 uM concentration from the synthesis company and that the scaffold stock (e.g., m13mp18 from NEB) is at ~ 100 nM. 15 fold excess is more than enough to get near 100% yield of excellent origami.
 * Compare with Paul Rothemund's original protocol: "Staple and remainder strands were purchased unpurified ... in water at 100 uM or 150 uM and stored at -20C. The desired set of up to 273 short strands was mixed with M13mp18 (typically 160 nM of each short strand, 1.6 nM M13mp18 circular or linear, a 100-fold excess of short strands) in a 100 uL volume of 1X Tris-Acetate-EDTA (TAE) buffer with 12.5 mM magnesium acetate (pH=8.3) and annealed from 95C to 20C in a PCR machine (Eppendorf) at a rate of 1C/minute in .1C steps."
 * NPRTF folding reaction for 2D origami (the whole anneal takes only ~ 1 hour):
 * 95C for 5 min
 * 90C for 1 min
 * Cool to < 40C at 1 minute per degree (1 degree steps are OK)
 * Take out the sample, or continue cooling gradually to room temperature
 * Right after folding (with no purification) they look like this under fluid tapping AFM (see below):
 * [[Image: 2011-07-03-TRO-afm.jpg|500x500px]]


 * Amicon purification of DNA origami (courtesy of Ralf Jungmann)
 * Buffer = 1X folding buffer
 * Amicon Ultra 0.5 ml Centrifugal Filters (100000 MWCO)
 * Put folded origami solution into filter
 * Add buffer solution to a total volume of 500 µl
 * Put filter device into filtrate collection tube
 * Spin for 5 min @ 14000 x g. This gives ~ 20 µl volume in the filter device
 * Fill it up again to 500 µl total volume and repeat the last step. Make sure to empty the filtrate collection tube beforehand!
 * After the second filtration step, turn the filter device upside down and spin for 3 minutes @ 1000 x g into a new filtrate collection tube
 * Fill up to the original sample volume with buffer
 * AFM imaging of 2D origami:
 * Nanoscope 3 or 5, fluid tapping mode
 * Cleave mica with scotch tape until a nice smooth surface is obtained
 * Add 10 uL of purified origami sample, wait 5 minutes
 * Cover with 60 uL of 1X folding buffer
 * Correct AFM imaging conditions for fluid tapping mode
 * Adjust laser as usual in AFM/LFM mode
 * Autotune w/ target amplitude = 0.4V, peak offset 8%, prop gain 0.3, int gain 0.15
 * Engage then switch mode on AFM body to TM setting
 * If vertical deflection = 9.99, increase amplitude setpoint until it goes down (to -9.99 is OK)
 * Adjust z limit so piezo is neither fully extended nor fully retracted
 * 1-3 Hz scan rate should be good

Making 3D Origami with "Complex Curvatures" (e.g. a sphere)

 * Based on Han et al.
 * JSON file: [[Media:Sphere_match.json]]
 * Excel staple file: [[Media:Sphere_staples.xlsx]]
 * Buffer: 5 mM Tris, 1 mM EDTA and 16 mM MgCl2
 * Formulation for 10x buffer: 2.5 mL of 1 M Tris, 1 mL of 0.5 M EDTA, 8 mL of 1 M MgCl2 and 38.5 mL H2O
 * Scaffold: 10 nm single-stranded M13mp18 DNA
 * Staples: 10 times molar excess in buffer
 * Cycle: 94°C to 86°C at 4°C per 5 minutes, 85°C to 70°C at 1°C per 5 minutes, 70°C to 40°C at 1°C per 15 minutes, 40°C to 25°C at 1°C per 10 minutes
 * Staple pools and reaction mix: [[Media:Staple_wells.xlsx]]

Making AuNP of Particular Sizes
Taken from Polysciences, Inc.; * one hour to develop.
 * To make 100 ml of gold solution, two stock solutions have to be prepared:
 * Solution A
 * 80 ml distilled water and 1 ml 1% aqueous gold chloride.
 * Solution B
 * 4 ml 1% tri-sodium citrate. 2H2O + 16 ml H2O + variable amount of 1% tannic acid (See below)
 * When 1 ml or more tannic acid is needed, add an equal amount of 25mM potassium carbonate for pH adjustment.
 * Warm up solutions A and B to 60°C on a hot plate, with a thermometer, and mix them while stirring with a magnetic stir bar. When the red color has formed heat up to 95°C and cool the solution on ice.
 * The larger particles (where lower concentrations of tannic acid are used) take longer to form and the red color can take up to

AuNP-DNA Conjugation

 * AuNP-DNA conjugation (1 or a few DNA strands per nanoparticle). This is a draft protocol based on one from Chenxiang Lin, with additional info from Shawn Douglas and Steve Perrault.
 * 50 mL gold solution (from Ted Pella, Sigma, or the one we made ourselves)
 * Add 10 mg phosphine powder, cover tube with aluminum foil, stir slowly for 24 hours (Day #1)
 * Add about 5 g NaCl powder until the color changes from red to purple/blue (Day #2): we're aiming for a NaCl concentration of about 3M in the solution
 * Centrifuge at 1000g in 50 mL conical tube (remember to use a balance tube) for 30 minutes. You will see a black pellet.
 * Remove supernatant (light yellow colored) with 25 mL electronic pipetter. Remove last 1 mL or so using standard pipette, being careful not to dislodge the black material from the pellet.
 * Add 1 mL of 2.5 mM phosphine buffer (66.9 mg phosphine powder in 50 mL H2O: http://www.wolframalpha.com/input/?i=535g%2Fmol+*+2.5+mM+*+50+mL) and resuspend pellet by pipetting. Solution will be red.
 * Add 2.5 mL methanol and centrifuge at 1000g for 30 minutes
 * Remove supernatant, resuspend in 500 uL of 2.5 mM phosphine and transfer to 1.5 mL tube, covered in foil
 * This can sit in 4C fridge for months at a time
 * Measure a UV-vis spectrum and record A520. If using the nanodrop with 1 mm path length (i.e., not in cuvette mode): add 1.5 uL of nanoparticle solution to the nanodrop, blanking against phosphine buffer, and record A520, then multiply this value by 10 to obtain your estimate of the nanoparticle concentration c_AuNP, which should be between 1 and 5, corresponding to 1 uM and 5 uM respectively.
 * Save some of these unconjugated AuNPs for comparision with the conjugated ones on a gel
 * Make a 1-1 molar solution of gold particles and DNA: if your DNA strand is at 100 uM and your nanoparticles are at 2 uM, add 400 uL AuNP solution to give 800 pmoles of nanoparticles and mix this with 8 uL of the DNA solution to also give 800 pmoles
 * Add 10x TBE to get ~0.5x TBE final concentration (20x dilution)
 * Add e.g. 5M NaCl to get 50 mM NaCl final concentration (100x dilution)
 * Leave for 40 hours on slow shaker, covered in foil
 * To extract nanoparticles via a gel, see the Sucrose Extraction protocol below.


 * Here is another protocol for AuNP-Gold conjugation: [[Media:Current protocols in nucleic acid chemistryedited by Serge L-1. Beaucage...-et al.- 2002 Taton.pdf | PDF]]

Nanoparticle Sucrose Extraction

 * Courtesy of Steve Perrault (Shih Lab)
 * They should be starting with AuNP which have been functionalized with oligo. I've found that the gel-based purification is more efficient (higher recovery) if I start with a higher concentration smaller volume of product.  This allows fewer wells to be used.  Anything up to about 1 mL can be put into a gel but if they can get it down to 100-200 uL, the job will be easier.  They can concentrate the solution by freeze-drying the sample or by an Amicon column if they want to try this.
 * 1) Pour a gel for sucrose extraction.  Pour 50 mL of 4% agarose to produce a base, without a comb.  Once set, pour the top using 140mL of agarose, I think it's 3% but you might want to double check that with Chenxiang.  Use standard 0.5X TBE for these, obviously no Mg is needed.  Use an appropriate comb with wells taped to give you enough space for the volume of product.
 * 2) You don't need to use DNA loading buffer, just add 1/5 volume of 40% sucrose solution to increase the density, and add it to the gel.
 * 3) Run the gel at 100V for however long you need to get good separation.  Again, it helps to start with a highly concentrated solution as you can see the bands really nicely.
 * 4) Once finished, you're going to cut the gel for sucrose extraction, but only the top of the gel, leaving the bottom base intact.  Cut a well in front of the band that you want to extract.  It should be slightly wide (5mm) on either side, and about 1 cm wide in the direction of electrophoresis.  Then, make cuts behind the band all the way up to the top, so that you remove all the other bands and wells.  You should have just a single bridge of agarose with your band in it, with a pool of buffer behind.
 * 5) Add the sucrose solution into the well in front of the band (0.5X TBE and I think it's 30 or 40% sucrose, check with Chenxiang).
 * 6) Run the gel for 5-10 min intervals, keeping an eye on the band.  The gold NP make it really easy to see when you need to extract the solution in the well and add new.  We typically find that ~8 min, 100V intervals work well.
 * 7) Do this as many times as necessary to retrieve the entire band product.  Purify it and do a buffer exchange with water using an Amicon column.  I recommend doing at least 2X good washes with H2O.
 * 8) Finally, you should do absorption spectroscopy, EM and DLS to characterize your product.  I've found (and Wei told me this) that having an NaCl solution of ~15 mM is necessary in your EM prep, to keep the particles from aggregating as they dry on the grid. DLS should show a population of (hopefully) monodispersed particles in the 10nm range, but that depends on the ligand you added to the surface.  This characterization is important b/c it will give you a comparator for down the road, if you think your particles might be aggregating.
 * 9) For long term storage, you can powder fractions of the product using the freeze-dryer, replace the atmosphere with Argon and throw them in the freezer.  They'll be good for a long time that way, probably only weeks to a couple months otherwise.
 * 10) Determine their concentration using absorption.  I have a function that will tell you the extinction coefficienct of AuNP based on their size, but it's at work so I can't get it to you until I'm back.  If they look pretty close to 5 nm by TEM (the cores), you can use the coefficient provided by Sigma or Ted Pella.

Other
 