License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: January 22, 2018
Last Modified: August 28, 2018
Protocol Integer ID: 9751
Abstract
The intention of this protocol is to isolate high molecular weight DNA. This means you should avoid any pipetting without using a wide-bore or cut off pipette tip, vortexing, mixer shakers or anything else which generate a velocity gradient which may shear the DNA. In addition you should be very careful not to introduce nucleases by making up buffers with nuclease-free water. Avoid unnecessary heating and do not freeze, isolated DNA should be stored in the fridge, a good extraction will be stable for months. Currently tested on E. coli and human cell lines, however it is likely to work with many gram-negative bacteria and mammalian cells.
Guidelines
Acknowledgements:
This protocol was developed by Josh Quick for the Nanopore WGS Consortium (https://github.com/nanopore-wgs-consortium). Updating this protocol to keep it working with each new version of the rapid kit would not be possible without the contributions of Nick Loman, Matt Loose and John Tyson. We would also like to thank David Stoddart, Simon Mayes and Daniel Turner at Oxford Nanopore for their support. Please follow on Twitter for latest updates and results:
20 µg/ml Qiagen RNase A (add fresh just before use)
EB+Triton-X100:
10 mM Tris-Cl pH 8.0, 0.02% Triton X-100
Expected Results for Human:
The NA12878 data in rel4 was generated using RAD002 kits. We put a lot of effort into reproducing these run metrics using the updated RAD003 kits and although we discovered many optimisations to the standard protocol, read lengths in particular fell short of what we had previously generated. The RAD004 kit featured changes to the tethering chemistry and we have observed a significant improvement in the performance of the kit. The yield for ultra-long read flowcells is expected to be lower than those with standard libraries, we attribute this to membrane damaged caused by the very high concentrations of DNA. We have been able to generate 50-100,000 reads per flowcell (1-2 Gb) although when working with HMW DNA we expect some variability due to DNA heterogeneity. A read length N50 of 100 Kb is considered an excellent result. The longest read sequenced with this protocol is 1.2 Mbp by Matt Loose.
Example run metrics for RAD004:
General summary:
Number of reads: 33,907
Total bases: 1,324,105,498
Median read length: 17,472
Mean read length: 39,051.1
Read length N50: 90,923
Active channels: 492
Top 5 longest reads and their mean basecall quality score:
Read length histogram view from MinKNOW 1.23 in playback mode after one hour. A lot of the data is off the scale which only extends to 72,000 events. Reads longer than this are not counted in the yield value which means the standard scaling factor of bases = events * 1.4 will not apply.
Channel states view from MinKNOW 1.23 in playback mode from the start of the run. Pore occupancy (strand / strand + single pore) of 0.75 should be achievable with this method. Overall numbers in strand with RAD004 are often >200 but fall quickly, regular restarts are needed to achieve the maximum yield.
Pulsed-field gel, taken from http://biorxiv.org/content/biorxiv/early/2017/04/20/128835.full.pdf showing different extractions, this method is in lanes 8 and 9. Smear extends to about 350 kb but closely matches E. coli DNA extracted using the same method demonstrating it is the extraction method not the starting size of the chromosomes driving the size.
Materials
MATERIALS
1X PBS (Phosphate-buffered saline )
RNase AQiagenCatalog #19101
Nuclease-free Water
AccuGene molecular biology water LonzaCatalog #51200
When handling phenol always wear PPE, keep a solution of 50% (w/v) PEG-400 nearby to treat the burn in the case of accidental splashes.
DNA Extraction
DNA Extraction
Take 5x107 human cell pellet fresh or stored at -80°C in a 50 ml Falcon tube (for E. coli I used an overnight culture of E. coli growing in 50 ml LB broth spun down at 4500 x g for 10 minutes)
Note
DNA extraction protocol adapted from Molecular Cloning by Sambrook and Russell (third edition) Chapter 6 protocol 1.
Resuspend by pipette mixing in 200 µl sterile PBS.
200 µL PBS
1X PBS (Phosphate-buffered saline )
Add 10 ml TLB and vortex at full speed for 5 seconds.
10 mL TLB
00:00:05
Note
Vortexing here is to thoughly resuspend the cells to achieve a homogenous DNA solution once cells are lysed, DNA will not be damaged as it is still packaged inside the cell.
Incubate at 37°C for 1 hour.
37 °C
01:00:00
Note
Solution will turn transparent as the cells lyse.
Add 100 µl Qiagen Proteinase K or other stock solution to a final concentration of 200 µg/ml. Mix by slowly rotating end-over-end 3 times.
100 µL Proteinase K
Proteinase K (2 ml)QiagenCatalog #19131
Incubate at 50°C for 2 hours, mix every 30 minutes by slowly rotating end-over-end 3 times.
50 °C
02:00:00
Add light phase-lock gel to 2 x 15 ml Falcons. If it is only available in 2 ml tubes, transfer it by cutting the lid off 3 x 2 ml tubes and spinning it out into each 15 ml Falcon.
Note
15 ml Falcons are used as they are narrower decreasing the surface area of the interface/gel. Using two means they balance each other in the centrifuge and gives phenol space to move which improves the emulsion.
Split the viscous lysate into the two 15 ml Falcon tubes prepared with phase-lock gel, this is easiest using a 10 ml serological pipette at slow speed.
Add 5 ml recently opened BioUltra TE-saturated phenol to each Falcon tube containing lysate.
Place on a HulaMixer at 20 rpm for 10 minutes, if a fine emulsion has not formed after a minute gradually increase the rotation speed.
00:10:00
Spin in a centrifuge at 4500 rpm for 10 minutes.
00:10:00
Pour the aqueous phases into two new 15 ml Falcon tubes containing phase-lock gel, try to avoid transferring any protein which may form a white layer above the phase-lock gel.
Add 2.5 ml buffer saturated phenol and 2.5 ml chloroform-isoamyl alcohol 24:1 to each tube.
Add 30 ml ice-cold ethanol and watch the DNA precipitate, bubbles will over time pull the mass of DNA to the surface so it looks like a Jellyfish with tentacles hanging down.
Make a hook by melting the tip of glass capillary in a blue flame so it curls over.
Hook out the DNA in one-piece if possible lift up and allow the excess liquid to drip off.
Submerge the DNA in a 50 ml Falcon tube containing 70% ethanol.
Note
It should have fully tightened up into a whitish opaque pellet.
Carefully work the pellet off the glass rod using the rim of an Eppendorf tube, let the pellet drop into the tube.
Go back for the rest of the DNA if it broke apart and repeat.
Add 1 ml 70% ethanol to the Eppendorf.
1 mL 70% ethanol
Spin down at 10,000 xg then remove as much of the 70% ethanol as possible.
Wash again with 1 ml 70% ethanol.
1 mL 70% ethanol
Spin down at 10,000 xg then remove as much of the 70% ethanol as possible.
Let the remaining ethanol evaporate by leaving at RT for 15 minutes.
00:15:00
20 °C
Add 100 µl EB + Triton-X100 and incubate without mixing at 5°C for 2 days to allow the pellet to fully resuspend into a translucent viscous gel.
100 µL EB+Triton-X100
48:00:00
Note
We have found that adding Triton-X100 to a final concentration of 0.02% dramatically improves the transposase activity. If you don't want to add it to the stock DNA you can add it when you make the library.
5 °C
QC DNA
QC DNA
Take a P2 pipette set to 1.5 µl with a cut-off tip and aspirate very slowly then try to work the DNA in the tip away from the DNA in the tube by continuously pulling and releasing the DNA to work it gradually away. Adjust the pipette volume to determine the volume of DNA you actually removed.
Quantify the DNA on the Qubit BR assay, the concentration should be >1 µg/µl, it is important to use BR as the buffer contains a detergent which appears to improve mixing.
Check the absorbance spectrum of the 1:10 diluted DNA on the NanoDrop, example trace below (ratios typically a little higher than other methods)
If necessary add more EB+Triton-X100 to the stock DNA to adjust concentration to 1 µg/ul but do not mix.
Leave overnight at 5°C before use.
5 °C
Library preparation with RAD004
Library preparation with RAD004
As slowly as you can pipette 16 µl DNA into a 0.2 ml PCR tube using a cut-off P20 pipette tip, retain the tip.
16 µL DNA
Remove 1 µl from the tube as before and quantify using Qubit BR assay.
1 µL
Note
DNA may still be heterogenous in concentration and as such this is the most reliable way to quantify actually input.
Add 1.5 µl FRA and 3.5 µl EB+Triton-X100. Using a P20 set to 18 µl and the tip from before, mix up and down as slowly as possible 8 times, retain the tip.
1.5 µL FRA
3.5 µL EB+Triton-X100
Note
Take care not to introduce bubbles as they are hard to remove.
Note
We have found that 1.5 µl FRA is suffient to generate an efficient library and any more can lead to over-fragmentation.
Using a thermocycler incubate at 30°C for 1 minute, 80°C for 1 minute then hold at 4°C.
30 °C
00:01:00
80 °C
00:01:00
4 °C hold
Add 1 µl RAP. Using a P20 set to 19 µl and the tip from before, mix up and down as slowly as possible 8 times, discard tip.
1 µL RAP
Incubate at room temperature while you prime the flowcell.
Prime flowcell
Prime flowcell
Add 30 µl FLT to tube of FLB, vortex briefly and spin down. This is the flush mix.
30 µL FLT
Place new flowcell on MinION and run platform QC .
Using a P1000 remove a little storage buffer from the inlet port using the volume adjustment screw. Load 800 µl flush mix via the inlet port slowly using the plunger. Wait 5 minutes.
800 µL Flush mix
00:05:00
Lift the cover off the SpotON port. Load 200 µl flush mix via the inlet port slowly using the plunger, try to dispense at a speed where a bead of liquid becomes visible over the SpotON port which then gets siphoned back in.
200 µL Flush mix
Load library
Load library
Add 34 µl SQT and 20 µl NFW to the library tube.
Note
Loading beads are not used as they clump when mixed with the library.
Using a P100 set to 75 µl with a cut-off tip mix up and down as slowly as possible 5 times. On the final mix slowly pipette the diluted library onto the SpotON port as it gets siphoned in, this can take much longer than usual due to the viscosity.
Note
If it gets blocked abandon the siphon and load by holding the pipette vertically against the SpotON port and positively pipetting the library directly in.