As John Doench from the Broad Institute of the Massachusetts Institute of Technology (MIT) and Harvard University is quick to note, disrupting the expression of a gene is the most tried-and-true way to determine its function. “For about a decade, RNAi was the only game in town for loss-of-function studies at scale. And while tremendously successful in many respects, no single technology is a magic bullet.”
An attractive genome editing tool, the CRISPR (clustered regularly interspaced short palindromic repeats)/Cas system relies on a single guide RNA (sgRNA) to direct the Cas9 protein to a specific DNA sequence, where Cas9 then generates a double-strand break. The break offers an opportunity to insert or delete sequences during repair by non-homologous end joining.
While still being perfected, recent studies are demonstrating clever, new applications for the CRISPR/Cas system that go beyond simple gene editing. And for scientists like Doench, this wave of CRISPR applications just might be the ticket. “A new technique that allows us to answer many of the same questions but in an orthogonal way is really going to accelerate our understanding of gene function,” he said.
Screening for Genes
Doench and his collaborators set out to tackle the challenge of using the CRISPR/Cas9 system to understand gene function by designing a single lentiviral vector to simultaneously deliver Cas9 and an sgRNA into target cells. They then designed a library of sgRNAs targeting the exons of about 18,000 genes in the human genome. “This delivery enables each cell in a large pool to get a different CRISPR and thus carry a different gene mutation,” explained study author Neville Sanjana, also at the Broad Institute and MIT.
The authors next selectively enriched a small group of cells that showed a particular feature of interest, in this case, resistance to a melanoma drug, and applied next-generation sequencing to this subset of cells to identify the mutated genes causing drug resistance. Their analysis revealed genes previously implicated in resistance to the drug as well as novel gene candidates, opening new avenues for testing hypotheses about the mechanisms of drug resistance.
“A challenging aspect of any functional genomics screen is to find a strong phenotype that can be screened in a robust manner,” said study author Ophir Shalem of the Broad Institute and MIT. “Given the clarity and strength of the results we observed, we are now in a position to screen more subtle phenotypes than previously possible.”
The researchers are now applying their approach to patient-derived cells to determine whether specific genetic backgrounds influence susceptibility to drug resistance or disease progression. In the end, the CRISPR/Cas9 screening approach could lead to a deeper understanding of the underlying causes of a broad range of diseases at the genetic level, as well as the discovery of new targeted therapeutics, according to study author Feng Zhang of the Broad Institute and MIT.
At the same time this study was published, another group of researchers reported a similar loss-of-function genetic screening approach in Science (3). David Sabatini and Eric Lander of the Broad Institute and their collaborators used a pool of sgRNA-expressing lentivirus to generate a library containing about 73,000 sgRNAs targeting about 7000 genes and then screened for genes whose loss conferred resistance to a chemotherapeutic drug.
“The concept of genome-wide knockout screening using the CRISPR/Cas system is a natural extension of its previous application in targeted gene editing, but clearly these two groups are the first to develop such screening platforms and demonstrate these important proof-of-concept results,” said Joanna Yeh, a CRISPR-Cas expert at Harvard Medical School and Massachusetts General Hospital who was not involved in the new studies. “I think we will see other genome-wide uses of the repurposed CRISPR/Cas system very soon.”
Despite the promise of these CRISPR/Cas techniques, the potential clinical impact may not become apparent for years to come. “Engineering genomes and probing gene function in the lab is all well and good, but developing this technology to the point where it can be used clinically is a major challenge,” Doench said. “The science will develop at its own pace, but as we've seen in the last 10 years of RNAi-based therapeutics, the road to commercial success can be quite bumpy and drawn-out.”
In Living Color
As Doench worked through his CRISPR/Cas screening platform, across the country at the University of California, San Francisco (UCSF) Bo Huang and his collaborators were searching for a way to visualize specific DNA sequences in living cells.
Previous imaging methods used fluorescently tagged DNA-binding proteins, which are restricted by their fixed target sequence and limited choices of native DNA-binding proteins, and fluorescence in situ hybridization (FISH), which offers target sequence flexibility through base pairing of the nucleic acid probes. With the emergence of CRISPR/Cas9, Huang realized that this system might help his imaging issues. He teamed up with Lei Stanley Qi, a CRISPR expert at UCSF, and together they used an endonuclease-deficient Cas9 protein (dCas9), which lacks the enzymatic ability to produce DNA double-strand breaks, and tagged this protein with enhanced green fluorescent protein (EGFP) (1). “This approach provides unprecedented flexibility for the dynamic imaging of specific genomic elements in living mammalian cells,” said Virginijus Siksnys of Vilnius University, who was not involved in the new study.
“FISH is probably the most popular chromatin imaging method, but it disrupts the native conformation of chromatin because it requires DNA denaturing,” Qi said. “The new CRISPR imaging technology allows us to visualize chromatin in live cells without disrupting its native conformation.”
To develop a genome-imaging technique that combines the flexibility of nucleic acid probes and the live imaging capability of DNA-binding proteins, Huang, Qi, and their collaborators had to re-optimize the CRISPR system, including the sgRNA design. “CRISPR imaging needs to maintain an extremely low level of dCas9 to reduce the background of free dCas9-GFP. At the same time, a sufficiently high sgRNA level is required to ensure efficient dCas9-sgRNA complex formation,” Huang explained. “In fact, the optimization that we did is not only essential for imaging but also helpful for CRISPR interference, and hopefully useful for gene editing, too.”
Using their optimized system, the researchers achieved robust live imaging of protein-coding genes as well as repetitive elements in telomeres. To visualize non-repetitive sequences in the human genome, they used an array of sgRNAs tiling along the target locus. They were able to monitor telomere dynamics as well as chromosome dynamics during mitosis. “Simultaneous expression of more than 30 sgRNAs for the imaging of non-repetitive sequences was a major challenge,” Huang said. “We were worried that treating the cells with so many different viruses at the same time was a crazy idea. But to our surprise, it worked in the very first trial.”
To improve the method, the number of sgRNAs needs to be reduced because high concentrations of sgRNA-encoding lentivirus can be toxic to cells, Qi explained. “Also, the sensitivity of chromatin imaging should be improved to reduce the background fluorescence signal. This is particularly important for imaging cells with a small nucleus volume or complex 3-D structures.”
Moving forward, the researchers plan to develop a two-color method for simultaneously visualizing different genome sequences in the same cell. This would allow them to study the relative movements between different chromatin regions. “As a long-term goal, we plan to develop imaging tools that enable the visualization and capture of chromatin with specific epigenetic states,” Qi said. He also envisions the future development of tools for epigenome editing—modifying chromatin marks.
But until then, CRISPR/Cas researchers have their work cut out for them. New approaches for reducing off-target effects will be important for further improving this system, experts say. And more information on the Cas9 protein is needed. “The Cas9 protein in wide use, from Streptococcus pyogenes, is still largely a black box,” Doench said. “A better understanding about all of its functions, and whether it is the best Cas9 to use, are important next steps in the development of this technology.”
1. Chen B, Gilbert LA, Cimini BA, Schnitzbauer J, Zhang W, Li GW, Park J, Blackburn EH, Weissman JS, Qi LS, Huang B. 2013. Dynamic imaging of genomic loci in living human cells by an optimized CRISPR/Cas system. Cell 155(7):1479-91. doi: 10.1016/j.cell.2013.12.001.
2. Shalem O, Sanjana NE, Hartenian E, Shi X, Scott DA, Mikkelsen TS, Heckl D, Ebert BL, Root DE, Doench JG, Zhang F. 2014. Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 343(6166):84-7. doi: 10.1126/science.1247005. Epub 2013 Dec 12.
3. Wang T, Wei JJ, Sabatini DM, Lander ES. 2014. Genetic screens in human cells using the CRISPR-Cas9 system. Science 343(6166):80-4. doi: 10.1126/science.1246981. Epub 2013 Dec 12.