blog

Guest post by Jeffrey Hannah:

Although genome editing tools have been around for years, the field essentially exploded in 2013 and has been expanding ever since. Seemingly overnight, the preferred gene knockdown technique in biomedical research has shifted away from RNA interference – which came from plant and invertebrate research – and towards CRISPR-Cas9 technology, which is used by bacteria to recognize and attack viral DNA. CRISPR-Cas9 has become the tool of choice over TALEN and zinc-finger systems because it requires minimal engineering – all you need to do is introduce the Cas9 nuclease and a designed guide RNA that is complementary to a short segment of the gene of interest. One of the reasons that the field has expanded so rapidly (one speaker claimed over 1,000 papers have already been published using the technology) is that much of the groundwork has already been laid with pre-existing genomic engineering strategies (lox-Cre, tetracycline induction, etc.) and nucleic acid delivery systems.

Last week, the New York Academy of Sciences (along with organizers from Pfizer, Inc.) held a one-day meeting focusing on the intersection of the nascent field of genome editing and translational research. As with any new technology, there are concerns and problems that need to be ironed out. At the meeting, some of the major concerns/limitations were discussed: non-specific cutting at unintended sites in the genome, the physiological effects of expressing a prokaryotic nuclease (Cas9) in mammalian cells and the relatively low efficiency of the CRISPR-Cas9 system. Both Yi Yang (Novartis) and Lukas Dow (Weill Cornell Medical College) explained how using a modified form of Cas9 that only cuts a single strand of DNA (a “nickase” as opposed to the wild-type Cas9 endonuclease which cuts both strands) and two separate guide RNAs profoundly reduced any observed off-target effects. Dr Dow went as far as to call this the “system of choice” for in vivo models.

Regarding Cas9’s effects on cellular and animal health, Randall Platt from Feng Zhang’s lab (MIT) described their transgenic mouse line, which expresses Cas9 ubiquitously and constitutively. He claims that they found no effects on mouse health and fertility and that even neurons (which are notoriously sensitive) performed basic functions normally. To reduce any further reservations, Cas9 can also be expressed under inducible or tissue-specific conditions or introduced transiently via adenoviral infection.

CRISPR efficiency still has much room for improvement but Chad Cowan (Harvard) found that CRISPR was more efficient than TALEN and caused fewer off-target mutation events (as determined by deep sequencing). For gene knockout, Dr Yang described in vitro experiments in which chemical reagents were introduced alongside CRISPR to shift DNA repair towards error-prone non-homologous end joining and away from highly efficient homologous recombination to increase the frequency of desired frame-shifting insertion/deletion events.

In terms of designing effective guide RNAs, the results are less predictable and researchers must design and test multiple guides. Although there are almost certainly multiple groups working on it, we don’t currently have a genome-wide library of guide RNAs that can be used for large-scale screening studies. This is the next logical step for CRISPR technology and should yield abundant information about genes that are important for drug responses in cancer and polygenic diseases. Jim Inglese (NIH) talked about the potential of CRISPR for high-throughput drug screening. Danilo Maddalo from Andrea Ventura’s lab (Memorial Sloan-Kettering Cancer Center) also spoke about how CRISPR-based screens could be used to identify mechanisms of drug resistance in cancer.

For now, genome editing is providing researchers with lots of interesting opportunities to manipulate their mouse and tumor cell models. Dr Maddalo described how his group is using CRISPR to induce chromosomal rearrangements involving the ALK kinase. Randall Platt provided examples of how CRISPR-Cas9 can be used to knock out several genes at once by using multiple guide RNAs. He claimed that CRISPR creates a more natural lung cancer model because the system is inefficient and doesn’t mutate all lung cells – which is closer to the natural progression of lung tumorigenesis. Dr Cowan showed how using CRISPR can be used to target genes in stem cells, which allows for the creation of multiple isogenic cell types.

However, CRISPR isn’t the only genome editing system available. Dr Yang extolled the virtues of zinc-finger nucleases (ZFNs) because they produce defined overhang sequences. As a result, it is possible to create an array of mutant proteins for functional analysis by using ZFNs along with various oligonucleotides (to replace the endogenous sequence). Michelle Luo (NC State Univ.) uses genome editing to study prokaryote biology but uses the type I CRISPR that involves crRNA, a DNA-binding protein, Cascade and the Cas3 nuclease. Rather than mediating Cas3-induced DNA hydrolysis, Cascade can be engineered to recruit transcription factors and alter target gene expression.

Regardless of the specific system or mode of delivery, genome-editing technologies are already firmly established in biomedical research. Let’s hope that funding agencies are taking notice since, once again, translational medicine stands to gain enormously from a discovery that originated in basic research.