Overcoming Challenges in Plasma Sample Prep
The product is only as good as the process
Fans of old war movies will recall a scene in which a medic hovers over a wounded patient and exclaims, “We need two units of plasma!” As a material used ubiquitously in research and to treat loss of blood due to trauma or illness, plasma holds a unique position among significant bodily fluids.
Plasma, the clear-yellowish blood component comprised principally of water, contains dozens of compounds of biological interest, including complexed and unbound minerals, proteins (mainly albumin), and genes. Interest in plasma is high these days, thanks to the emergence of cell-free DNA cancer testing, which is accessible through plasma.
Plasma preparation typically involves collection and anti-coagulant treatment, low-temperature centrifugation, and storage at -80 °C “but never at -20 °C because, despite appearances, plasma does not freeze at this temperature,” says Chad R. Borges, PhD, associate professor at the School of Molecular Sciences, Arizona State University (Tempe, AZ). “If samples are shipped on dry ice, the headspace should be cleared of CO2 before thawing to prevent acidification due to CO2 dissolving in the plasma.”
While established protocol dictates deep-freezing plasma to preserve its biochemical features, samples may be thawed and re-frozen multiple times between collection and use. Accidental thawing is also a potential issue, particularly when samples are stored in group refrigerators. Since these incidents are poorly documented, there is need for an objective measure of how much time a plasma sample has spent above its optimal storage temperature.
Borges has devised such a method, a dilute-and-shoot, intact-protein mass spectrometric assay. The test measures the cumulative exposure of archived plasma (or serum) samples to thawed conditions based on the relative abundance of S-cysteinylated (oxidized) albumin—higher levels mean longer exposure. “Our delta-S-Cys-Albumin marker places a timestamp on samples for how long they’ve been exposed to the equivalent of room temperature,” Borges explains. “Then, if the stability of a particular target analyte in plasma (say, a particular miRNA) is known, one can determine the suitability of the sample for analysis of that particular analyte.”
Determining the sample’s suitability for genetic analysis assumes that a stability study for that species was previously undertaken. Plugging the exposure time value from the delta-S assay into the stability curve could, theoretically, provide a more-or-less exact value for levels of the intact gene of interest. This is often overkill, though, as scientists are usually only interested in “go/no-go”: Is the sample suitable for genetic analysis or not? A simple time cutoff value for accepting or rejecting a plasma sample will usually suffice.
Plasma provides researchers and clinicians with a cornucopia of interesting genetic material. The bad news is that biological samples, particularly liquids, are notoriously unstable. Cells present in plasma upon collection rapidly die and genetic material degrades or becomes inactivated after getting stuck to other rapidly degrading components.
Quality and reproducibility
- In addition to issues related to plasma’s complexity and fragility, there is the potential for error after the collection step, for example:
- Using unsuitable collection tubes, which can lead to perturbed expression profiles
- Non-optimal centrifugation protocols
- Inappropriate sample handling, transport, and storage after collection and before plasma preparation, leading to hemolysis that perturbs microRNA analysis, or resulting in cellular breakdown that interferes with analysis of cell-free DNA
These quality and reproducibility issues, both in plasma preparation and downstream analyses, are under study by two European consortia, CANCER-ID and SPIDIA4P. CANCERID recently published a paper addressing the high variability of circulating tumor DNA (ct-DNA) fractions. Using automated methods, researchers comparing commercial ct-DNA extraction kits found that the kits varied significantly in their preference for a spiked mononucleosomal DNA or endogenous cell-free DNA. In a previous article the consortium used quantitative real-time PCR and next-generation sequencing to compare five extraction protocols for cell-free microRNA. Extraction efficiency was compared by measuring levels of spiked microRNAs. They found that column-based extraction methods were highly effective, whereas protocols employing phenol extraction followed by column purification resulted in lower yield and quality. They concluded that, “The choice of plasma miRNA extraction methods affects the selection of potential miRNA marker candidates and mechanistic interpretation of results, which should be done with caution, particularly across studies using different protocols.”
“Spiking is a common technique for generating concentration curves and reference standards, and is particularly useful to assess interference in molecular analyses of plasma samples,” says Dr. Mikael Kubista, founder of the TATAA Biocenter (Gothenburg, Sweden). RNA and DNA spikes are available from many providers, TATAA included.
TATAA uses a variety of instrument platforms to prepare plasma samples for genomics studies. Their preferred system for microRNA is the EdgeSeq instrument from HTG Molecular, which eliminates several hands-on steps and features an intuitive graphical user interface. “But for targeted analysis, we use Two-Tailed qPCR, a technology invented at TATAA and commercialized by Czech-headquartered BioVendor,” Kubista says.
“Two-tailed” refers to the sequences that overcome size differences between conventional primers and targets, where two primers are too long and one primer shows insufficient selectivity. Two-tailed PCR uses two “hemiprobes”—half-length sequences connected by a chemical tether—that bind to distinct sequences on the target microRNA. BioVendor claims sensitivity down to 10 molecules, up to a nine-log concentration dynamic range, and microRNA detection from serum, blood, and solid tissues in addition to plasma.
TATAA employs a different technique, sensitive mutation detection using sequencing (SiMSen-seq), a technique that is barely three years old. With a focus on rare variant alleles from complex matrices, SiMSen-seq generates targeted barcoded libraries from very low DNA inputs, with flexible target selection and a short (four-hour) library construction protocol.
For analysis of circulating tumor cells, TATAA scientists turn to their own product, the GrandPerformance assays, which are qPCR assays based on pre-amplification, and which quantify 96 specific expression markers per cell. In addition to covering cell analysis, these assays are suitable for qPCR, next-generation sequencing, and quality assessment.
Kubista notes the many sources of variability and uncertainty in plasma genomic preps. “Every step in the workflow, from sampling, extraction, and handling to analytics, introduces method- and kit-specific bias and confounding variation. Every workflow should therefore be carefully optimized and validated.” Commercial manufacturers perform kit-level validation but end-users must establish and optimize their own protocols.