Even use-as-directed assays may require some adaptation based on a host of different considerations. What type and how many samples are to be run, and from what matrix?
How precious are the samples? Is throughput as important as accuracy and reproducibility? How many parameters need to be assayed, and will it be done in monoplex or multiplex? How sensitive and selective does the assay have to be, and over what dynamic range? Will it give consistent results across lots—and can you prove it?
Scientists gathered recently at CHI’s “Biomarker World Congress” to share their insights about developing assays to measure DNA, protein, and even RNA.
There are well-established, reliable methods to visualize where specific proteins and DNA sequences lie in a tissue or in a cell: to wit, respectively, immunohistochemistry (IHC) and fluorescence in situ hybridization (FISH). “Yet there’s a lack of technology to look at RNA in situ,” commented Yuling Luo, Ph.D., founder, president, and CEO of Advanced Cell Diagnostics (ACD). “That is the space that our RNAScope technology fills.”
RNA in situ hybridization (ISH) has undergone incremental improvements over the past 40 years, but not enough to endow it with the sensitivity, specificity, robustness, or simplicity to be routinely used for biomarker analysis and diagnostic applications, Dr. Luo noted. Boosting signal at the same time boosted the background, he said, yielding only limited improvement in signal:noise ratio.
ACD designed a system “that allows selective amplification of target-specific signal without amplifying nonspecific hybridization signal,” Dr. Luo explained. Two independent oligonucleotide probes simultaneously bind adjacent RNAs targets, and are recognized by a single specific “pre-amplification” molecule.
This, in turn, is bound by up to 20 amplifier molecules, each having 20 binding sites for the label probe—making a “Christmas tree-like hybridization structure”. And since a 1 kb stretch is typically targeted by 20 probe pairs, an RNA molecule can be decorated with 8,000 labels.
All the RNA resides in the cytoplasm and the assay conditions are all the same, so RNAScope can easily be multiplexed, Dr. Luo pointed out. By contrast, “in immunofluorescence analysis, because there are membrane proteins and nuclear proteins, the assay conditions are different—you have to get different antibodies in different locations to work together, which is much harder to do.”
RT-PCR, on the other hand, can assay single RNA molecules but fails to deliver tissue context. And this would be a problem when searching for loss of expression as a tumor biomarker, for example, since “grind-and-bind” methods such as RT-PCR cannot differentiate whether the expression is found in tumor or stroma.
For protein biomarkers—when context is not an issue—the immunoassay “can be the most specific assay you’re going to get,” said Lynn Zieske, Ph.D., vp for commercial solutions at Singulex.
In a typical sandwich ELISA, antibody that has randomly adhered to a plastic plate captures target from a sample of interest. Another antibody, to a different epitope on the same target, then either directly labels the target or itself becomes the target for a direct or indirect label.
The Erenna® immunoassay system offers resolution at subpicogram concentrations—up to two orders of magnitude greater than a standard ELISA, claimed Dr. Zieske. First, antibodies are coated onto paramagnetic particles in a way that orients them for maximum exposure to binding and capture of antigen.
Using particles in suspension “allows us a lot more flexibility in being able to not worry about nonspecific binding—because most nonspecific binding occurs on the plastics of wells.” The antigen is then translated into a signal using a fluorescently conjugated detection antibody.
Once the sandwich is formed and washed, the detection antibody is dissociated so that only the detection antibody, antigen, and elution buffer remain as a solution to be read in the Erenna immunoassay reader, which is in concept similar to a capillary flow cytometer.
A laser focuses on the sample within a 100 µm capillary tube, and a single molecule at a time is counted. “All we’re looking for is literally the fluorescence intensity of a molecule,” he said. An algorithm back-calculates the concentration of target in the original sample.
“We’re focused on very high precision and high sensitivity, though we have a broad dynamic range for targets not requiring exquisite sensitivity,” Dr. Zieske said. Users can run the equivalent of four 96-well assays at a time, without the cross-talk inherent in multiplex assays. He suggests we “think of it as being a multiple monoplex,” giving the same amount of information while potentially using less precious sample than a multiplex ELISA.
Single System, Multiple Assays
High-density microarrays can whittle 20,000 targets, found from assaying just a few samples, down to a few dozen or hundreds that seem to be telling a story. After that is “the point when you really want to hone in on what markers are important for that particular disease or condition that’s being studied,” said Sherry Dunbar, Ph.D., director of scientific marketing for Luminex. “We think of ourselves as the step after high-throughput screening.”
She touted the ability of the company’s eponymous barcoded bead-based platforms to quickly, and at low cost, multiplex multiple tests: “You can run—depending on which of our analyzers you’re using—a 96-well plate in 17 minutes to an hour, and you can have a multiplexed result, up to 500-plex, on 100 samples in an hour or less. So that’s a lot of data you can get in an hour.”
It’s difficult to get a good picture of what’s happening immunologically by looking at just one cytokine. In a study on ischemic brain injury, researchers used Luminex technology to simultaneously examine the expression of 30 or so cytokines from samples of just 50–200 laser microdissected brain cells. “There was really no other way to do a very extensive analysis on that,” Dr. Dunbar said.