Read the summary and answer questions in bold below:
In this post, weâll be focusing on Tyler Jacks and his graduate student Francisco J. Sánchez-Riveraâs review, âApplications of the CRISPR-Cas9 system in cancer biology,â circa 2015 in Nature Reviews. So we understand the context of the authors, Jacks is the director of the Koch Institute for Integrative Cancer Research at MIT. His lab has a large mouse experimental component (MIT 2017). This sets the stage for the angle weâll be reading from. In all, itâs a nice roundup of his and colleagues' cancer applications of CRISPR-Cas9 with great graphics, but leaves a bit desired in crossing the bridge from mouse to human.
Briefly, the CRISPR-Cas9 system stems from the prokaryotic immune system, where the RNA-guided DNA endonuclease Cas9 uses a single guide RNA (sgRNA) containing a CRISPR RNA (crRNA) and trans-activating crRNA (tracrRNA) that binds Cas9 and directs it to the genomic sequence of interest, adjacent to a photospacer adjacent motif (PAM). Cas9 causes a precisely-directed double-stranded break (DSB) in the DNA, that gets repaired via non-homologous end joining (NHEJ) or homology-directed repair (HDR).
With NHEJ, this can disrupt the gene by creating Indels: frameshifts and stop codons. In cancer, individual oncogenes or tumor suppressors can be targeted, or drug response modulators, or synthetic lethality genes. With a multiplexed knockout approach, combinatorial vulnerabilities and epistatic relationships can be assessed.
For gene modification via HDR, target single or double-stranded DNA can be introduced, resulting in precise gene knock-ins, with potential effects of introducing point mutations to mimic cancers, fluorescent reporters/synthetic tags can be created, conditional (Cre-lox/Flp-FRT) constructs can used to turn off/on a mutation, non-coding elements like enhancers/insulators/promoters can turn on proto-oncogenes, and ultimately, HDR can be used for gene correction.
With Cas9 and 2 sgRNAs, there can be genomic rearrangements, leading to deletions, inversions (like EML4-ALK), and translocations (like BCR-ABL).
Dead Cas9 (dCas), which lacks its endonuclease activity via mutation to its HNH and RuvC domains, can be used for reversible, transcriptional repression or activation of coding and non-coding genes, or chromatin modification, via an effector.
Much of the practical applications of CRISPR-Cas9 is in ES cells and mouse models, both germinal and somatic (GEMMs and nGEMMs). With this technology, a plethora of GEMM/nGEMM models of cancer can be created, some with the ability to turn on/off the Cas9 system, and multiplexed sgRNAs can help gain further understanding of multi-hit cancers. Patient derived xenografts (PDX) mice can be assayed with CRISPR-Cas9 to determine different treatment efficacies. Entire libraries of sgRNA can be screened in cancer cell lines for enrichment/depletion with Cas9, and dCas9-activators/repressors can be used for complementarity, both of which providing insights into drug targets, synthetic lethality, and chemotherapy modulators.
The delivery mechanisms for the system are diverse, and include plasmids, lentivirus, adenovirus, nanoparticles, hydrodynamic gene transfer, and protein-based. The authors also cite integrating Cas9 into the GEMMs, so itâs one less thing needed in the delivery vehicle. This of course would not be applicable to human.
Ultimately, human cancer treatment is the goal. Yes, we know the boundaries between ES/mice experimental models and human embryos have already been broken, with an international team targeting the autosomal dominant mutation in MYBPC3, which causes cardiac myopathy (Ledford 2017). But the current compromise position of CRISPR-Cas9 technology toward therapeutics and treatment, beyond basic science understanding, is using primary cells, organoids, PDX, or personalized GEMMs to create genetic and drug screens.
In theory, what would it take to bring CRISPR-Cas9 to the ultimate, targeted gene therapy via either NHEJ or HDR, in human cancer? Are there cancers that are better bets for CRISPR-Cas9 than others, delivery vehicles that are safer or better optimized? Is CRISPR-Cas9 really âon targetâ enough to introduce into a person, and how do we best assess?
