NEW YORK (GenomeWeb) — Researchers from Virginia Commonwealth University have developed a method of quantifying multiple nucleic acid targets in a single sample by combining low-cycle-number, target-specific PCR with atomic force microscopy.
The technique, according to its developers, has the potential to absolutely quantify target nucleic acids much in the same way that digital PCR does, but without the use of relatively expensive microfluidic platforms. Instead, the method makes use of standard laboratory equipment such as thermal cyclers and AFMs.
However, the researchers also believe that their method could be adapted to a dedicated laboratory instrument combining thermal cycling, AFM imaging, and data analysis as AFMs become cheaper and faster, Jason Reed, an assistant professor of physics at VCU and corresponding author of a recent Analytical Chemistry paper describing the technique, told PCR Insider.
"AFMs are actually very common in universities … [but] a dedicated instrument could be put together much more cheaply, and no one offers that commercially at the moment," Reed said. "But it could be a box that would be very similar to an [Agilent] Bioanalyzer-type device — a benchtop box that does its thing and you get the data out, because it doesn't need to have all these features that a research instrument has."
Reed and colleagues originally developed their method for a specific application — detecting isoforms of closely related splice variants. They specifically wanted a technique that could detect multiple nucleic acid targets in a single reaction — a common goal for many groups developing gene expression or molecular diagnostic assays.
Although RT-qPCR is the current gold standard for gene expression assays due to its high degree of sensitivity, it is difficult to multiplex for a number of reasons, including non-uniform amplification of lower-abundance targets, off-target binding, and primer-dimer formation — all problems that are enhanced by a higher number of amplification cycles.
The method developed by the VCU researchers achieves multiplexing by amplifying several specific target nucleic acids at once using a relatively low number of cycles — approximately 15, as opposed to the usual 30 or more in standard qPCR — which helps to alleviate differential amplification. "When we pick which genes or targets we're interested in, we design the primers so that the amplicons have a specific size," Reed said.
Then, the researchers take the multiplex PCR reaction droplet and put it on a piece of mica, and image all of the molecules present, Reed explained. "Each of the amplicons has a distinct size, in base pairs or nanometers, and the precision with which we can determine that size is very high with an AFM," he said. "That allows us to tell which species we're looking at, and we simply count."
The method has orders of magnitude higher sensitivity than bulk fluorescent techniques, according to the researchers, and is easier and cheaper to implement because it doesn't use any fluorescent dyes or other types of labels.
"In normal qPCR you are quantitating some sort of fluorescence curve," Reed said. "In this case, we're only doing a small number of PCR cycles. The reason for that is to make the multiplex PCR reaction behave well. And the reason that we can get away with that low number of cycles is that we use single-molecule detection with the AFM."
In fact, the researchers pointed out, the number of PCR cycles should be optimized for each assay conducted, as too few may result in decreased specificity and an insufficient amount of amplicons, while too many may distort the initial distribution of nucleic acid targets.
In the Analytical Chemistry study, which was published earlier this month, the researchers demonstrated their technique by measuring the relative expression levels of 10 human genes in two different total RNA samples, finding a high concordance between data generated by the single-reaction multiplex PCR/AFM and by 20 independent single-plex qPCR assays.
In general, the researchers noted, the technique could prove especially useful to quantify multiple nucleic acid targets in situations where molecular concentration is relevant, like studies of genomic copy number variation, mRNA isoform detection, and chromosomal translocation analysis.
In this regard, the new method shares much in common with digital PCR, another alternative to qPCR that has proven superior at absolute quantitation and is being increasingly adopted for many of these same applications.
"It's very similar in many ways, except we don't have to deal with a microfluidic device," Reed said. "We have basically an infinite number of wells and we can determine how many molecules we want to count. It's very easy for us to count 1,000 or 10,000, depending upon how much sampling depth we want. It's also very easy to multiplex in the same assay. People have shown that they can do some multiplexing with digital PCR, but it's … not a simple process, whereas this is pretty robust."
Current drawbacks to the method include the fact that AFMs, while typically ubiquitous in university core laboratories, is still regarded as a relatively expensive and complex instrument to operate. In addition, the general-purpose AFM used by Reed and colleagues takes approximately 25 hours to scan just two samples.
These AFM attributes are changing, however, as some of the most well-known commercially available platforms from vendors such as Bruker and Asylum Research boast scan rates that could theoretically reduce the assay time to less than 1.5 hours — similar to the time it takes to run 30-plus qPCR cycles and much faster than hybridization-based detection schemes.
And although most general-purpose AFMs would be prohibitively expensive to dedicate to an application such as this, Reed noted that such a platform could be stripped down to its basic capabilities. In fact, the researchers noted, a suitable system for this application could be assembled from commercially available components for about $30,000.
The VCU group has had several discussions with external parties who are potentially interested in commercializing the method, but no concrete progress has been made on this front yet. "They want to see the technology mature to the point where it's something that's kind of a benchtop or desktop box," Reed noted.