to BioTechniques free email alert service to receive content updates.
Digital PCR: Separating from the Pack

03/05/2014
Nathan S. Blow, PhD

Could the next revolution in PCR be digital? Nathan Blow takes a look at the history of digital PCR and why the methodology might have finally reached a tipping point in development.


When Bio-Rad reported their fourth quarter earnings in late February, the company said that sales of their QX200 Droplet digital PCR system were one of the key drivers of the 5% revenue increase for the period. In the earnings call, Christine Tsingos, Chief Financial Officer and Executive Vice President at Bio-Rad, noted continued strong sales growth associated with the droplet digital PCR products, with more than 500 units now in operation.

QX200 Droplet Digital PCR System, Source Bio-rad.com

Clearly digital PCR has come a long way in recent years, thanks in large measure to the development of commercial systems like the QX200. These technology advances seem to indicate a tipping point where a greater number of researchers will soon have access to the technology, which will spur development of new applications that take advantage of the full capabilities of digital PCR and move scientists towards more robust biomarker studies and even single cell analyses.

A long time in development

What makes digital PCR different from traditional PCR is the endpoint of the assay—digital PCR directly quantitates the number of PCR targets in a sample by counting rather than by amplifying those targets in aggregate as done in traditional PCR or generating a semi-quantitative measure of target abundance as in real-time PCR.

Early attempts to implement digital analysis of targets using PCR were described in 1992 in BioTechniques. Here, the authors applied the process of limiting dilution of a sample to obtain only a single template molecule in each well. By counting the number of wells with amplification signals following PCR, the researchers could determine more accurately the number of starting molecules. The key to the technique was separating templates such that each well held only one or zero templates.

Following those initial reports in 1992, interest in digital PCR quickly took off as researchers realized the potential for quantitation. In 1999, Kinzler and Vogelstein first used the term “digital PCR” in an article in the journal Proceedings of the National Academy of Science. This article was also one of the first to apply digital PCR to clinical samples, looking at mutations in the ras oncogene from patients with colorectal cancer. Five years later in 2003, Vogelstein and Kinzler, along with other colleagues, reported on a technique that they called BEAMing, which employed beads, emulsion, amplification, and magnetics to provide another quantitative method for the analysis of specific target sequences in complex biological samples. BEAMing would later be applied to the development of next-generation sequencing technologies, but its development highlighted a need for new ways to separate single molecules.

Although advances in digital PCR continued among academic labs, commercialization was a much slower process, with a digital PCR platform not arriving until 2006. This lag in time begs the question: Given the possibility and potential of digital PCR, why did it take so long for the technology to catch on with developers and researchers? The challenge can be traced back to that first paper in 1992 and the 2003 BEAMing article—finding robust ways to separate single molecules from complex samples.

A question of partitions

Stephen Quake, Source thebigone.stanford.edu

Limiting dilution is a sensible approach for partitioning a single template to each reaction and is quite effective on a small scale. But it can also be error-prone when done by hand. The first solution to this dilemma appeared in the form of small channels and miniaturized plumbing.

Microfluidics and lab-on-a-chip designs first appeared in the early 2000s but were not applied to partitioning until the mid-2000s. Fluidigm, a company founded by Stephen Quake from Stanford University, was the first to develop a microfluidic-based platform capable of partitioning DNA templates for amplification and thus provide a digital readout. Today, the company has both a digital PCR array and a quantitative real-time digital PCR fluidic chip that can process up to 48 samples with nearly 37,000 reaction chambers.

On the heels of microfluidic platforms, another partitioning approach was advanced; instead of fabricated wells and channels, this approach used a large number of small droplets generated in a liquid.

Droplet digital PCR, as it is now known, uses individual emulsion droplets, each acting as their own reaction vesicles (like a single well on a microplate). The scalability of these droplet-based systems allow tens or even hundreds of thousands of droplets to be created for reaction analysis, making droplet digital PCR a particularly powerful option adopted for instruments by RainDance Technologies and the QX200 system from Bio-Rad.

RainDrop Digital PCR System, Source Raindancetech.com

More and more droplets and channels

For template quantification, digital PCR is not the only game in town—real-time PCR (a so called analog method) also enables relative quantification of target templates in a sample. But when it comes to quantification, how do the two methods compare?

In 2013, researchers from the Fred Hutchinson Cancer Research Center in Seattle, Washington compared absolute quantification by droplet digital PCR and real-time PCR and reported their findings in the journal Nature Methods. The authors found that when quantifying microRNAs, droplet digital PCR had greater precision and comparable sensitivity. The surprising finding was that digital PCR improved day-to-day reproducibility by up to 7-fold when compared with real-time PCR. Robust reproducibility means that studies done over a period of days can be directly compared— something that is often a challenge for labs using real-time PCR.

The upshot of all these findings is that when it comes to clinical diagnostic applications and biomarker analysis, digital PCR might be a strong option—even opening the door to the use of circulating DNA or miRNA as biomarkers, given the sensitivity of the technique.

While still in the early stages of adoption, both digital PCR and droplet digital PCR are showing strong acceptance. In 2013 and so far in 2014, more than 40 articles have been published describing droplet digital PCR usage or developments, according to PubMed. This is a sharp increase from prior years. And with applications ranging from biomarker analysis to detection of waterborne viruses, it is clear that in the years to come, digital PCR will grow towards realizing the potential first described over 20 years ago.

References

Sykes P. J.; Neoh S. H.; Brisco M. J.; Hughes E.; Condon J.; Morley A. A. Quantitation of targets for PCR by use of limiting dilution. Biotechniques 1992, 13, 444–449.

Vogelstein B, Kinzler KW. Digital PCR. Proc Natl Acad Sci U S A. 1999 Aug 3;96(16):9236-41.

Dressman D, Yan H, Traverso G, Kinzler KW, Vogelstein B. Transforming single DNA molecules into fluorescent magnetic particles for detection and enumeration of genetic variations. Proc Natl Acad Sci U S A. 2003 Jul 22;100(15):8817-22.

Hindson CM, Chevillet JR, Briggs HA, Gallichotte EN, Ruf IK, Hindson BJ, Vessella RL, Tewari M. Absolute quantification by droplet digital PCR versus analog real-time PCR. Nat Methods. 2013 Oct;10(10):1003-5.