Introduction

This protocol describes a method to extract cell-free DNA (cfDNA) from 12 human blood samples collected in EDTA K2 vacuum tubes (step 1), or human plasma (step 3). The extraction is performed using the QIAGEN QIAamp MinElute ccfDNA Midi Kit.

Note: We recommend that blood samples are processed while fresh, as we have observed potential gDNA contamination arising from blood that has been stored in certain types of collection tubes.

Following this extraction method, we recommend the extracted human blood cfDNA samples are prepared for multiplex sequencing using our Ligation sequencing V14 - Human cfDNA multiplex (SQK-NBD114.24) protocol for PromethION.

We have tested our library preparation method with cfDNA inputs of 6, 4, and 2 ng (per barcode), and have seen that ≥6 ng provides optimal flow cell performance and data yield. This is achievable from as little as 1 ml of plasma (~3.5 ml of blood) (see Table 1 in the results section of this document).

The read length distribution was shown to match the expected cfDNA characteristics, with nucleosome-associated peaks.

For further information on fragment and sequencing read length of cfDNA libraries, see our cfDNA knowhow.

Materials

• ≥3.5 ml blood in EDTA K2 vacuum tube or ≥1ml plasma, per sample
• QIAamp MinElute ccfDNA Midi Kit

Consumables

• Qubit dsDNA HS Assay Kit (Invitrogen, Q32851)
• 100% ethanol
• Isopropanol
• Freshly prepared 80% ethanol in nuclease-free water
• Nuclease-free water (e.g. ThermoFisher, cat # AM9937)
• Qubit™ Assay Tubes (Invitrogen, Q32856)
• 5 ml DNA LoBind Eppendorf tubes
• 15 ml Falcon tubes
• 0.2 ml thin-walled PCR tubes
• 1.5 ml Eppendorf DNA LoBind tubes

Equipment

• Centrifuge with capacity for 5 ml and 15 ml tubes, and a swing out and fixed angle rotors
• Microfuge
• Hula mixer, or a gentle rotation mixer
• Magnetic rack for 15 ml tubes
• Magnetic rack for microcentrifuge tubes
• Vortex
• P1000 pipette and tips
• P100 pipette and tips
• P10 pipette and tips
• Thermal cycler
• Ice bucket with ice
• Qubit fluorometer (or equivalent for QC check)

Method

Preparation of plasma from blood

1. Centrifuge ≥3.5 ml of fresh blood (overnight chilled delivery) in the EDTA K2 vacuum tube at 1900 x g for 10 minutes at 4°C in a swing out rotor centrifuge.

2. Pipette and transfer the supernatant (this is the plasma fraction) to a fresh 5 ml DNA LoBind Eppendorf tube.
Note: We recommend a minimum plasma volume of 1 ml is used, although volumes up to 4 ml have been validated.

3. To remove residual cells from the plasma, centrifuge the plasma at 16,000 x g for 10 minutes (or 6,000 x g for 30 minutes depending on the spin capacity of the centrifuge) at 4°C, in a fixed angle rotor.

Note: It is important to remove residual cells from the sample when the blood/plasma is still fresh (from an overnight chilled delivery). Failing to do so will result in increased amounts of gDNA contamination in the sequencing library.

4. Aspirate the supernatant and transfer it to a fresh 15 ml tube.

Purification of cfDNA from 1–5 ml serum or plasma Using the QIAamp MinElute ccfDNA Midi Kit

5. Before starting the extraction:

• Prepare a shaker for microcentrifuge tubes at room temperature for use in step 14.
• Preheat a second shaker at 56°C for use in step 26. (Alternatively, equilibrate the first shaker to 56°C after step 14).
• Resuspend Magnetic Bead Suspension (from the QIAGEN QIAamp MinElute ccfDNA Midi Kit) by pulse-vortexing for 1 min.
  Note: Do not let the suspension settle for more than 2 min before use. Pipette from the centre of the suspension.

6. Prepare the buffers for extraction:

• Add 8 ml isopropanol (100%) to 12 ml Buffer ACB concentrate to obtain 20 ml Buffer ACB. Mix well after adding isopropanol.
• Add 30 ml ethanol (96–100%) to 13 ml Buffer ACW2 concentrate to obtain 43 ml Buffer ACW2. Mix well after adding ethanol.

7. Mix the following components according to the instructions below in a 15 ml tube.

Component	Volume for 1 ml plasma (µl)	Volume for 2 ml plasma (µl)	Volume for 3 ml plasma (µl)	Volume for 4 ml plasma (µl)	Volume for 5 ml plasma (µl)
Plasma	1,000	2,000	3,000	4,000	5,000
Magnetic Bead Suspension	30	60	90	120	150
Proteinase K	55	110	165	220	275
Bead Binding Buffer	150	300	450	600	750
Total volume	1,235	2,470	3,705	4,940	6,175

8. Incubate the reaction for 10 min at room temperature while shaking (at a slow speed) end-over-end.

9. Spin the tube down briefly (30 seconds at 200 x g) to remove any solution in the cap.

10. Place the tube containing bead solution into a magnetic rack for 15 ml tubes. Let the tube stand for at least 1 min, until the solution is clear.

11. Remove and discard supernatant.

12. Remove the tube from the magnetic rack and add 200 µl of Bead Elution Buffer to the bead pellet. Vortex to resuspend beads, and pipette up and down to mix and rinse residual beads from the tube wall.

13. Transfer the full volume of mixture (including the beads) into a Bead Elution Tube.

14. Incubate for 5 min on a shaker for microcentrifuge tubes at room temperature and 300 rpm.
Note: If the same shaker for microcentrifuge tubes is to be used in step 26, remove the tubes after the room temperature incubation and equilibrate the shaker to 56°C

15. Place the Bead Elution Tube containing the bead solution into a magnetic rack for 2 ml tubes. Let the tube stand for at least 1 min, until the solution is clear.

16. Transfer the supernatant into a new Bead Elution tube. Discard the bead pellet.
Note: Avoid transferring any magnetic beads in this step. Carryover may result in reduced cfDNA yield.

17. Add 300 µl Buffer ACB to the Bead Elution tube containing the supernatant, and vortex to mix. Briefly centrifuge the tube to remove drops from inside the lid.

18. Pipet the supernatant–Buffer ACB mixture from the previous step into a QIAamp UCP MinElute column.

19. Centrifuge for 1 min at 6,000 x g.

20. Place the QIAamp UCP MinElute column into a clean 2 ml collection tube, and discard the flow-through.

21. Add 500 µl Buffer ACW2 to the QIAamp UCP MinElute column.

22. Centrifuge for 1 min at 6,000 x g.

23. Place the QIAamp UCP MinElute column into a clean 2 ml collection tube, and discard the flow-through.

24. Centrifuge the QIAamp UCP MinElute column at 20,000 x g for 3 min.

25. Place the QIAamp UCP MinElute column into a new 1.5 ml elution tube and discard the 2 ml collection tube.

26. Open the lid of the tube and incubate the assembly in a shaker for microcentrifuge tubes at 56°C for 3 min to dry the membrane completely.

27. Carefully pipet 23 µl of ultra-clean water into the centre of the membrane. Close the lid and incubate at room temperature for 1 min.

28. Centrifuge at 20,000 x g for 1 min to elute the DNA.

29. To maximise yield from the elution: Place the QIAamp UCP MinElute column in a clean 1.5 ml elution tube. Aspirate the eluate from the previous step and reload it onto the centre of the membrane. Close the lid and incubate 1 min at room temperature.

30. Centrifuge at 20,000 x g for 1 min to elute the DNA.

31. Quantify 1 µl of eluted sample using a Qubit fluorometer.

Note: From 1 ml of plasma, you can expect a yield of ≥ 6 ng cfDNA.

We recommend that the fragment length profiles of extracted cfDNA samples are analysed using a Femto Pulse (Agilent), or equivalent (example in Figure 1).

Library preparation and sequencing

For instructions on library preparation, sequencing and data analysis; proceed with ≥6 ng of recovered cfDNA, per sample, as input to the "DNA repair and end-prep" section of our library preparation protcol:

• Ligation sequencing V14 - Human cfDNA multiplex (SQK-NBD114.24)

This protocol has been developed for optimal performance and sequencing of 12 human cfDNA samples along with this extraction method.

Results

Figure 1. Fragment length profile of extracted cfDNA, run on a Femto Pulse (Agilent). This example shows the characteristic nucleosome peaks with minimal gDNA contamination.

Sequencing performance

Libraries were prepared using the recommended Ligation sequencing V14 - Human cfDNA multiplex (SQK-NBD114.24) protocol.

Figure 2. Total sequence data output from single PromethION flow cells (72 hour runs), for 12x barcoded cfDNA inputs of 6 ng, 4 ng and 2 ng inputs. (QScore ≥10). a) Total sequencing data (Gb), b) Total primary aligned data (Gb) and c) Read count (millions of reads).

Figure 3. Per barcode sequence data output from single PromethION flow cells, using 12x barcoded libraries (72 hour runs) (Q-score ≥10) . a) Raw sequencing data (Gb). b) Primary aligned data (Gb). 6 ng of cfDNA was required to generate >3Gb of aligned data which equates to approximately 1X human genome coverage. c) Read count (million reads).

Figure 4. Read length profiles (Q-score ≥10) of total data output (multiplexed) for a) Sequencing data, b) Aligned data. The reduction in length of aligned data is due to the removal of native barcode and adapter sequence. These profiles are equivalent for each of the 12x demultiplexed samples.

Table 1. Yields (ng) of cfDNA extracted from a range of plasma volumes (1–5 ml) when eluting into a final volume of 20 μl. We show that our recommended ≥6 ng of cfDNA can be achieved using starting plasma volumes ≥1 ml, which can be achieved from starting blood volumes ≥3.5 ml.

Change log

Version	Change
v2, 10th October 2024	Updated extraction method to elute sample in 23 µl (was 50 µl) Secified expected yield to be ≥ 6 ng for 1 ml of plasma.
v1, 15th May 2024	Initial publication