Nan Cher Yeo and colleagues in the Church lab report in Nature Methods the identification of a combination of repressor domains which increases gene repression and could be widely useful.
The CRISPR-Cas system is widely used as a powerful gene editing tool thanks to the easily reprogrammable feature of the gRNA. In addition, a catalytically inactive Cas9 (dead Cas9, dCas9) where mutations specifically inactivate the endonuclease domain, without affecting the reprogrammable gRNA-dependent DNA binding, can be used to modulate or silence gene expression.
dCas9 alone, or dCas9 fused to effector domains such as KRAB, targeted to a window around the transcription start site (TSS) can repress gene expression by interfering with the transcriptional machinery or due to recruitment of chromatin-modifying factors.
In an attempt to improve the efficacy of CRISPR gene repressors, Yeo et al. identify a combination of repressor domains which increases gene repression. The authors first tested repressors in a EYEP fluorescent assay in HEK293T cells and of the more than 20 available repressor domains chose the strongest and fused it to dCas9 in a single or double conformation. Ultimately, they identified the bipartite dCas9-KRAB-MeCP2 combination to be the strongest (up to 60-fold stronger repression than dCas9 or dCas9-KRAB).
The authors then performed a thorough analysis of dCas9-KRAB-MeCP2 performance and advantages compared to dCas9 or dCas9-KRAB in singe gene silencing, multiplexed gene repression and genome-wide repression screen using gRNA libraries.
Exploiting a combinatorial repression screen using dual gRNA libraries targeting genes involved in DNA repair, the authors showed again how the improved dCas9-KRAB-MeCP2 performs better than previous repressor systems allowing the generation of genetic interaction mapping and clustering gene interaction in hierarchical heat maps.
Finally, the authors analysed the increased performance of their repressor in a synthetic gene circuits, directly repressing a fluorescent protein, by derepressing a TALE repressor or by deoxicyclin-inducible Pol-II repression. Interestingly, a functional three-layer synthetic gene circuit was built to demonstrate that dCas9-KRAB-MeCP2 can be used in complex genetic interactions more efficiently than dCas9 or dCas9-KRAB.
The work of Yeo et al. provide a very useful addition to the CRISPR repressor toolbox, at least in vertebrate research. Although not directly applicable to insects, the thorough investigation of efficacy of the newly generated repressor highlights the benefit and at the same time the limitations of CRISPR repressor systems, namely specificity and variation.
As for CRISPR-Cas9 nuclease, the specificity of the gRNA to limit its activity to the target sequence must be assessed, likewise for dCas9 repressors, off-target effects must be taken in consideration. In this respect, the authors proposed the use of high fidelity Cas9 variants or truncated gRNAs as obvious solutions to a known problem. In the second place, likewise for any CRISPR/Cas9 there are sets of gRNAs with limited or no activity, requiring the use of different gRNA arrays to target the same gene to maximize gene depletion.