LAUSR

Dear Editor, The majority of mammalian genome is transcribed to RNA transcripts, of which only a very small percentage code for proteins. As a result, thousands of RNAs that do not code for proteins are produced in cells, including microRNAs (miRNAs) and long noncoding RNAs (lncRNAs). These noncoding RNAs exert regulatory functions in various physiological and pathological conditions. In addition, numerous noncoding RNAs are expressed in a tissueand cell-specific manner. Thus, a reporter that faithfully reflects the expression or activity of noncoding RNAs can provide useful tools not only for uncovering the regulators of noncoding RNAs, but also for tracking cell fate and disease status. Previously we have designed a miRNA inducible CRISPR-Cas9 platform that can serve as a sensor for miRNA activities. However, designing a reporter for long noncoding RNAs has not been easy due to its untranslatable nature and low expression level. Here, we design an sgRNA precursor in an intron (GRIT) strategy that can monitor the promoter activity of lncRNAs (Fig. 1a). Furthermore, we show that GRIT can be used to track differentiation status of stem cells. The design of GRIT includes three key elements (Fig. 1a): dCas9-VPR expressed under the control of a constitutively active CAGGS promoter, an RFP gene under the control of a tetracycline-inducible promoter (TRE), and a transfer RNA (tRNA)-flanked sgRNA that is integrated in an endogenous noncoding RNA locus through homologous recombination. To minimize the impact of tRNA-sgRNA knock-in on lncRNAs, we chose genome region that will be expressed as an intron to knock-in tRNA-sgRNAs. In addition, for lncRNA gene without an intron, an artificially designed mini-intron-containing tRNA-sgRNA fusion sequence was knocked in. The tRNA flanking design was chosen based on our observation that tRNA-flanked sgRNA induced higher level of RFP expression when compared to ribozyme-flanked sgRNAs (Fig. 1b, c; Supplementary Fig. S1a-c and Table S1). We then knocked the tRNA-flanked sgRNA into the second intron of Lncenc1 in mouse embryonic stem cells (ESCs) in which CAGGS-dCas9-VPR and TRE-RFP have been transgenically integrated (Fig. 1d; Supplementary Table S1). Lncenc1 is a lncRNA specifically expressed in mouse ESCs. In ESCs with GRIT successfully integrated (Lncenc1-GRIT ESCs), we observed high level of RFP expression (Fig. 1e, f). In addition, the knock-in of tRNAsgRNA have little effect on the expression of Lncenc1 and pluripotency genes including Nanog, Oct4 (also known as Pou5f1) and Sox2 (Supplementary Fig. S2a, Tables S2 and S3). Importantly, the transcription activity of Lncenc1 locus was found not affected based on qPCR analysis of precursor RNA of Lncenc1 (Supplementary Fig. S2a). Lncenc1 is downregulated during ESC differentiation. To check whether GRIT can report the expression of Lncenc1 during ESC differentiation, we induced differentiation of Lncenc1-GRIT ESCs with all-trans retinoid acids (ATRA) and measured RFP expression during differentiation process. Interestingly, RFP was significantly decreased upon ATRA induced differentiation (Fig. 1g, h; Supplementary Fig. S2b). More importantly, RFP level was highly correlated to the RNA level of Lncenc1 (Supplementary Fig. S2c).