CRISPR-mediated modular RNA-guided regulation of transcription in eukaryotes

Abstract

The genetic interrogation and reprogramming of cells requires methods for robust and precise targeting of genes for expression or repression. The CRISPR-associated catalytically inactive dCas9 protein offers a general platform for RNA-guided DNA targeting. Here, we show that fusion of dCas9 to effector domains with distinct regulatory functions enables stable and efficient transcriptional repression or activation in human and yeast cells, with the site of delivery determined solely by a coexpressed short guide (sg)RNA. Coupling of dCas9 to a transcriptional repressor domain can robustly silence expression of multiple endogenous genes. RNA-seq analysis indicates that CRISPR interference (CRISPRi)-mediated transcriptional repression is highly specific. Our results establish that the CRISPR system can be used as a modular and flexible DNA-binding platform for the recruitment of proteins to a target DNA sequence, revealing the potential of CRISPRi as a general tool for the precise regulation of gene expression in eukaryotic cells.

Figure 1. A modular CRISPR fusion system for efficiently repressing and activating transcription in human cells

(A) dCas9 fused to effector domains can serve as an RNA guided DNA binding protein to target any protein to any DNA sequence. (B) A minimal CRISPRi system in human cells contains an sgRNA expression plasmid and dCas9 or dCas9 fused to the repressive KRAB effector domain. Both dCas9 constructs are fused to two copies of a nuclear localization sequence and a blue fluorescent protein. (C) A dCas9-KRAB fusion protein efficiently silences GFP expression. 8 sgRNAs targeting GFP are transfected into GFP+HEK293 stably expressing either dCas9 (light grey) or dCas9-KRAB (dark grey). GFP fluorescence is quantified by flow cytometry 6 days following transfection and displayed as signal normalized to a vector control. The data is displayed as mean ± standard deviation for 3 independent experiments. See also Figure S1. (D) CRISPRi gene repression is stable over time. GFP+HEK293 cells were infected with lentivirus constructs expressing a negative control sgRNA or a sgRNA targeting GFP and either dCas9 or dCas9-KRAB. Cells were grown for 9 days and then analyzed for GFP expression. A histogram displays GFP fluorescence for each sample and a control population of HEK293 cells which do not express GFP. Data are representative of 3 independent experiments. (E) Two dCas9 fusion proteins were constructed with VP64 or p65AD. The sgRNA is expressed as before. A diagram of the Gal4 UAS-GFP reporter and data showing transient transfection of either dCas9-VP64 or dCas9-p65AD and sgGAL4–1 can activate gene expression in HEK293 cells. Cells were transfected with the indicated plasmids and 48 hours later analyzed by flow cytometry for GFP expression. The data are displayed as mean ± standard deviation for 2 independent experiments. See also Figure S2.

Figure 2. CRISPRi is highly specific in human cells

(A) RNA sequencing RPKM (reads per kilo base per million) are plotted for GFP+ HEK293 cells stably expressing dCas9-KRAB and either a negative control sgRNA targeting Gal4 or an sgRNA targeting GFP. Total RNA was collected 15 days following lentiviral transduction. The data are representative of 2 independent biological replicates. See also Figure S3. (B) A histogram showing the fold changes in gene expression from (A) plotting sgGFP-NT1 over the negative control. The GFP is indicated with an arrow. The data are representative of 2 independent biological replicates. See also Figure S3.

Figure 3. CRISPRi can stably suppress expression of endogenous eukaryotic genes

(A) A diagram of sgRNA binding sites for CD71 and a graph displaying suppression of CD71 by dCas9 or dCas9-KRAB in HeLa cells. Cells stably expressing dCas9 or dCas9-KRAB were transfected with sgRNAs targeting CD71. After 72 hours, cells were harvested and analyzed by flow cytometry for CD71 protein expression. The data are displayed as mean ± standard deviation for 2 independent experiments. See also Figure S4A. (B) A diagram of sgRNA binding sites for CXCR4 and a graph displaying suppression of CXCR4 by dCas9 or dCas9-KRAB in HeLa cells. Cells stably expressing dCas9 or dCas9-KRAB were transfected with sgRNAs targeting CXCR4. After 72 hours, cells were harvested and analyzed by flow cytometry for CXCR4 protein expression. The data are displayed as mean ± standard deviation for 2 independent experiments. See also Figure S4A. (C) Stable suppression of CD71 by dCas9 or dCas9-KRAB in HeLa cells. Cells stably expressing dCas9 or dCas9-KRAB were transduced with sgRNAs targeting CD71. After 7 days, cells were harvested and analyzed by flow cytometry for CD71 protein expression. The data are displayed as mean ± standard deviation for 2 independent experiments. (D) Stable suppression of CXCR4 by dCas9 or dCas9-KRAB in HeLa cells. Cells stably expressing dCas9 or dCas9-KRAB were transduced with sgRNAs targeting CXCR4. After 8 days, cells were harvested and analyzed by flow cytometry for CXCR4 protein expression. The data are displayed as mean ± standard deviation for 2 independent experiments. (E) Double knockdown of both CD71 and CXCR4 using sgRNAs targeting CD71 and CXCR4 in HeLa cells that stably express dCas9-KRAB. Cells were dissociated and stained for CD71 and CXCR4 expression 5 days after transfection. The data are displayed as mean ± standard deviation for 2 independent experiments. See also Figure S4B. (F) Robust repression of endogenous genes in yeast. A bar graph shows fluorescence intensity of a strain expressing TEF1-GFP transformed with the indicated sgRNA and dCas9 or dCas9-Mxi1. The two dotted lines indicate fluorescent signals from untagged yeast cells that do not express GFP or TEF1-GFP yeast cells without dCas9. The data are displayed as mean ± standard deviation for 3 independent experiments. See also Figure S4C.

Figure 4. CRISPRi can suppress gene expression from the promoter of a gene

(A) A dCas9-KRAB fusion protein efficiently silences GFP expression when targeted to the SV40 promoter of an SV40-GFP reporter. 6 sgRNAs targeting different regions of the SV40 promoter as indicated are transfected into GFP+HEK293 cells stably expressing either dCas9 (upper panel) or dCas9-KRAB (lower panel). GFP fluorescence is quantified by flow cytometry 6 days following transfection and displayed as signal normalized to a vector control. The data are displayed as mean ± standard deviation for 3 independent experiments. (B) The dCas9 construct and an sgRNA (sgTET) construct were co-transformed into a yeast strain expressing a TetON-Venus reporter and the rtTA protein. Doxycycline was added to cells expressing rtTA alone or rtTA, dCas9 and sgTET. The data are displayed as mean ± standard deviation for 3 independent experiments.