CRISPR: New Frontiers

Clustered, regularly interspaced, short palindromic repeat (CRISPR) loci and their associated (Cas) proteins provide adaptive immunity against viral and plasmid infection in prokaryotes. The CRISPR-Cas immune response can be divided into two phases. Upon infection, short phage or plasmid sequences known as spacers integrate between CRISPR repeats. This is known as the immunization stage. During the second phase, the targeting phase, spacers are transcribed into small RNA guides that identify the viral or plasmid targets of CRISPR immunity. The CRISPR RNA guides are loaded into Cas nucleases and direct them to complementary sequences in the invading genome. Cleavage of the target genome results in the end of the infection. I will discuss recent work in my lab which shows how these two phases of the CRISPR-Cas immune response are fundamentally linked.

Canonical CRISPR–Cas systems maintain genomic integrity by leveraging guide RNAs for the nuclease-dependent degradation of mobile genetic elements, including plasmids and viruses. Here we describe a remarkable inversion of this paradigm, in which bacterial Tn7-like transposons have co-opted nuclease-deficient CRISPR–Cas systems to catalyze RNA-guided integration of mobile genetic elements into the genome. Programmable transposition of Vibrio choleraeTn6677in Escherichia coli requires CRISPR- and transposon- associated molecular machineries, including a co-complex between the DNA-targeting complex Cascade and the transposition protein TniQ. Integration of donor DNA occurs in one of two possible orientations at a fixed distance downstream of target DNA sequences, and can accommodate variable length genetic payloads. Deep-sequencing experiments reveal highly specific, genome-wide DNA insertion across dozens of unique target sites. This discovery of a fully programmable, RNA-guided integrase lays the foundation for genomic manipulations that obviate the requirements for double-strand breaks and homology-directed repair.

Conventional functional genomics approaches enable scalable investigation of gene function and have been critical for elucidating molecular pathways. CRISPR-based genetic technologies, including those for genome editing and gene expression modulation, have enhanced these efforts by improving and expanding our ability to make genetic perturbations in human cells. Now, recent advances allow high-resolution phenotyping of CRISPR-based screens, enabling complex readouts to be generated for thousands of genetic perturbations in single experiments. These phenotypes provide rich information about the effects of individual perturbations on cells and can therefore be used for sophisticated analyses, including functional classification of interrogated genes and delineation of the genetic pathways that control complex or rare events.We have recently built a high-resolution screening platform to systematically map the many integrated pathways of DNA repair and are using this platform to study genome editing.

Forward genetic screens using CRISPR–associated nucleases like Cas9 are a powerful tool to pinpoint genes involved in disease. Initial screens capitalized on genome-scale libraries to perturb nearly all protein-coding genes in the human genome to examine therapeutic resistance in cancer and gene essentiality. Recently, several engineered Cas9 variants with relaxed target sequence requirements have been developed. To understand how they compare with the widely-used S. pyogenes Cas9, we developed a novel pooled genetic screen with multiple Cas9 variants in the same pool and evaluated how these variants performed at gene knock-out, inhibition and activation. We found that flexibility in target sequences comes at a price of reduced genome editing performance and construct a new Cas9 variant with improved gene activation at a broad range of target sites.

In addition to genome-scale screens targeting protein-coding genes, we have recently adapted CRISPR screens into noncoding regions of the genome to identify cell type-specific enhancers and other functional elements that control gene expression. Despite these advances, high-throughput screens of noncoding RNAs have been more challenging since small mutations are unlikely to abrogate function. For this reason, we recently developed a RNA-targeting, Cas13-based platform for massively-parallel forward transcriptomic screens in human cells. We compared several Cas13 orthologs and, using ~25,000 Cas13 guide RNAs, we determined optimal parameters for target RNA knockdown with Cas13. Using protein-coding transcripts to evaluate performance, we find that gene essentiality from Cas13-based screens compares well with known essential genes from prior screens using Cas9 and RNA interference.

A simple benefit of single-cell methods is that each cell can serve as a tiny experiment, enabling massive parallelization. For example, Perturb-seq pairs single-cell RNA sequencing with an approach for capturing the identity of CRISPR-mediated genetic perturbations, enabling thousands of genotype-phenotype measurements to be made in pooled format.These rich experiments have great potential for understanding how phenotypes arise from the genes controlling them.As an example, we applied Perturb-seq to understanding genetic interactions, which explain why perturbing a pair of genes together differs phenotypically from perturbing either gene alone. To identify candidate interactions, we activated ~22,000 pairs of genes in K562 cells using CRISPRa and scored them for unexpectedly strong fitness defects. We then profiled ~300 strong interactions via Perturb-seq. The resulting transcriptional profiles yielded several insights. First, they allowed us to identify biological mechanisms underlying interactions, such as gene pairs inducing cell differentiation. Second, we created a simple mathematical model of interaction that enabled us to classify interactions (e.g. identifying suppressors). Finally, using a mathematical technique called matrix completion we could computationally predict interactors using far fewer experiments, suggesting a strategy for searching vast spaces of genetic interactions that would be inaccessible experimentally.

The mechanisms through which genetic variants interact to contribute to complex genetic disorders are unclear. Integrated genetic and transcriptomic studies have identified candidate genes, whose expression is affected by SZ risk variants. For the functional validation of such variants and genes, we established a genetics-driven,human induced pluripotent stem cell (hiPSC)-based approach, integrating various CRISPR technologies.By utilizing CRISPR-mediated allelic conversion of one putative causal variant, our isogenic platform recapitulated genotype-dependent gene expression differences as well as downstream effects on neuronal activity. Further integration of CRISPR activation and interference to perturb SZ-genes in the disease-relevant direction, individually and in combination,suggested considerable downstream effects specific to a combinatorial perturbation of SZ-genes. Modulation of the genes in concert, but not individually, mirrored transcriptional changes in postmortem brain analyses of SZ. Downstream genes that were affected more than predicted in an additive model,were significantly enriched for synaptic and disorder signatures, including both rare and common risk variant genes.

These synergistic effects emphasize the complex polygenic nature of SZ, where a combination of variants contributes to disease. We propose that the links between rare and common variants constitute a potentially generalizable phenomenon occurring more widely in complex genetic disorders.

T cells became the central focus of new cancer therapeutics. Immune checkpoint inhibitors targeting T cell signaling pathways and cellular therapeutics utilizing chimeric antigen receptors (CARs) have shown success in the clinic. Discovery of previously unknown genes that modulate T cell function is of urgent need to open different avenues for immunotherapies, as a large fraction of patients still do not respond to, or have undesired side effects to currently approved ones. Systematic approaches to identify new regulators of T cell functions in vivo can provide potentially orthogonal or complementary opportunities. We recently performed in vivo CRISPR screens in CD8 cytotoxic T cells, both in a genome-scale and in a focused manner, in tumor models of immunotherapy (Dong et al. 2019 Cell). Furthermore, we developed a novel AAV-SleepingBeauty system that enhanced the power of genetic screening in primary T cells (Ye et al. 2019 Nature Biotechnology). Our screen re-discovered prime immunotherapy targets such as PD-1, TIM-3 and LAG3, as well as previously undocumented targets. We characterized novel targets such as DHX37, ODC1, MGAT5 and PDIA3 for their activity in mouse and human CD8 T cells including CAR-Ts.

Current major types of immunotherapy include checkpoint blockade, adoptive cell transfer, human recombinant cytokines, and cancer vaccines. However, immunotherapy has met challenges in immunologically cold tumors. These challenges urge for new types of immunotherapies that are more potent and potentially less toxic. Recently, we have developed CRISPRa-mediated Multiplexed Activation of Endogenous Genes as an Immunotherapy (MAEGI) (Wang et al. 2019 Nature Immunology, in press). The CRISPR activation (CRISPRa) system uses a catalytically inactive Cas9 (dCas9), enabling simple and flexible gene expression regulation through dCas9-transcriptional activators paired with single guide RNAs (sgRNAs). This enables precise targeting of large gene pools of endogenous genes in a flexible manner. We demonstrate that MAEGI has therapeutic efficacy across three tumor types. Mechanistically, our preliminary work showed that MAEGI treatment elicits anti-tumor immune responses by recruiting effector T cells and remodeling the tumor microenvironment. We will perform advanced development, characterization and optimization of MAEGI, as a novel immune-gene therapy approach to elicit a potent and specific immune response to tumors based on their unique genetic composition.

Detailed characterization of molecular effects of genetic variants is essential for understanding biological processes that underlie genetic associations to disease driven by noncoding regulatory variants. One of the earliest approaches to characterize regulatory effects has been population mapping of genetic variants associating to gene expression and other transcriptome traits. The Genotype Tissue Expression project is the largest effort of this type, with the final release having 17,382 RNA-sequencing samples from 54 tissues of 948 post-mortem human subjects. In this talk, I will describe the comprehensive GTEx catalog of genetic associations for gene expression and splicing, or eQTLs and sQTLs, and characterize their shared and distinct molecular mechanisms across tissues and cell types that mediate complex traits. Furthermore, I will describe genome editing approaches to follow up eQTLs discovered in population studies from an experimental angle. This includes validation of regulatory and nonsense variants in GTEx, as well as experimental follow-up of an autoimmune disease -associated locus close to the IRF1 gene. Altogether, our integration of population genetic and experimental approaches to analyze functional variation provides interesting insights into their respective strengths and limitations, and the future potential for a unified analytical framework.

Proteins must be maintained in specific three-dimensional conformations to ensure functional fidelity and to prevent aberrant and potentially toxic interactions with cellular machinery. Subcellular compartments including the cytoplasm, ER, and mitochondrion have elegant mechanisms to identify misfolded proteins and target them for re-folding or clearance. In contrast, little is known about how the mammalian nucleus senses and responds to misfolded proteins, in spite of the numerous incurable diseases associated with the accumulation of misfolded proteins into nuclear inclusion bodies. To address this, we developed cell culture models of a mutant Huntingtin allele that forms nuclear inclusion bodies as protein quality control becomes compromised. We leveraged a FACS-based method to perform genome-wide CRISPR/Cas9 knockout screens in this model, and identified genes that modulate the frequency of nuclear inclusion bodies. To better distinguish real signal from noise, we developed a simple but effective way to reduce the impact of mean-dependent variance – a particular challenge in marker-based screens where ‘real’ hits may only be expected to produce a modest fold-change in the relevant phenotype, or where the phenotype of interest is rare. This analysis strategy recapitulated known regulators of nuclear protein degradation and uncovered two particularly promising novel hits that seem to act in concert to regulate protein accumulation in the nucleus.

Coauthors: Gregory Cajka, Yevgeniy Serebrenik, and Ophir Shalem, University of Pennsylvania and Children’s Hospital of Philadelphia.

The development of CRISPR-Cas systems for genome editing has transformed the way life science research is conducted and has already successfully been used to treat patients. The initial demonstration of Cas9-mediated genome editing launched the development of many other technologies, enabled new lines of biological inquiry, and motivated a deeper examination of natural CRISPR-Cas systems, including the discovery of new types of CRISPR-Cas systems. These new discoveries in turn spurred further technological developments, the pace of which has been accelerated by the open and innovative community of researchers in the CRISPR community. As genome editing technology advances, we must also collectively consider the prudent implementation of these tools to ensure the greatest benefit to the world.