Trillions of microbes wriggle around in our guts. They help us break down our food with metabolic activity so high that they are almost like another organ. But even though they are an integral part of our digestive process, scientists still don’t quite understand these microbes. It’s not that they haven’t tried, but rather scientists haven’t had the tools needed to unravel the genomes of these bacteria.
Likewise, whether researchers are looking at cancer-causing mutations or DNA recombination in individual sex cells, scientists who study populations of organisms struggle to piece together fragments of DNA from the same cell and to get enough, accurate copies of the DNA to analyze and sequence reliably. Recognizing these challenges, Guo and other lab researchers, as well as commercial companies, have been looking for an answer.
One possible solution has been microdroplet PCR, a physics-inspired process where DNA from an individual cell is captured in a small oil-emulsified droplet and then merged with only one other droplet containing the correct primer pair to target the portion of the genetic code that a scientist wants to sequence.
From Hundreds to Billions
Other methods, such as micromanipulation and flow cytometry, can also separate cells from one another. “However, this conventional approach is only practical for up to a few hundred cells. It would take liters of reagents and years to separate one billion cells this way,” says Guo. The microdroplet PCR techniques developed by Weitz’s group can compartmentalize one billion cells individually into droplets using about 1 mL of reagents in about 10 hours, and smaller, nanoliter-scale microfluidic chambers can efficiently compartmentalize a few thousand cells on a single palm-sized chip (1).
In the microdroplet PCR process, scientists still design and synthesize forward and reverse primer pairs to target their specific sequences of interest. But instead of mixing and amplifying all the primer pairs and genetic material together, the primers and DNA are prepared separately. First, scientists mix a single pair of PCR primers with a drop of oil and a polymer. To prep the DNA, the scientists still have to fragment and purify it. Then, it's mixed with all of the other PCR reaction reagents, except for the PCR primers, and made into a separate droplet.
Next, scientists send one primer-pair droplet down one lane of a microfluidic chip and a droplet of the DNA mixture down a second channel at the same time. The drops meet and flow through an electric field, which forces the two droplets to merge. Each PCR droplet is then collected into a single PCR tube and the entire library is processed in a standard thermal cycler for targeted amplification. After thermocycling, the PCR droplet emulsion is cracked like an egg to release the amplified DNA into solution for genomic DNA removal, purification, and sequencing (2).
In 2004, Weitz co-founded the company RainDance Technologies, Inc., along with two European collaborators and one of the postdocs from his lab to commercialize the high-throughput microdroplet PCR technology developed in his lab. In 2010, the company marketed a system that could process a million PCR reactions at the same time, in a single day. Now, the company has turbo-charged the system to generate more than a billion reactions in that same amount of time, while providing a digital readout of the PCR products. According to the company, the system promises to speed scientists’ detection of uncommon gene variants that can cause tumors.
For example, in a 2011 paper published in Lab on a Chip, French scientists sought out to identify mutations within the DNA of tumor cells (3). If identified, these mutation sequences could act as biomarkers to distinguish cancer cells from healthy ones. Because previous qualitative techniques lacked the required sensitivity, the researchers turned to RainDance’s digital microdroplet PCR system.
As a result, the team detected a single mutated copy of the cell-division-regulating gene KRAS from 200,000 healthy copies. The technique also let the scientists screen six common mutations in a single codon of KRAS at the same time in the same experiment.
Cheaper and Easier
For the past three years, Olivier Harismendy, a microbiologist and translational genomicist at the Moores Cancer Center of the University of California, San Diego, and his colleagues have been using microdroplet PCR to identify non-inherited tumor mutations as biomarkers. Their hope is to use that information to predict a patient’s response to treatment with specific targeted therapies.
Validating potential biomarkers is slowly progressing, mainly because of the way cancer clinical trials are set up and because tumor DNA samples can be mixed with healthy cells. "Clinical tumor samples are 100% cancer,” says Harismendy. But to better understand how tumor mutations spread to other cells, scientists need to study samples of mixed tissue. Because the population of cells is mixed, “standard PCR is not realistic for us,” he says.
So, Harismendy and his collaborators used RainDance’s microdroplet PCR system to create a procedure to process mixed cell populations and look for cancer mutations with extremely low prevalence of in the cell population, between 1% to 5%. In their experiment, they screened 71,000 cancer mutational hotspots in 42 cancer genes. The samples were from a colon, breast, ovarian, and sarcoma carcinoma and were mixed with DNA from all four patients' blood. The study showed that the team’s experimental design offered a streamlined way to simultaneously sequence a mass of cancer mutational hotspots in mixed cell-type samples and detect low-prevalence mutations appearing in just 5% of the sample (4).
“We were trying to set this up for a clinical setting where technicians might not have a background in microbiology,” says Harismendy. “We needed a way to prepare PCR amplicons on sets of genes that was quite streamlined, and microdroplet PCR was the easiest method at time.”
With the microdroplet PCR system, the technicians can separate the cells into droplets, add the correct PCR primer and at the end of the process, the samples are ready to go onto sequencing. “There’s not a lot of lab prep like you see with whole exome sequencing or other capture methods. The droplet process is a very simple process, with a limited number of steps, yet the results are robust and streamlined, which is very important,” he says.
Simply put, the new technology is "just as good as regular PCR in a plate, but it is much cheaper and easier," says Harismendy.
Beyond Digital PCR
While RainDance has developed systems that are focused on digital emulsion PCR, Weitz’s lab has continued to develop other technologies as well. "We have several projects that involve digital PCR, but that specific technique is just one small part of what we do,” she says. “The lab’s broader goal is to develop general applications and perform specific research on various topics, regardless of whether that involves digital PCR. RainDance has the industrial capacity to optimize off-the-shelf products and offer solutions that are in high demand, while the lab can develop novel tools tailored to unique applications.”
One specific application is sequencing the trillions of bacteria in the human gut (5). “I'm hoping that [Guo]'s work on microfluidics will help us to address a number of long-standing questions [in this field],” says Harvard systems biologist Peter Turnbaugh. “We would like to know which bacterial species harbor genes that allow the metabolism of drugs or antibiotics. Are these genes widespread or limited to individual microbes? What environmental factors might promote the spread of these genes within microbial communities? The development and application of high-throughput methods for doing successive amplification reactions on single cells is a key first step towards these studies.”
1. Agresti, J., E. Antipov, A.R. Abate, K. Ahn, A.C. Rowat, J.C. Baret, M. Marquez, A.M. Klibanov, A.D. Griffiths, D.A. Weitz. 2010. Ultrahigh-throughput screening in drop-based microfluidics for directed evolution. Proc. Natl. Acad. Sci. 107 (9): 4004–4009.
2. Tewhey, R., J.B. Warner, M. Nakano, B. Libby, M. Medkova, P.H. David, et al. 2009. Microdroplet-based PCR enrichment for large-scale targeted sequencing. Nat. Biotechnol. 27:1025 – 1031.
3. Pekin, D., Y. Skhiri, J.C. Baret, D. Le Corre, L. Mazutis, C.B. Salem, et al. 2011. Quantitative and sensitive detection of rare mutations using droplet-based microfluidics. Lab Chip. 11:2156-2166.
4. Harismendy, O., R.B. Schwab, L. Bao, J. Olson, S. Rozenzhak, S.K. Kotsopoulos, et al. 2011. Detection of low prevalence somatic mutations in solid tumors with ultra-deep targeted sequencing. Genome Biol. 12:R124.
5. Guo, M. et. al. 2012. High-throughput single-cell PCR using microfluidic emulsions. Bulletin of the American Physical Society. PS March Meeting 2012. 57, 1: Abstract K1.00156.