The long process of drug discovery can take over a decade, from the identification of a target molecule, to release of a clinically approved drug that can be incorporated into therapeutic regimes. However, even after this long, convoluted, and financially risky process is complete, treatment strategies using the drug can be hindered by resistance, detrimentally impacting patient outcomes.
Drug resistance makes for an uncontrollable disease with limited treatment options and higher mortality. The mechanisms governing drug resistance are complex, and in many cases remain poorly understood. Cancer therapies in particular are susceptible to the development of resistance and subsequent patient relapse because the same mechanisms that drive cancer initiation and progression—uncontrolled proliferation, resisting cell death, avoiding growth suppression, and genomic instability—enable cancers to mutate rapidly and circumvent cytotoxic selection pressures applied by the drug. Elucidating the molecular mechanisms involved in drug resistance can therefore aid in the identification of optimal first-line therapies to avoid resistance, as well as having major implications for drug discovery for new therapeutic targets.
Enter CRISPR-Cas9
Functional genomic screening is a powerful tool for determining the relationship between phenotype and genotype and identifying molecular events that underpin processes such as drug resistance. The premise is simple—knock out gene expression and see which genes confer resistance to a drug. The CRISPR-Cas9 gene editing system allows for precise and specific gene knockout to be performed quickly, cost-effectively, and at high-throughput, enabling the entire genome to be screened at once.
The first successful genome-wide CRISPR-Cas9 knockout screens came within a year of its debut, from work carried out at the Broad Institute. In two papers published simultaneously, a pooled, genome-wide library of sgRNA was delivered by lentiviral vector to perform a systematic loss-of-function study, identifying genes conferring resistance to 6-thioguanine and BRAF-inhibitor vemurafenib.1,2 Following this, CRISPR-Cas9 genome-wide screens were successfully performed in a myriad of human, animal,and cancer cell-lines.3
Elucidating mechanisms of resistance
Identifying genes involved in drug resistance by genome-wide knockout screening is now a widely used application of the CRISPR-Cas9 technology, transforming the way in which we understand and question drug-gene interactions. By characterizing those genes involved in cellular mechanisms, we can further our understanding of the molecular basis of de novo and acquired resistance.
In addition to gene knockout, the gene-editing repertoire of CRISPR-Ca9 has been expanded with the development of CRISPRi and CRISPRa, where gene expression can be modulated using a with the development of CRISPRi and CRISPRa, where gene expression can be modulated using a catalytically inactive version of Cas9 (dCas9) tethered to a transcriptional repressor or activator domain. In some cases, knocking down gene expression with CRISPRi rather than complete ablation of gene expression by CRISPR-knockout better mimics the action of certain drugs, such as small molecule inhibitors, as well as reduces the incidence of false positives that arise when targeting amplified regions. CRISPRa also enables gain-of-function studies to be performed more easily compared to traditional cDNA overexpression libraries. This adaptation has given rise to a combined screening approach, where both CRISPRi and CRISPRa are used in parallel to identify those genes involved in both resistance and sensitivity.4
“Unbiased efforts to explore drug response have benefitted from outstandingly clean datasets, and transition of the discoveries to clinical application should now be much faster. Scanning for drug resistance with CRISPR ahead of clinical analysis in oncology drug development has started to become an almost obligatory exercise of due diligence for pharma companies—the data is that good.” Dr. Benedict Cross, head of functional genomic screening at Horizon Discovery
As of February 2019, a search for CRISPR screens on PubMed finds 541 publications published in the preceding five years. Here we highlight some recent examples where CRISPR screens have been used to elucidate mechanisms of drug resistance.
Mechanisms of endocrine therapy resistance
In a 2018 paper, Xiao et al.,5 utilized a genome-wide pooled lentiviral CRISPR-Cas9 screen to identify an estrogen-regulated negative feedback loop with CSK loss and PAK2 overexpression as the key components that drive resistance to endocrine treatment. Endocrine therapy, such as Tamoxifen, is used to treat estrogen receptor (ER) signaling that is activated in over 70% of breast cancers, but resistance is a common development. The discovery of a previously unknown feedback loop suggests a potential therapeutic target for combination strategies to combat both de novo and secondary resistance that could significantly improve patient outcomes.
Resistance in leukemia
A genome-wide CRISPR screen using the GeCKO library was performed in acute myeloid leukemia cells to identify mechanisms of resistance to Quizartinib (AC220), a potent and selective second-generation inhibitor of FLT3.6 Consequently, Hou et al. found that SPRY3 and GSK3, through which FGF/Ras/ERK and Wnt signaling is reactivated, are involved in resistance—a finding that was confirmed by subsequent analysis of patient samples from resistant tumors.
Identifying synthetic lethal relationships in cancer
Functional genomic screens have classically been used to aid the discovery of predictive biomarkers for clinical response and identify synthetic vulnerabilities of cancer cells for novel targets and optimal combination therapies. In their 2018 paper, Pettitt et al.7 uncover both resistance mechanisms and novel synthetic lethal relationships in the same CRISPR screen.
PARP/BRCA is the classic example of a synthetic lethal relationship, whereby tumors containing mutations in the BRCA1 or BRCA2 are highly sensitive to treatment with PARP inhibitors, but this treatment regimen is no less susceptible to resistance developing. Genome-wide CRISPR screening uncovered mutations in PARP1 responsible for resistance to talazoparib, a PARP inhibitor (PARPi) often used to treat cancers with defects to BRCA1 and BRCA2. As well as identifying the key mutations that alter the DNA-binding capability of PARP1, they also discovered a double PARP1/BRCA1 mutant that displayed distinct sensitivities to chemotherapeutic agents.
More recently, a 2019 Nature Communications paper presented work investigating resistance mechanisms to the DNA topoisomerase I inhibitor (TOPi) topotecan in ATM-deficient cells using CRISPR screening.8 The authors suggest that the expression of components of the BRCA1-A complex and NHEJ DNA repair pathway could aid in predicting responses to TOP1 or PARPi, and that certain resistance mechanisms may confer sensitivity to alternative drugs. These publications show the capacity of CRISPR screening to provide new opportunities not only for drug target identification and validation, but for patient stratification to optimize treatment strategies for immediate impact.
Screening for sensitivity
Screening for resistance allows drug developers and clinicians to anticipate where and how treatment regimens are likely to fail, and plan accordingly. Screening for gene knockouts that confer sensitivity can open up new treatment regimens or improve existing ones. One recent example is from Manguso et al.,9 who used CRISPR screening in vivo to discover gene knockouts that sensitize tumors to immunotherapy, offering a new means to discover immuno-oncology targets. They highlight Ptpn2, the deletion of which dramatically increased the response of tumors to immunotherapy.
In summary
In only the short time since publication of the first CRISPR screens, the technology has been improved across multiple facets to further enhance the efficiency and applications of the technique. These include developing algorithms to predict the most active sgRNA, improved strategies around guide targeting, improvements in the lentiviral backbones for enhanced delivery efficiency, and, more recently, combining orthologous Cas9 variants to enable dual screening—targeting two genes in every cell.10 Additionally, scientists are now moving beyond cancer cell lines and optimizing CRISPR-Cas9 resistance screening in primary cells.11
For obvious reasons, much of the high-profile work has focused on uncovering the mechanisms of disease resistance in human cancer, with the clinical impact of these findings likely to play out in the coming years. More recently however, scientists are adapting CRISPR screening to other paradigms: for example, the CRISPRi approach using dCas9 has been used to screen for phage-resistance mechanisms in E. coli, providing insights into the design of phage therapies.12
The simplicity of the approach means that adoption of CRISPR-Cas9 screening is likely to increase, as requiring only the ability to synthesize RNA, basic molecular biology techniques, and access to next-generation sequencing, genome-wide resistance screening is now within the grasp of most laboratories. Consequently, gene knockout has gone from being a tool for hit validation, to now being utilized as a first-pass screen widely adopted for drug discovery.
Recap: Pooled lentiviral CRISPR-Cas9 screen methodology
- A library of sgRNA designed to target every gene in the genome is delivered by lentiviral vector.
- Viral titer is optimized to ensure a single integration, and therefore single gene knockout per cell.
- The stable integration of the transgene acts as a barcode for each knockout, allowing screens to be performed in a pooled format.
- Following drug treatment, sgRNA are identified by highly-parallel next-generation sequencing
- Cells resistant to the drug will be enriched by the drug treatment, whereas those that are depleted have been sensitized .
References
1. Shalem, O. et al. Genome-scale CRISPR-Cas9 knockout screening in human cells. Science 343,84–87 (2014).
2. Wang, T., Wei, J. J., Sabatini, D. M. & Lander, E. S. Genetic Screens in Human Cells Using the CRISPR-Cas9 System. Science (New York, N.Y.) 343, 80–84 (2014).
3. Joung, J. et al. Genome-scale CRISPR-Cas9 knockout and transcriptional activation screening. Nature Protocols 12, 828–863 (2017).
4. Le Sage, C. et al. Dual direction CRISPR transcriptional regulation screening uncovers gene networks driving drug resistance. Scientific Reports 7, (2017).
5. Xiao, T. et al. Estrogen-regulated feedback loop limits the efficacy of estrogen receptor–targeted breast cancer therapy. Proceedings of the National Academy of Sciences 201722617 (2018).
6. Hou, P. et al. A Genome-Wide CRISPR Screen Identifies Genes Critical for Resistance to FLT3 Inhibitor AC220. Cancer Research 77, 4402–4413 (2017).
7. Pettitt, S. J. et al. Genome-wide and high-density CRISPR-Cas9 screens identify point mutations in PARP1 causing PARP inhibitor resistance. Nature Communications 9, (2018).
8. Balmus, G. et al. ATM orchestrates the DNA-damage response to counter toxic non-homologous end-joining at broken replication forks. bioRxiv 330043 (2018).
9. Manguso, R. T. et al. In vivo CRISPR screening identifies Ptpn2 as a cancer immunotherapy target. Nature 547, 413–418 (2017).
10. Najm, F. J. et al. Orthologous CRISPR-Cas9 enzymes for combinatorial genetic screens. Nature Biotechnology 36, 179–189 (2018).
11. Parnas, O. et al. A Genome-wide CRISPR Screen in Primary Immune Cells to Dissect Regulatory Networks. Cell 162, 675–686 (2015).
12. Rousset, F. et al. Genome-wide CRISPR-dCas9 screens in E. coli identify essential genes and phage host factors. PLOS Genetics 14, e1007749 (2018).