NEW YORK (GenomeWeb) – Cas9 is, at heart, a nuclease. And while scientists have found that its killer app is gene targeting and editing in eukaryotic cells, new studies are returning it to the test tube and the microbial cell to simply cut up DNA.
In the last six months, several published studies have found new applications for CRISPR/Cas9, using it as a programmable nuclease. In March, scientists at the University of California, San Francisco led by Joe DeRisi published a paper in Genome Biology describing the use of Cas9 to remove unwanted DNA species from next-generation sequencing libraries and PCR-based diagnostics. And last month, scientists from China's Tsinghua University led by Ting Zhu published a protocol using Cas9 to clone long 100-kilobase genomic sequences in bacteria. Other studies are demonstrating increased efficiency in vitro, expanding possibilities for synthetic biology. As many have predicted, CRISPR is well on its way to becoming a trusted, handy tool in the biologist's kit.
"It's the tip of the iceberg in terms of in vitro uses of Cas9 and Cas9-like proteins for molecular biology applications at the bench," DeRisi told GenomeWeb. "They're coming out of the woodwork."
Alejandro Chavez, a clinical fellow in George Church's lab at Harvard Medical School, said that while the bench applications of CRISPR/Cas9 have so far been "kind of obvious," there's a lot of new science that can be done with them. "It has opportunities for anyone," he said.
Early on, scientists found ways to use Cas9 to improve basic lab work. In 2014, South Korean scientists led by Jin-Soo Kim of Seoul National University published a method to do genotyping by restriction fragment length polymorphism (RFLP) using CRISPR/Cas9.
"RFLP analysis is one of the oldest, most convenient, and least expensive methods of genotyping, but is limited by the availability of restriction endonuclease sites," the authors wrote in Nature Communications.
With Cas9, there's now an easy workaround to the problem of not having a restriction enzyme site in one's genomic area of interest. An Agilent Technologies spokesperson told GenomeWeb that customers are using its SureGuide CRISPR/Cas9 kit for cloning applications, but declined to provide an example.
About a year after Kim's team showed that Cas9 created new opportunities for simply cutting DNA, scientists from the National Cancer Institute demonstrated the ability to do transformation-associated recombination cloning of entire genes, complete with distal regions, as well as longer chromosomal loci in yeast. They improved the efficiency of this method by an order of magnitude, from between 0.5 and 2 percent all the way up to 32 percent. They published their work in Nucleic Acids Research.
Around that same time, Zhu's team at Tsinghua University described a similar method of cloning large microbial genes and gene clusters. In an email, Zhu noted that his paper addresses the problem of cloning large bacterial genome sequences, up to 100 kilobases, in a single step.
Zhu published the details of this method, called Cas9-assisted targeting of chromosome segments (CATCH) this month in Nature Protocols. It's performed in an agarose gel, which he said helps prevent DNA shearing. Bacteria are embedded into the gel, the cell is lysed, RNA-guided Cas9 digests the bacterial chromosome, and the target is then ligated into a cloning vector. Finally, the vector is electrotransformed into a new cell for production.
"That was incredibly hard to do before," Chavez said. Zhu's method could accelerate metagenomic mining. Chavez said scientists have been mining bacterial genomes for operons, sections that encode a promoter and an entire genetic pathway, for research in bioproduction and drug discovery. Sometimes, the operon is found in bacteria that are difficult to culture and is too big to clone with PCR. CRISPR-based cloning could help scientists harness the ocean of enzymes that microbes have evolved to do particular tasks and put them to use for humans.
Zhu and his co-authors noted in their Nature Communications paper that CATCH could also have sequencing applications. "The ability of CATCH to isolate near-arbitrary DNA segments from the known, flanking sequences may greatly accelerate efforts to fill the gaps in sequenced genomes," they said, adding that it could also enable targeted sequencing of disease-specific genes in humans."
That's exactly what DeRisi and his colleagues at UCSF used Cas9 cleavage for, enhancing other transformative technologies in modern biology, namely NGS and PCR. They've developed a process called Depletion of Abundant Sequences by Hybridization (DASH) and it's a sort of enrichment by depletion.
"It's exactly the opposite of enrichment," DeRisi said. "You actually get enrichment by destroying the stuff you don't care about."
DeRisi's team showed in the Genome Biology paper that it could both reduce mitochondrial rRNA in HeLa cells and enrich pathogen DNA present in patient samples. The concept is simple: impede the amplification of unwanted DNA species by cutting them up with CRISPR/Cas9. Any sequence that gets cleaved won't have flanking adaptors and thus won't get amplified. For NGS, this takes place after library construction and before sequencing.
"All it takes is 50,000 to 100,000 guide RNAs to remove all the stuff I don't want to see, with super high specificity," DeRisi said. "You get target enrichment without having to know what the target is."
While bead-based depletion is a similar approach, DeRisi chided that it was "sloppy, less precise, and less quantitative," than DASH. "Cas9 enforces specificity. Unless [there is an] exact match, you're not going to take it out," he said. Rice University professor David Zhang and his company, Searna, are also working on a hybridization-based method for depleting wild-type DNA for sequencing and PCR applications
Both DeRisi and Chavez pointed to the applications for diagnostics.
"If I go sequence a human sample to see if they have a pathogen, 99 percent of [the DNA] is going to be human," DeRisi said. "That's not what I care to see. What I care to see is the pathogen."
For something like cancer diagnostics, DASH could help solve the problems that deep tumor sequencing couldn't solve, Chavez said. "There are naïve NGS errors that arrive during the amplification step. So you find things you think are [cancer] alleles, but they're not really there."
Removing wild-type DNA could prove to be especially useful in liquid biopsies. Chavez said that for solid tumors, smaller tumors shed less DNA than larger ones. Enriching the circulating tumor DNA by depleting the wild type could help monitor and diagnose small nodules that may or may not be benign.
The key is specificity, which clinical diagnostics require and CRISPR/Cas9 provides.
"If you're depleting the wild type at the same rate as the allele of interest, then you're not helping yourself," Chavez said.
DeRisi said that DASH is cheap, at about $4 per reaction. He said he has no plans to commercialize the technology. "I'm sure anybody could kitify it," he said. "But at $4 per reaction, anybody could boot it up in their own lab."
"Tools like this shouldn't be owned, nor should the applications," he said.