Researchers from the Wellcome Sanger Institute and collaborators used human stem cells and neurons to investigate what features influence how well CRISPR activation works for different sets of genes. Their study “Massively parallel characterization of CRISPR activator efficacy in human induced pluripotent stem cells and neurons,” which appears in Molecular Cell, was designed to explain the rules determining to what extent genes respond to CRISPR activation, ensuring that future research can be designed as efficiently as possible.
CRISPR activation (CRISPRa) is a type of CRISPR gene editing that is used to overexpress certain genes. Although this technique is broadly used, predicting its efficiency when aimed at certain points in the genome can be challenging, making it hard to reliably overexpress certain genes, according to the scientists.
Key tool for perturbing transcription
“CRISPR activation (CRISPRa) is an important tool to perturb transcription, but its effectiveness varies between target genes. We employ human pluripotent stem cells with thousands of randomly integrated barcoded reporters to assess epigenetic features that influence CRISPRa efficacy. Basal expression levels are influenced by genomic context and dramatically change during differentiation to neurons,” write the investigators.
“Gene activation by dCas9-VPR is successful in most genomic contexts, including developmentally repressed regions, and activation level is anti-correlated with basal gene expression, whereas dCas9-p300 is ineffective in stem cells. Certain chromatin states, such as bivalent chromatin, are particularly sensitive to dCas9-VPR, whereas constitutive heterochromatin is less responsive. We validate these rules at endogenous genes and show that activation of certain genes elicits a change in the stem cell transcriptome, sometimes showing features of differentiated cells.
“Our data provide rules to predict CRISPRa outcome and highlight its utility to screen for factors driving stem cell differentiation.”
The team uncovered multiple features that impact CRISPRa efficiency, including expression level, chromatin status, cell state, and gene location. They found that bivalent genes, which are key developmental regulator genes that have both repressing and activating marks in the same region, can be robustly activated by CRISPRa. They also discovered that CRISPRa could achieve the same overexpression levels that are necessary to drive significant changes in cell state and cause them to differentiate.
Team found a differential response
The team found that while most genes can be activated by CRISPRa, not every cell responded to the same extent. For example, genes that contain H3K9me3 repressing regulators, and therefore are shielded from activation, showed greater variation in response.
The researchers add that these data demonstrate that CRISPRa is generally applicable across chromatin states and cell types, and highlights the factors that impact the degree of gene activation and how easy it is to reproduce the effects. Understanding these factors is important in the design and analysis of CRISPRa screens, which are used to look for genes involved in genetic diseases, the team points out.
Further study is required to continue to add to these rules and to see whether different CRISPRa or CRISPR interference techniques behave in a similar way.
“Our research has established a system for reporting the effectiveness of CRISPR activation in stem cells, allowing us to gain a better understanding of how CRISPRa works in multiple cell states,” says Qianxin Wu, PhD, first author from Wellcome Sanger. “We also showed that CRISPR gene activation is powerful enough to induce stem cells to differentiate into other cell states. This suggests that CRISPRa screens can be used to search for genes involved in cellular processes or to generate more accurate models of cell types in the body, aiding research into genetic diseases and regenerative medicine.”
“By investigating the factors that impact CRISPRa efficiency in a systematic way, we have created a set of rules that show where it is most or least effective and deepened our understanding of the factors involved,” notes Andrew Bassett, PhD, senior author, also from the Wellcome Sanger Institute.
