Two separate groups at Stanford University published improved DNA amplification methods last week that could help researchers prepare DNA from a single cell, or select thousands of DNA targets, for next-generation sequencing.
Both groups say that their approaches could be used in sequencing projects where scientists have to work with minute quantities of starting material, such as microorganisms that cannot be cultured, tumor biopsies, or stem cells.
The first method is an improvement of an approach to amplify DNA from individual microbial cells using a microfluidic chip. Based on multiple displacement amplification, it uses nanoliter instead of microliter reaction volumes to amplify the DNA. This method was published last week in PLoS Genetics by a team led by Stephen Quake at the department of engineering at Stanford University.
The other method, invented by a group of scientists led by Nader Pourmand at the Stanford Genome Technology Center and published last week in PLoS ONE, enables, in principle, multiplex amplification of thousands of DNA targets from a single cell.
But the researchers have not yet used their method on single cells, and have so far only amplified and sequenced two genes in parallel. However, they plan to develop the approach further in order to analyze thousands of genes in single stem cells.
Stephen Quake’s group at Stanford, in its study, used a microfluidic chip to isolate individual E. coli cells. The researchers amplified these by multiple displacement amplification in 60-nanoliter reactions and showed that the quality of the genome amplification, especially amplification bias, was improved over standard 50-microliter reaction volumes.
In their study, the researchers used 454’s sequencing platform to sequence amplicons and showed that “the specificity from single cells is extraordinarily high,” suggesting that “it may be possible to perform full genome assemblies with this procedure,” according to the article.
Pyrosequencing has “the advantage of simplified library construction,” the authors note, and although read lengths are shorter than in Sanger sequencing, “substantial de novo assembly was achieved.
“The ability to rapidly acquire even substantial portions of the genome, when used to aid in assembly of existing metagenomic shotgun data, promises to greatly accelerate genomic discovery of new microbial species,” they write.
Indeed, the researchers now plan to use the method for studying microbial diversity, Quake told In Sequence by e-mail. His group, in collaboration with the Joint Genome Institute, has already applied the method to sequence “most of the genome” of a bacterium from the human mouth, a study they published in PNAS in July.
He believes the method “can be modified for human genome amplification, but we have not tried this yet.”
In humans, the method could be “very useful for projects where there is limited material, such as cancer needle biopsies or pre-implantation screening for in vitro fertilization,” he said.
Researchers can obtain the “basic” microfluidic chips with the nanoliter reactors from Quake’s lab, he said. Also, Fluidigm is making a higher-throughput commercial version, which includes control hardware and software, he added.
“The results presented in this paper are really encouraging,” Kun Zhang told In Sequence by e-mail. Zhang, who is an assistant professor in the department of bioengineering at the Univerisity of California, San Diego, published a paper on sequencing genomes from single cells using polymerase cloning last year. “I would not be surprised if many unknown microbes will be amplified and sequenced using this microfluidic device.”
But the researchers still have to show that they can scale up their approach. “The next big challenge would be to amplify thousands, or tens of thousands, of microbes on microfluidic chips at once,” Zhang said.
Connector Inversion Probes
Pourmand told In Sequence last week that his team’s method is “basically an upstream technique for the next-generation sequencing platforms, starting from minute [amounts], or single-cell material.”
The method uses so-called padlock probes, single oligonucleotide primers that contain, at their ends, sequences that are complementary to the target. The research team’s version of these probes, which it calls connector inversion probes, bind to the ends of a target sequence. “If one of [the ends] lands on a gene, and the other side is not present, it will not work,” Pourmand explained. “Both of them have to bind to be able to create a target.” The gap in between them is then filled by DNA polymerase, followed by ligation, and the target DNA is amplified using universal primers.
“The ability to rapidly acquire even substantial portions of the genome, when used to aid in assembly of existing metagenomic shotgun data, promises to greatly accelerate genomic discovery of new microbial species.”
The CIP technology is compatible with any sequencing technology, he said, including next-generation platforms from 454 Life Sciences, Illumina, Applied Biosystems, and Helicos BioSciences.
While other DNA-amplification methods are based on random, or global, amplification, this one selects regions of interest first, then amplifies the material, Pourmand said.
In contrast to a method based on so-called selector technology, published by yet another Stanford team in PNAS earlier this year (see In Sequence 5/22/2007), his method does not require the genomic DNA to be digested with a restriction enzyme first, which could lead to the loss of DNA, making it incompatible with single-cell procedures, he said.
Although his group has only amplified a few genome regions in parallel, he expects that it will be possible to amplify at least 30,000 targets in parallel.
Researchers using a related technology, molecular inversion probes, have already published that they can use it to genotype 12,000 SNPs in parallel, and Pourmand knows they have used it to an even higher degree of multiplexing. “We have no reason to believe it will be any different” for connector inversion probes, he said.
Molecular inversion probes, commercialized by ParAllele Bioscience, which Affymetrix acquired two years ago, use a modified version of padlock probes for SNP detection rather than gene amplification, he explained.
Connector inversion probes can also be barcoded, enabling researchers to interrogate several different samples in parallel, he added.
Pourmand, who is taking a faculty position at the University of California, Santa Cruz, in January, plans to use the technology to sequence thousands of targets in stem cells. “The first thing I want to do there … is use this technology with stem cells, and sequence certain targets in a multiplex fashion,” he said.
The goal is to optimize the method for use in single cells. “The concentration of probes that we are putting in is really critical,” he said. “If it’s too much, they interfere with each other; if it’s too little, they don’t find a target. We have to find a sweet spot, basically. It’s just optimization and a lot of patience.”
But it might not be all that easy. According to UCSD’s Zhang, scaling up from two targets to several thousand is “not trivial.” He also does not believe that the method can currently be used to amplify genes from single cells, based on the amount of starting DNA the researchers used in their published study.
By the end of the year, Pourmand said, UCSC plans to acquire at least two next-generation sequencers but has not decided on the platforms yet.
Other researchers can use the method by following the published protocols, he said, which include a software application to design connector inversion probes.