By Julia Karow
This article, originally published March 1, has been updated from a previous version to include additional information from Life Technologies.
Life Technologies last week provided its long-anticipated answer to Pacific Biosciences' single-molecule real-time sequencing technology, publicly presenting details of its own technology for the first time at the Advances in Genome Biology and Technology conference.
Life Tech's approach — internally named "Starlight" — uses a single quantum dot nanocrystal tethered to a DNA polymerase as a photon source for four-color terminally labeled nucleotides and measures light emitted from both the Qdot and the labels in real time as bases get incorporated by the polymerase.
The technology is at a relatively early stage — so far, company scientists have only sequenced synthetic DNA of about 100 bases on a dozen or so prototype machines — but early-access collaborations with research groups are slated to start in the fourth quarter. Next year, Life Tech plans to start putting the technology into customers' hands, company officials told In Sequence last week.
According to Joseph Beechem, chief technology officer and head of single-molecule sequencing research at Life Tech, the system will provide "continuously tunable" read lengths and accuracy. He presented the technology in front of a packed auditorium during the last session of the AGBT meeting on Saturday.
The company is aiming for a so-called "fundamental read length" of 1 to 1.5 kilobases by the end of the year, which can be increased several-fold by stripping off inactive polymerase and adding fresh enzyme to continue the read.
Instrument runs are expected to take on the order of 30 to 60 minutes. So far, the system runs at 1 to 5 bases per second "routinely," though company scientists have already obtained faster speeds.
At present, Life Tech scientists can monitor signals from about 50,000 single-stranded DNA strands at a time that are bound to the surface of a glass slide above a microscope objective. After letting Qdot-labeled polymerase bind to the primer-template complexes, nucleotides are added to start DNA synthesis. Similar to PacBio's technology, each of the four nucleotide types is labeled with a different organic fluorescent dye at their terminal phosphate group.
The 10-nanometer Qdot nanocrystals, which consist of inorganic CdSe and ZnS atoms and are "very photostable," were designed specifically for the sequencing technology, Beechem said, since the commercially available crystals were too large for high-resolution sequencing. He said that the Qdots do not change the properties of the DNA polymerase, which still "behaves absolutely identical to Mother Nature's version."
Each Qdot soaks up light from a single 405-nanometer excitation laser — about 100 times more light than a "good organic dye," according to Beechem — and emits all of that light again, leading to a bright signal. These properties, he told In Sequence, allow the company to use a weaker laser than other platforms, conferring less damage to the polymerase and making the system easier to build.
When the correct nucleotide binds to the polymerase, the Qdot transfers some of its light to the nucleotide dye in a process called Förster resonance energy transfer, or FRET. The dye, in turn, emits light in the 630- to 800-nanometer range — depending on its type — which is separated spectrally, collected by an EMCCD camera over time, and translated into base calls by software in real time. The polymerase incorporates the nucleotide, generating natural DNA, and cleaves off the dye.
Because FRET only happens in the vicinity of the Qdot, the signal is spatially confined, and nucleotides in solution rarely light up, keeping the background noise low.
A second signal for each base incorporation comes from the Qdot itself: as it transfers photons to the nucleotide label, it becomes dimmer for a short time. "This extra sequencing signal allows you to make a base call at a whole new level of accuracy," Beechem said. He told In Sequence that the aim is to get the single-read accuracy percentage "in the 90s range."
[ pagebreak ]
In order to bring the accuracy up further, Life Tech scientists can also strip the newly synthesized strand off the template and resequence the same template again — a process they call "recursive sequencing" that is reminiscent of PacBio's "circular consensus sequencing" but does not require the template to be circularized. So far, they have been able to sequence the same strand six times over, according to Beechem.
Another feature that distinguishes Life Tech's technology from PacBio's is its ability to replace inactive polymerase with fresh enzyme and continue sequencing the same strand, leading to "continuously tunable" read lengths. Beechem said that, so far, he and his colleagues have gone through six such sequencing cycles and have seen "no loss of signal" and "no gaps in sequence" for five cycles, though he did not mention the expected upper read length limit.
The company is leveraging several existing labeling technologies from its various branches for the technology: for example, it uses organic dyes from its Molecular Probes brand and Big-Dye chemistry to label the nucleotides.
Parts of the FRET-based technology for the detection of base incorporations — or at least the intellectual property — came from VisiGen Biotechnologies, which Life Tech's Invitrogen unit acquired in 2008, but Beechem did not acknowledge any contribution from VisiGen. Life Tech is currently in litigation with VisiGen co-founder and former CEO Susan Hardin over alleged breach of contract and other accusations (see In Sequence 1/20/2010).
Beechem emphasized that Life Tech sees its single-molecule technology as a complement rather than a replacement for its SOLiD short-read, high-throughput sequencer. "Each one has their strength, their weaknesses and unique characteristics," he said.
While the "gen-3" single-molecule sequencer would be used for de novo sequencing, haplotype phasing, structural variation analysis, RNA structure studies, and the analysis of gene families and repeat regions, other applications, such as gene expression analysis and whole-genome sequencing, will remain the domain of the SOLiD system.
Beechem said that starting in the fourth quarter, a small group of scientific collaborators "will be putting this single-molecule sequencer through its paces for a couple of applications."
One of the first early-access customers for the technology appears to be the J. Craig Venter Institute. "At JCVI, we are eager to apply the technology in a number of projects, including haplotype phasing of the human genome, environmental sequencing, and rapid sequence identification in our infectious disease work," Venter said in a statement released by Life Tech this week.