Despite rapid progresses in the genome-editing field, in vivo simultaneous overexpression of multiple genes remains challenging. We generated a transgenic mouse using an improved dCas9 system that enables simultaneous and precise in vivo transcriptional activation of multiple genes and long noncoding RNAs in the nervous system. As proof of concept, we were able to use targeted activation of endogenous neurogenic genes in these transgenic mice to directly and efficiently convert astrocytes into functional neurons in vivo. This system provides a flexible and rapid screening platform for studying complex gene networks and gain-of-function phenotypes in the mammalian brain.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
We thank D. Li, E. Zuo, Y. Shi, Y. Liu, J. He, J. Pan, Y. Zhong, Y. Lu, Y. Zhang, J. Yang and X. Tang for technical assistance and valuable discussion. This work was supported by National Science and Technology Major Project (2017YFC1001302), CAS Strategic Priority Research Program (XDB02050007, XDA01010409), the MoST863 Program (2015AA020307), NSFC grants (31522037, 31500825, 31571509, 31522038), China Youth Thousand Talents Program (to H.Y.), Break through project of Chinese Academy of Sciences, Shanghai Sailing Plan for the Young Scientific Talents (15YF1414700), and The Ministry of Science and Technology of China (MOST; 2016YFA0100500).
Integrated supplementary information
(a) mCherry reporter is driven by a minimal CMV promoter and activated by an upstream sgRNA target site. Note that miniCMV is a weak promoter, thus the basal fluorescence level is very low. See also18. (b) BFP was co-transfected to control the transfection efficiency (n = 2 cell cultures). Scale bar: 100 μm.
(a) Schematic of the vectors used to evaluate SPH-based activation. Cre recombinase was used to remove the LSL cassette in the SPH vector. (b) Representative images show that co-transfection of sgRNA-miniCMV, SPH and Cre led to enhanced expression of mCherry (n = 4 cell cultures, 4 images). mCherry was activated in 93 ± 1% of the cells expressing SPH activator system. All values are presented as mean ± s.e.m.
Supplementary Figure 3 SPH shows more potent activation efficiency compared to SAM and VPR in primary astrocytes.
(a) Schematic illustration of all plasmids. The expressions of all functional elements were probed by the fluorescence reporter mCherry or GFP. (b) 2 days after transfection, cells were prepared for FACS. The mCherry and GFP double-positive cells were collected for qPCR. (c) SPH achieved the highest level of activation, which supports the observations in cell lines (n = 2 cell cultures).
(a) Schematic of the integration sites of SPH in a SPH transgenic mouse genome. (b) Genotyping of SPH transgenic mice. “+” indicates positive mice. Percentage of positive progenies was calculated on 101 mice. Gel image was cropped. (c) Systemic evaluation of the Alb-Cre effectiveness in cleaving LSL cassettes specifically in the liver. Note that Alb is a liver-specific promoter (n = 1 mouse). Gel image was cropped. (d) Fluorescence images of the primary SPH fibroblasts infected by lentivirus containing sgRNA-EF1a-Cre-mCherry, confirming expression of dCas9 and activators were induced by Cre recombinase (n = 3 mice). (e) Expressions of dCas9 in Cre-expressing fibroblasts. Gel image was cropped. (f) Characterization of Cre effectiveness in the liver after hydrodynamic injection of the plasmids expressing Cre and mCherry (n = 1 mouse). (g) Stereotactic Injection of AAV-hSyn-mCherry-Cre to induce neuron-specific expression of EGFP (n = 1 mouse). Red arrowheads indicate primers. Scale Bars: d, f, g, 20 μm.
Supplementary Figure 5 Targeted activation of individual genes and lncRNA in the liver through tail vein injection of plasmids.
(a) Experimental design. Plasmids with Cre, sgRNA and mCherry expression were delivered into the liver by hydrodynamic injection via tail vein. Note that mRNA was extracted from the liver without isolating mCherry-positive cells and used for qPCR analysis (n = 1 mouse). Scale Bar: 100 μm. (b) Targeted activation of individual genes in the liver after injection of plasmids expressing indicated sgRNAs (number of mice denoted within graph). Increased amount of ASCL1 protein was confirmed by western blot (n = 1 mouse). Gel images were cropped. (c) Induction of lncRNAs in the liver (n = 3 mice). (d) Activation of multiple genes in the liver through tail vein injection of a mixture of sgRNA-expressing vectors (n = 2 mice). The sgLacZ serves as a control sgRNA.
(a) Schematic of the plasmid used for sgDkk1 expression. (b,c) SPH mice were hyperdynamically injected with sgRNA-mCherry plasmids. Immunofluorescence staining were performed 4–6 days after plasmids injections on liver sections. mCherry (red) was stained to indicate cells receiving plasmids. Both GS and CYP2E1 (cyan) were β-catenin target genes locating inpericentral area of the liver lobule. Arrows indicate mCherry-positive cells showing downregulation of GS and CYP2E1. Note that expression of GS and CYP2E1 in pericentral hepatocytes receiving sgDkk1-Cre-mCherry was inhibited. Scale bar = 50 μm. (d) Increased Dkk1 transcripts were quantified by qPCR (n = 5 mice). (e, f) Percentage of cells showing suppressed expression of GS (88/439 VS 0/165, sgLacZ: n = 2 mice, sgDkk1: n = 2 mice) or CYP2E1 (127/829 VS 0/66, sgLacZ: n = 2 mice, sgDkk1: n = 5 mice) in plasmid-expressing hepatocytes. (g) Schematic illustration of DKK1-mediated suppression of Wnt/β-catenin signaling. Activation of Dkk1 leads to reduced expressions of two β-catenin target genes; green circles indicate mCherry-positive cells showing downregulation of GS and CYP2E1. The sgLacZ serves as a control sgRNA.
(a) Prescreening in primary fibroblasts derived from SPH mice (n = 2 cell cultures) and (b) N2a cells (n = 1 cell culture). The sgRNAs used in the sgRNA arrays are denoted as red. The sgLacZ serves as a control sgRNA.
Supplementary Figure 8 Simultaneous activation of 8 genes and 2 lncRNAs in mCherry-positive cells via hydrodynamic injection of all-in-one plasmids.
(a) Schematic of the all-in-one vector used for sgRNAs and Cre expression, and experimental flow of gene activations in the liver of SPH transgenic mice. Note that mRNA was extracted from mCherry-positive cells. (b) Simultaneous activation of 8 genes and 2 lncRNAs in the liver (n = 3 mice). The sgLacZ serves as a control sgRNA.
Supplementary Figure 9 Simultaneous activation of 10 genetic elements including 8 genes and 2 lncRNAs in the liver via hydrodynamic injection of dual-AAV vectors.
(a) Experimental flow of gene activations in the liver of SPH transgenic mice. (b) Schematic of the AAV-10-sgRNA-array and AAV-Alb-Cre. (c) Simultaneous activation of 8 genes and 2 lncRNAs (n = 2 mice). SPH mouse injected with vehicle serves as a control.
Primary astrocytes derived from the SPH mice were transfected with an all-in-one vector expressing Cre, mCherry and an sgRNA array targeting the promoters of 8 genes. Due to detection limit, 3 (Ascl1, Slc6a4 and Slc7a11) out of 8 genes (Neurog2, Slc6a4, Dkk1, Hbb, Slc7a11, IL10, Ascl1, Neurod1) were detected in single-cell RNA-seq data. The average expression levels (FPKM values) of Ascl1, Slc6a4 and Slc7a11 were normalized to control group (n = 3 cells). 6 cells receiving the all-in-one vector were analyzed. Control cells were introduced with sgLacZ constructs.
Supplementary Figure 11 Enhance transcription of Hbb and Dkk1 was confirmed by in situ hybridization.
Enhance transcription of Hbb and Dkk1 was detected in the cerebral cortex after injecting AAV-hSyn-Cre and AAV-sgDkk1-sgHbb (n = 1 mouse). Scale Bar: 20 μm. The contralateral cerebral cortex serves as control.