Precise measurement of molecular phenotypes with barcode-based CRISPRi systems

Abstract

Genome-wide CRISPR-Cas9 screens have untangled regulatory networks and revealed the genetic underpinnings of diverse biological processes. Their success relies on experimental designs that interrogate specific molecular phenotypes and distinguish key regulators from background effects. Here, we realize these goals with a generalizable platform for CRISPR interference with barcoded expression reporter sequencing (CiBER-seq) that dramatically improves the sensitivity and scope of genome-wide screens. We systematically address technical factors that distort phenotypic measurements by normalizing expression reporters against closely-matched control promoters, integrated together into the genome at single copy. To test our ability to capture post-transcriptional and post-translational regulation through sequencing, we screened for genes that affected nonsense-mediated mRNA decay and Doa10-mediated cytosolic protein decay. Our optimized CiBER-seq screens accurately capture the known components of well-studied RNA and protein quality control pathways with minimal background. These results demonstrate the precision and versatility of CiBER-seq for dissecting the genetic networks controlling cellular behaviors.

Figure 1:. Eliminating background in barcode-based genetic screens.

(A) Schematic of paired guide-barcode libraries for CRISPRi screening. (B) Workflow for CiBER-seq screen and determination of phenotypic effects. Sequencing libraries are prepared from samples before and after guide induction, normalizing RNA barcodes to DNA barcodes. (C) An idealized CiBER-seq platform using closely matched transcription factors that each express a barcode from similar promoters with common genetic dependencies. (D) Characterization of the dose response of each hormone-inducible transcription factor by flow cytometry of yeast transformed with a YFP expressed from the cognate promoter. Mean YFP was fit to a simple binding isotherm for biological replicates, see Methods (n=2). (E) Cross reactivity of Z3PM or Z4PM expressing YFP from a P(Z3) or P(Z4) promoter. YFP expression is normalized to the transcription factor and its cognate promoter (n=2). (F) Schematic for evaluating technical variations between DNA and RNA barcode-based comparisons. All barcodes are isolated from the same sample before and after guide induction. (G) Analysis of genome-wide CiBER-seq screen with Z3PM RNA barcodes normalized to DNA barcodes levels. Each point is a single guide, with significant guides colored red. A q-value < 0.01 and > 1 log₂-fold change and is represented by dashed lines. (H) Same screen as in (G), except Z3PM barcode expression was normalized to the control barcodes driven by Z4PM.

Figure 2:. High-efficiency, single-copy integration of reporters with a Bxb1 recombinase-based system.

(A) Schematic of yeast-based Bxb1 integration system. Bxb1 is constitutively expressed until recombined with a donor plasmid, which reconstitutes a URA3 selectable marker. (B) Flow cytometry of yeast transformed with a mixture of plasmids encoding yECitrine or yEmScarlet through plasmid-based or Bxb1-mediated recombination approaches. Fluorescence was recorded both before and after removal of selective pressure. (C) Distribution of yECitrine fluorescence of yeast transform through different approaches (D) Transformation efficiency of plasmid-based or Bxb1-mediated recombination approaches (n=3).

Figure 3:. Precise identification of degron regulators with optimized CiBER-seq.

(A) Schematic for CiBER-seq to characterize regulators of the CL1 degron. (B) RT-qPCR of reporter transcripts expressed by Z3PM or Z3PM-CL1 in a wildtype or ∆doa10 background, compared to the normalizer transcript (n=3). (C) Analysis of genome-wide CiBER-seq screen for regulators of the CL1 degron. Each point is a single guide and colored based on molecular function in legend. Significant and robust guides were assessed by a q-value < 0.01 and > 2 log₂-fold change, which is represented by dashed lines. (D) Schematic of established CL1 turnover pathway with yeast protein names displayed. (E) RT-qPCR of reporter barcodes expressed by Z3PM-CL1 with guides induced (n=3).

Figure 4:. CiBER-seq directly measures regulators of an RNA quality control pathway.

(A) Schematic for using CiBER-seq to interrogate RNA-level phenotypes. Barcodes are embedded in the 3′ UTR of a reporter and normalizer, which are constitutively expressed from identical promoters. (B) Proteins involved in NMD, with yeast gene names displayed. (C) Workflow for CiBER-seq to investigate regulators of NMD and their dependency on active translation. (D) RT-qPCR of the NMD reporter, compared to the normalizer transcript without the PTC, at all steps during CiBER-seq (n=2). (E) Analysis of genome-wide CiBER-seq screen for regulators of NMD. Guides with a q-value < 0.01 and > +1 log₂-fold change are labelled in red, with thresholds represented by dashed lines. (F) RT-qPCR of PTC-containing mRNA compared to normalizer with guides induced (n=2–3).