neWs and VieWs Epigenome editing to the rescue Epigenetic dysregulation of gene expression is a major contributor to many human dis- eases. Targeted alteration of the epigenome became possible with the advent of customiz- able DNA binding domains, and the approach was quickly tested in animal models1 and in clinical trials. In recent years, the develop- ment of easily programmable genome editors based on CRISPR–Cas9 has renewed inter- est in epigenome editing technologies and their therapeutic applications2. Now, a new Reproduced with permission from Liu, X.S. et al. Cell 172, 979–992.e6 (2018) study in Cell3 has demonstrated the power of using specific editing of epigenomic marks to reverse the effects of a genetic mutation. repeat methylation was so closely linked to the neurons derived from the implanted cells Epigenome editing makes use of the same removal of heterochromatin marks and the still expressed FMR1. Interestingly, at least customizable DNA binders (zinc finger pro- appearance of active chromatin marks at the in vitro, maintaining expression of FMR1 did teins, TALEs or CRISPR–Cas9) that are used promoter,” says Charles Gersbach, professor of not require sustained dCas9–Tet1 activity. for genome editing or for general transcrip- biomedical engineering at Duke University in When the authors expressed a Cas9 inhibi- tional activation or repression. But instead of Durham, North Carolina. tor, FMR1 expression was unchanged for at being fused to a nuclease or to a transcrip- Off-target demethylation events were rare. least two weeks. tional activator or repressor, the DNA binder Although ChIP-seq detected >1,000 sites that Further characterization of the therapeu- carries an enzyme that puts in place or erases were at least transiently bound by the dem- tic effects of dCas9–Tet1 was complicated by a specific epigenetic mark2. Although epi­ ethylase, only 29 of these loci showed substan- the limitations of the available mouse models. genome editors have been a boon to scientists tial demethylation. Of the 28 off-target genes Inserting the extended repeats into mouse investigating the mechanisms of epigenetic affected, most showed no change, and none Fmr1 does not result in DNA hypermethyla- regulation, their utility for therapeutic pur- showed more than a fourfold change, in expres- tion or gene silencing. The limitations of the poses has not yet been tested. sion levels. “One of the lessons that we learned fragile X mouse models also make it difficult As a first step toward therapeutic epi­ from gene therapy is that we have to carefully to assess whether post-natal reactivation of genome editing, Liu et al.3 studied a Cas9- assess safety of the treatments before going FMR1 will be sufficient to achieve a cure. based DNA demethylase in a model of fragile into a clinical trial,” says Angelo Lombardo of “While this is an important first step, it X syndrome. This condition affects about the San Raffaele Telethon Institute for Gene remains to be seen how well this approach 1:3,600 males and is the most common cause Therapy in Milan. “It is reassuring that the off- will translate into clinical applications. of male intellectual disability. It is caused by a target effects observed here are limited, but one Delivery is an important issue as the dCas9– trinucleotide repeat expansion in the 5′ UTR needs to keep in mind that epigenetic off-target Tet1 fusion is too large for commonly used of the FMR1 gene. In individuals with more effects might be much more context-dependent AAV [adeno-associated virus] vectors, than 200 of these repeats, the repeat region than genomic off-target events.” although it should be possible to split the is hypermethylated, leading to formation of Restoration of FMR1 expression was suf- construct and deliver it in separate vectors,” heterochromatin at the gene promoter and ficient to rescue cellular phenotypes associ- says Lombardo. And Gersbach highlights gene silencing. ated with fragile X syndrome. Neurons that the potential issue of continuous transgene To reverse the hypermethylation, the differentiated from methylation-edited iPSCs expression: “Ideally, one would like to avoid authors designed single-guide RNAs that retained close to normal FMR1 expression lev- long-term expression of the epigenome edit- target a catalytically inactive Cas9, which els and showed none of the electrical hyper- ing tool, but additional research is necessary has been fused to the catalytic domain of the activity of affected neurons. Similar results to evaluate if a hit-and-run approach would DNA methylcytosine dioxygenase TET1, were obtained if iPSC-derived neurons were work here.” to the hypermethylated repeats. Testing the treated after differentiation, although in this Markus Elsner constructs in patient-derived induced pluri- case demethylation and expression restora- Senior editor potent stem cells (iPSCs), they observed a tion remained incomplete. The authors also 96% reduction in the methylation levels of tested whether reactivation of FMR1 is main- the repeats and an almost complete restora- tained in vivo by transplanting methylation- 1. Rebar, E.J. et al. Nat. Med. 8, 1427–1432 (2002). 2. Thakore, P.I., Black, J.B., Hilton, I.B. & tion of FMR1 expression. “A surprising and edited neural precursor cells into the brains Gersbach, C.A. Nat. Methods 13, 127–137 (2016). important finding was that the reversal of of mice. After three months, about half of the 3. Liu, X.S. et al. Cell 172, 979–992.e6 (2018). NATURE BIOTECHNOLOGY VOLUME 36 NUMBER 4 APrIl 2018 315.