Coordinated regulation by lncRNAs results in tight lncRNA–target couplings

The determination of long non-coding RNA (lncRNA) function is a major challenge in RNA biology with applications to basic, translational, and medical research. Our efforts to improve the accuracy of lncRNA-target inference identified lncRNAs that coordinately regulate both the transcriptional and post-transcriptional processing of their targets. Namely, these lncRNAs may regulate the transcription of their target and chaperone the resulting message until its translation, leading to tightly coupled lncRNA and target abundance. Our analysis suggested that hundreds of cancer genes are coordinately and tightly regulated by lncRNAs and that this unexplored regulatory paradigm may propagate the effects of non-coding alterations to effectively dysregulate gene expression programs. As a proof-of-principle we studied the regulation of DICER1—a cancer gene that controls microRNA biogenesis—by the lncRNA ZFAS1, showing that ZFAS1 activates DICER1 transcription and blocks its post-transcriptional repression to phenomimic and regulate DICER1 and its target microRNAs. CRISPRi screen followed by RNA-Seq: We used a high-throughput parallel CRISPRi screening platform that combines live-cell imaging with a scalable RNA-seq workflow to generate unbiased analyses of lncRNA regulation. For each lncRNA target of interest, a pool of up to 10 single-guide RNAs (sgRNAs) was produced by high-throughput in vitro transcription of sgRNA templates generated by multiplex PCR. The resulting sgRNA pools were delivered by electroporation to HEK293T cells with stable dCAS9-KRAB expression. QuantSeq RNA-seq library preparation (Lexogen) was performed according to the manufacturer’s protocol using 5 µl of cell lysate as input. Libraries were quantified by qPCR, pooled, and sequenced on a NextSeq 500 System (Illumina). FASTQ files were processed using an in-house RNA-seq analysis pipeline. FastQC (v0.11.8) was first used for data quality control, after which adapter sequences, polyA readthrough, and low-quality reads were removed by BBMap v38.26. Reads were then mapped against the hg38 reference genome with STAR v2.6.0c, and gene counts were determined by HTSeq v.0.11.0. The number of reads for each gene was adjusted to account for differences in sequencing depth and presented as counts per million (CPM). Comparisons were made to plate-specific pooled negative-control NT sgRNAs. siRNA followed by RNA-Seq: Three human cell lines, ECC-1, NCI-H460, and PC-3, were transfected with 25 nM siRNAs targeting ZFAS1 and DICER1 for 24 hours in 96-well plates. Quadruplicate wells containing 10,000 cells each were pooled for each sequenced sample (N = 18: 6 per cell line, non-targeting control (NT), siZFAS1, and siDICER1, each in duplicate). Subsequently, aliquots from each sample were submitted to Novogene Corporation Inc. (Sacramento, CA) for library preparation and RNA sequencing (NovaSeq). Libraries were prepared using a polyA-selected approach and yielded over 20 million high-quality, 150bp paired-end reads per sample. RNA-seq raw reads were aligned to the hg19 reference genome with GENCODE v16 gene annotation using STAR v2.3.0e. Alignment files were processed with Picard tools v1.54 (http://broadinstitute.github.io/picard/) and indexed using SAMtools v0.1.11. Transcript quantification was performed in Cufflinks v2.02 quantification mode with the GENCODE v16.gtf annotation. Transcripts Per Million (TPM) values were used for relative abundance estimation.