Josh P. Roberts has an M.A. in the history and philosophy of science, and he also went through the Ph.D. program in molecular, cellular, developmental biology, and genetics at the University of Minnesota, with dissertation research in ocular immunology.
The cliché “garbage-in, garbage out” is as true in research as anywhere else. With the explosion of next-generation sequencing (NGS), having the appropriate tools and methodology to extract high-quality, reasonable quantities of nucleic acids from precious or technically challenging samples such as blood, liquid biopsies and formaldehyde-fixed paraffin-embedded (FFPE) tissue is critical. This article focuses on kits, reagents and methodologies that can enable researchers to gather higher-quality nucleic acids for further analysis.
Where from?
As a rule of thumb, it’s best to start with really good sample material. But how researchers define “good” varies. Extracting NGS-quality nucleic acids from cell culture is generally not a problem; however, blood, fresh or archived tissue, soil, swabs or otherwise nonoptimal sources can present challenges.
Unless the samples are to be processed right away, the first step should be to preserve them so parameters such as osmolality, temperature, oxygen concentration and nutrition have minimal effects on the sample. RNA transcript levels can be affected by the presence of nucleases, for example, as well as by the induction of stress-inducing genes. There are numerous products now on the market—such as Whatman (part of GE Healthcare Lifesciences) FTA cards, RNAlater (from Ambion, now part of Thermo Fisher Scientific), QIAGEN’s Allprotect Tissue Reagent, Zymo Research’s DNA/RNA Shield™ and several PAXgene products from PreAnalytiX (a joint venture between QIAGEN and Becton, Dickinson and Company (BD))—that stabilize blood or tissue at room temperature for storage and transport. Although most of these require reagent removal prior to processing, “there is no reagent removal with our product,” says Ryan Kemp, director of nucleic acid solutions at Zymo Research.
Yet often the researcher isn’t given such options, and samples are already in the form of the molecular biologist’s bane: FFPE tissue. This format is just what the pathologist ordered to keep morphology intact, so it has historically been the preservation method of choice. But the process leaves nucleic acids crosslinked and degraded. Numerous protocols and kits exist to de-paraffinize FFPE tissue and remove the crosslinkage, after which the researcher must make the best of a fragmented genome or transcriptome.
Another problem plaguing FFPE samples is deamination. When deaminated DNA is “put through an NGS workflow, you will read a different sequence, as a pure storage artifact,” remarks Markus Sprenger-Haussels, senior director of sample technologies development at QIAGEN. The company’s GeneRead DNA FFPE Kit, “incorporates a step during extraction that cuts out deaminated nucleotides, so that … these sequencing artifacts are reduced by a factor of about twenty.”
Cell-free
There is much interest of late in capturing, quantifying and sequencing circulating cell-free DNA (cfDNA or ccfDNA), and to a lesser extent RNA (cfRNA), as these molecules provide researchers with new insights on cellular regulatory events and can uncover potential therapeutic biomarkers.
There is much interest of late in capturing, quantifying and sequencing circulating cell-free DNA (cfDNA or ccfDNA), and to a lesser extent RNA (cfRNA), as these molecules provide researchers with new insights on cellular regulatory events and can uncover potential therapeutic biomarkers.
These nucleic acid fragments are released by dying cells (among other sources) and can be found in serum, plasma, urine, mucus, amniotic fluid, cerebrospinal fluid, saliva, semen and the like.
There are several challenges to extracting these macromolecules from fluids that are not seen when dealing with cellular DNA and RNA.
“There is a lower concentration of nucleic acids because there are no cells, no nuclei full of chromosomes to give you DNA,” points out Marianna Goldrick, senior scientist at Bioo Scientific. For example, a typical plasma sample—especially from a healthy individual—may yield a concentration of DNA too dilute to be reliably detected using standard methods (according to Bioo’s website).
“It’s also fragmented: The typical size recovered from cell-free biological fluids is about 170 base pairs (bp), reflecting the size protected by histone protein complexes,” Goldrick adds. Multiples of 170 bp are also observed in lesser amounts.
A common problem encountered when working with cell-free nucleic acids is contamination by cellular nucleic acid, especially in clinical fluids that are not stabilized upon collection. “The problem is that once you collect the sample, DNA fragments are metabolized over time and are eaten up by the immune system, while at the same time new DNA fragments are released,” diluting the signal, says Sprenger-Haussels. “This is what we have addressed with the PAXgene ccfDNA system, where we stop apoptosis, we stop DNA degradation and release of new DNA.”
Kits are available that give researchers the tools to extract their choice of nucleic acid – or combination of nucleic acids -- from their choice of sample, in their choice of format. And since some vendors offer several kits with overlapping capabilities, “you can’t just give the vendor’s name and know specifically what kit” is being referred to, but need the kit name or part number as well, Goldrick points out.
Challenging samples
Not all samples are from human cell culture or bodily fluids, and some may require different procedures.
Many kits and protocols rely on silica-based columns or beads, eschewing toxic phenol/chloroform-based extraction methods. Yet Shahrouz Ghafoory, now a post-doc at the Oklahoma Medical Research Foundation, found that collagen and other debris from fibrotic liver homogenates clogged up the silica-based columns he tested, and that silica-based magnetic beads required too much precious sample. In contrast, the Lexogen Split RNA Extraction Kit “was really easy, and I got a really high quality of RNA,” he says. The kit, explains Lukas Paul, head of services at Lexogen, relies on an organic phase extraction before running the sample through a silica column, because “it’s very efficient in extracting all the contaminants.”
Samples such as feces and soil are rich in humic acids, polyphenols and other DNA polymerase inhibitors, which can co-purify with standard extraction techniques, points out Kemp. Zymo offers a “very simple column to remove those very efficiently and easily.” Other companies offer other solutions, for example the patented Inhibitor Removal Technology® from MO BIO Laboratories’ (now part of QIAGEN) that selectively precipitates a laundry list of inhibitors. “The art of nucleic acid purification is to develop lysis, binding and washing procedures which effectively differentiate between wanted nucleic acids and unwanted contaminations or impurities,” says Sprenger-Haussels.
Homogenization, whether manual or chemical, is a crucial step in nucleic acid extraction. “It has to be quick so you don’t have degradation, and it has to be really efficient and complete so you get really good RNA extraction efficiency,” says Paul. “And sometimes it requires special treatment, like incubation with enzymes to disrupt yeast cell walls or remove polysaccharides from plant tissues. Upstream treatments and organic phase extraction steps address many but not all of these issues. … No one kit fits all.”
Companies like QIAGEN and Norgen, for example, offer a range of kits for different sample types. And sometimes it’s a matter of adjusting a protocol to fit one’s needs. Kuba Kwintkiewicz, lab manager of the University of North Carolina School of Medicine’s Microbiome Core Facility, encounters all sorts of sample sources, from oceanic ice to vaginal swabs to fermenting tobacco. “For most of my work, I use buffers that are parts of QIAGEN kits, but I really do not use kits as I often have to modify things for my own purposes,” he says.
DNA? RNA? What size?
In some cases, kits and protocols can be adjusted to yield DNA, RNA or both, and in some cases even protein. In addition, there are options that enrich for specific populations—for example: small RNA, including microRNA; large RNA, such as messenger RNA; or total RNA. “They really get all these ‘omes from the same sample, from the same cells,” notes Paul.
It’s important to be aware of biases that may be associated with different protocols, as well. This may be intentional—to recover the smaller, fragmented RNA from FFPE tissue, for example, or to recover cfDNA fragments in preference to contaminating genomic DNA (gDNA). Or it may be an unintended by-product of, for example, purification methods utilizing organic separation, which Kemp says “strongly biases results away from proper miRNA recovery.”
The good news is, with all these choices, there’s likely a kit or protocol that meets your nucleic acid extraction needs.
Image: Shutterstock Images