Abstract
Nishiyama J, Mikuni T, Yasuda R. Neuron 2017;96,755–768.
Precise genome editing via homology-directed repair (HDR) in targeted cells, particularly in vivo, provides an invaluable tool for biomedical research. However, HDR has been considered to be largely restricted to dividing cells, making it challenging to apply the technique in postmitotic neurons. Here we show that precise genome editing via HDR is possible in mature postmitotic neurons as well as mitotic cells in mice brain by combining CRISPR-Cas9-mediated DNA cleavage and the efficient delivery of donor template with adeno-associated virus (AAV). Using this strategy, we achieved efficient tagging of endogenous proteins in primary and organotypic cultures in vitro and developing, adult, aged, and pathological brains in vivo. Thus, AAV- and CRISPR-Cas9-mediated HDR will be broadly useful for precise genome editing in basic and translational neuroscience.
Commentary
Gene editing as a laboratory tool and as a future treatment for epilepsy gives us reason to be optimistic. Advanced gene editing technologies could have the capacity to insert deleted genes, modify transcription factors, or repair pathological de novo single nucleotide variations. This has obvious implications for genetic, and potentially even acquired, epilepsies. Paired with affordable whole genome sequencing and bioinformatic analysis, gene editing could be a tool as flexible as the causes of epilepsy are variable. However, the future is not yet here. One significant hurdle is the difficulty precisely editing genes in neural tissue. Until recently, the available gene editing techniques for postmitotic neurons (1) were prone to insertion/deletion (indel) errors, thus risking knockout of the target gene. In a major step toward overcoming this challenge, Nishiyama et al. report remarkably accurate gene editing in mature neurons in vivo. Their progress in precisely editing neuronal genes gives the epilepsy research community an important new laboratory tool and moves us toward a future with lasting genetic fixes.
Nishiyama et al. achieved this step forward through adeno-associated virus (AAV) delivery of homology directed repair (HDR) machinery and CRISPR-associated endonuclease Cas9. Cas9 plays a prominent role in the ongoing gene editing boom, impacting a wide range of biological sciences. Two exemplar Cas9 applications include genome-scale knockout screening in cell culture (2–4) and building artificial immune systems against HIV (5). Both leverage Cas9's ability to make double-stranded DNA breaks at a chosen location based on a guide RNA (gRNA). However, cleaving the DNA is only a first step toward editing a cell's genome. Cas9 gene editing relies on endogenous double-stranded DNA repair machinery for nonhomologous end joining (NHEJ) or homology directed repair (HDR) to make a mistake and insert exogenous DNA. NHEJ often makes random indel errors during repair, even causing frame shifts, and thereby loss of function. HDR, in comparison, is highly accurate using either the homologous chromosome or exogenous template DNA to repair or insert DNA. However, HDR is generally limited to dividing cells in G2 and S phase, and therefore does not work well in (postmitotic) mature neurons. Highlighting the importance of this issue, a recent review on Cas9 applications to neuroscience set forth the goal of HDR gene editing in adult neurons (6).
Nishiyama et al. report achieving this very goal, demonstrating HDR in not only embryonic or cultured neurons, but importantly also in adult neurons in vivo in aged mice. To do this, they delivered HDR machinery (including the gRNA and the HDR template) via AAV, which may have been key to their success. AAVs robustly infect nondividing cells, have a single stranded vector genome that gets localized to the nucleus, can be used at a high titer to achieve high copy numbers, and speculatively might even trigger activation of gene editing machinery as part of an immune response. Using this method, together with Cas9 expression, the authors were able to achieve genetic modification in up to 30% of infected cells. The observed gene editing required Cas9-mediated double strand breaks, as delivery of the HDR machinery alone (without Cas9) did not induce detectable gene alteration. They also demonstrated that HDR underlies the alterations in experiments using organotypic slice culture from mice and rats. These experiments targeted the CaMKIIa locus, allowing the researchers to leverage subtle sequence differences between mice and rats in their design of the homology arm of the AAV-delivered template DNA. Evidence of genetic modification only occurred when the correctly matched sequence was delivered, supporting the idea that HDR (rather than e.g., NHEJ) was underlying the edits; their HDR constructs designed for the mouse sequence didn't work in the rat and vice versa, showing that homology was required. The authors also cleverly designed the constructs such that the double strand breaks would occur in the noncoding region to reduce the risks of unintended indels affecting gene expression, and found that the endogenous gene was not deleted in 98.1% of infected neurons.
Two things are apparent: 1) these findings mark an important step forward in gene editing in neurons, and 2) there remains room for further advances. Notably, while not deleting the targeted gene in 98.1% of targeted neurons is a good start, losing gene function in 2% of cells could have detrimental effects in many circumstances. Similarly, while achieving successful modification in up to 30% of infected neurons is remarkable, it may not be sufficient to treat many neurological disorders. There is also the possibility of negative off-target effects that went undetected in the current study, such as mutations in genes not examined, or the long-term consequences of expressing Cas9, exogenous HDR machinery, and high-copy number AAV infection. Many advances to improve the system are already in the works, including more selective, and therefore more error-resistant, Cas9 variants (7–9). Even once the required technologic hurdles are cleared, and safe and efficient gene modification is achievable, significant considerations remain. These include 1) that developmental alterations or other negative consequences of the neurological disorder (including cell death and circuit reorganization) will not necessarily be reversed once the pathological gene is altered, 2) insufficient knowledge surrounding which genetic variations are indeed pathological, and 3) ethical considerations: gene editing in neurons has the potential to make lasting changes to the patient's brain and thereby their personality, cognition, capabilities, and interpersonal relationships (arguably not unlike neurosurgery).
While the road ahead for successful genetic modification is neither short nor easy, Nishiyama et al. have brought us a crucial step closer. Additionally, even before future advances are made that would allow a clinical application of their work, the current advances provide an important new experimental tool. For example, Nishiyama et al. used the technology to express a fluorescent protein fused to endogenous β-Actin in a mouse model of Alzheimer's disease. Notably, this did not require the generation of a new mouse line nor the often arduous and time consuming crossing of existing lines, and could be done at a specific time point in the disease state. The hard fought advancements in gene editing technology clearly give us new ways to study the epilepsies, and can in turn also help advance future gene therapy options clinically, for example, by improving characterization of de novo single nucleotide variations.
As the first applications of Cas9 in human patients begin (10) and as research (like that reported by Nishiyama et al.) moves us closer to safe and efficient neuronal gene modification, we progress cautiously optimistic. Hopefully additional, more accurate, highly efficient gene editing methods for adult neurons will continue to be developed. These technologies propel basic research and could eventually lead to gene modification clinical trials for epilepsy. After all, DNA is often central to development and pathology; could it be central to treatment?