How to Automate cfDNA and cfRNA Extraction

Sample Transfer

It can be a challenge to transfer large volumes of liquid quickly. To improve speed, be sure to use 5ml pipettes (or 1ml pipettes at the least) to transfer liquid from the vacutainer to the tubes or plates. Many labs choose to manually perform these preprocessing steps of centrifuging and pipetting. Plasma to be used for liquid biopsy is generally prepared using the double centrifugation protocol. This ensures a clean capture of the plasma that is free of contaminating cellular DNA that is typically higher molecular weight. While an instrument could be used to prepare the plasma, it requires expensive and complex camera systems to inform the pipette where the layers are. Without a camera system, there is no way to know where the plasma level starts, which could result in cellular contamination. However, if you have enough sample, you could pipette from the top at a fixed volume. For example, if 2ml of plasma is enough, there should be sufficient sample in a 10ml blood collection that a fixed volume far from the border of the layers could be used without disrupting the cellular fraction post-centrifugation.

Digestion and Lysis

The next step where the volume and viscosity of the samples plays a role is in the digestion and lysis step. Although mixing large volumes of a viscous solution can pose some challenges, they can be easily overcome with proper equipment. We provide a large volume Heater Shaker Magnet (HSM) module for this purpose that can be adapted to all major liquid handlers or used separately.

Binding

The binding step is the most universally challenging because the volumes can become quite large. For example, 3ml of plasma will typically require 1–3 volumes of binding buffer (depending on the supplier and kit), leading to a total volume of 6–12ml that requires mixing. This step requires extra caution as insufficient mixing is one of the leading causes of low yield. Visually inspect the sample to ensure the beads are fully suspended during the binding and that a vortex fully forms, suggesting thorough mixing. If these visual parameters are met, you can be confident that binding is occurring with sufficient time. Confirmation can be made by further processing the sample manually to avoid the risk of other steps influencing the result. For more information on experimental design and troubleshooting, download the "Automating Nucleic Acid Extraction — A How-to Experimental Guide".

There are a variety of methods to complete the binding step: 

• Using a Promega Heat Shaker Magnet (HSM) or rotisserie (most efficient)
• Sequential binding using a KingFisher™ Presto Purification System (Thermo Fisher Scientific), a particle mover that can be integrated onto most liquid Handlers (e.g., Hamilton or Tecan workstation)
• Sequential binding on a liquid handler (least efficient)

The most efficient method is to use the HSM and 5ml pipettes, or the Presto System. If using the HSM, it can be helpful to add the magnetic particles while the liquid is shaking to keep them suspended. Some particles settle rapidly and can be difficult to resuspend in viscous solutions full of proteins that could cause bead aggregation. When using the HSM approach, the binding itself is very efficient, but the liquid removal step and bead transfer is less efficient compared to the Presto System. The Presto System supports a lower maximum volume and therefore requires sequential binding, but it gains speed during the transfer steps. Sequential binding on a liquid handler is the least effective. Unless you are working with low volumes, the number of sequential binding steps makes it extremely slow.

Following this binding (mixing) step, the samples will need to be transferred from large tubes to a more convenient format, such as a 96 well plate. This requires magnetizing the beads and removing the solution, followed by resuspension of the particles and then transferring. This tends to be a very slow process as the instrument will need to aspirate multiple times per tube before transferring the beads.

Washing

During the wash steps, magnetic particles can clump due to the quantity of protein being inadvertently concentrated from the large sample volume. This is especially true during the first wash. The best way to track your progress is to visually verify that the clump of particles is disrupted, otherwise your eluates will be impure and possibly inhibited downstream.

To learn more about overcoming this problem, read the "Automating Nucleic Acid Extraction" guide, which walks through the protocol for automating nucleic acid extraction. You will find a detailed guide including video instruction, experimental design and troubleshooting tips to support your method development.

Preserving Samples

Strict control over the quality of the initial sample is necessary because the way plasma is prepared and shipped strongly affects the output. Processed samples collected in EDTA tubes should be kept on ice and processed within 1–2 hours for optimal results. Otherwise, cell-free DNA or RNA blood collection tubes (e.g., Streck) are preferred. Cells that are not stabilized such as those in the EDTA tube will begin to slowly lyse, leading to contamination of cell-free fraction with WBC (white blood cell) DNA. When looking for a rare target cfDNA, this is equivalent to dumping a few hundred more bales of hay on top of the needle in the haystack. RBC (red blood cells) are even more prone to lysis and need to be handled with extreme caution, especially for RNA analysis. Proper preservation is critical to achieving consistent results. Learn more about sample preservation in this webinar: Detection of Circulating Tumor DNA (ctDNA) in Blood: Standardization of Pre- and Post-Analytical Conditions.

Centrifugation

Prior to purification, a double centrifugation procedure is recommended to separate the cellular fraction from the plasma. Typically, a lower-speed centrifugation is performed at 1,000–1,900×g for 10 to 15 minutes at temperatures ranging from 4°C to 25°C . Then, the plasma is carefully removed and a second centrifugation is performed, often at much higher speeds (i.e., 16,000×g). Some protocols advise higher speeds for this step while others repeat the 1,500–1,900×g centrifugation for another 10–15 minutes. Some methods even recommend a third centrifugation step. While most labs have settled on the notion that double centrifugation is preferred, the ideal method was up for debate.

In 2020, Greytak et al. did a deep review on the preanalytical conditions to harmonize and improve consistency across labs. Here is a summary of their advised method:

Collection Tube

• EDTA tubes when immediate processing is possible
• cfDNA stabilizing tubes when processing delays are unavoidable

Pre-Centrifugation Processing Delay

• ≤ 2 hours at room temperature or on ice (EDTA tube)
• ≤ 3 days at room temperature (stabilizing tube)

Tube Agitation

• Minimize (after initial inversions)

Centrifugation

• Two centrifugation steps
  • 800–1,600×g for 20 minutes
  • 14,000–16,000×g for 10–20 minutes

Plasma Storage

• ≤ –80°C, with ≤1 freeze-thaw event

cfDNA Storage

• ≤ –20°C, with ≤1 freeze-thaw event

Quantification

• Real-time or digital PCR

In the above protocol, real-time or digital PCR is recommended for quantitation. This is because the fluorescent dye measurements tend to overestimate the useable nucleic acids. In addition, the detection limit is around 0.01ng/µl, and cannot provide information on the amount of genomic DNA contamination. The TapeStation System (Agilent) can be used here to assess the fragment sizes present; however, it is only semi-quantitative and cannot assess amplifiability. See below to learn about the pros and cons of various methods for quantifying cfDNA, and how the ProNex® DNA QC Assay can be used to quantify amplifiable nucleic acids down to 3.2pg/µl and assess genomics DNA contamination.

Benefits of the ProNex® DNA QC Assay

Not all QC analysis is the same. Only the ProNex® DNA QC Assay includes everything you need, with minimal sample usage.