Target Validation with CRISPR

Comments

October 28, 2022

A key step in drug discovery is target identification—the discovery of a biological target that is “druggable”, meaning its activity can be modulated by a therapeutic agent. But after a target has been identified, it must then be validated, a process that aims to prove a functional relationship between target and disease phenotype while also ensuring safety. Good target validation should result in increased confidence in the relationship between target and disease phenotype for therapeutic benefit and ensure only the most promising targets are taken through to the next phase of the drug discovery pipeline.

Search CRISPR products
Search Now Search our directory to find the right CRISPR products for your research needs.

The recent development of CRISPR-Cas9 gene-editing technology has had a tremendous impact on drug discovery, allowing researchers to deliberately activate or inhibit genes to elucidate and understand those cellular pathways that play a role in disease progression. It can also be used to create accurate models in which to better study disease phenotypes and screen small molecules to rapidly identify numerous potential targets. But CRISPR is also proving its worth further downstream in the drug discovery pipeline. This article discusses how CRISPR can be used to validate targets and identify those with the highest likelihood of being ideal therapeutic targets, translating to an increased chance of success in the clinic.

Scaling down with CRISPR

In models where a phenotype or response can be easily monitored, such as cell fitness, drug resistance, or morphological changes, drug discovery research can begin with a large, genome-wide primary screen—either RNAi or a pooled CRISPR library screen—and can result in hundreds or even thousands of potential targets that then require validation. A good first step is to perform a follow-up screen, albeit on a smaller scale, to narrow down the larger gene set to a feasible number of putative targets. When following up, researchers usually turn to array-based methods, where each sgRNA is present in a single well. Arrayed screens are not usually feasible for large-scale primary screens due to cost and the requirement for automation, they are ideally suited for these smaller follow-ups. As well as using CRISPR screening to reduce the number of targets to take forward, the robust and flexible nature of CRISPR means it can also be used for further phenotypic analysis for validation, such as:

Knocking out a target gene or the gene’s family members to see if this elicits the same phenotype to increase confidence it plays a role in the disease state
Use CRISPR to systematically knockout an entire gene pathway, with the aim of finding more or better druggable targets up- or downstream
Expanding the scope to multiple cell-line models with smaller pools of guides
Performing screens in alternative models, such as moving from cancer cells to iPSCs, organoids, or animal models

These experiments can not only potentially find new targets, but also result in increased confidence that the target should be taken down the pipeline.

Better cell models mean better validation

The discovery of new drug candidates for therapeutic use has the potential to save and change lives through the treatment of numerous diseases, including cancer, pathogenic infections, and inherited genetic disorders. But drug discovery is a long, complex, and expensive process—taking up to 15 years and over $1 billion to bring a drug to clinical use, with only a 1 in 5000 chance of a drug making it to market. Validation of potential hits increases the chance of success and requires a number of cell-based assays that need an accurate model of the disease to perform functional analysis and rescue studies.

Precision genome editing with CRISPR-Cas9 allows researchers to create useful cellular models of disease rapidly and economically for accurate assessment of a drug target for therapeutic benefit. Crucially this allows researchers to not just confirm findings but eliminate false positives and save on costly downstream work in the clinic. For example, this can be seen with the validation of the potential target of the MutT Homolog 1 (MTH1) that arose from RNAi knockdown studies and showed promise for the treatment of cancer. CRISPR-Cas9 was used to create MTH1 knockout cell lines, but treatment with specific MTH1 inhibitors showed no difference in cell viability between the knockout cell line and wildtype.¹ This CRISPR approach enabled the devalidation of the MTH1 target, therefore saving time and money pursuing a false target.

CRISPR-Cas9 technology can also be used in more biologically relevant models, such as iPSCs, primary cells, and even in vivo, to recapitulate the disease phenotype more accurately. Again, this allows researchers to build a more solid case around the validity of a target prior to progression to the clinic. For example, the inhibition of salt-induced kinase (SIK) was identified as a potential target for chronic obstructive pulmonary disease (COPD) as it altered macrophage cytokine profile to promote an anti-inflammatory response. However, SIK is expressed in one of three isoforms in the body—CRISPR-Cas9 technology was used to create biallelic point mutations of the different isoforms in iPSCs, which served as an appropriate model system and allowed researchers to identify the correct target.²

Utilizing the CRISPR toolkit

Although gene knockouts with CRISPR-Cas9 are a powerful tool in functional screens for drug discovery, they do come with limitations. The production of a null phenotype may not accurately mimic drug inhibition as suppression of gene products by a drug is rarely absolute. When validating a target, it is a good idea to use orthogonal approaches to confirm results seen using CRISPR knockouts. For example, CRISPRi can be used to modulate gene expression rather than full knockdown to better mimic the action of a drug and understand the functional response to gene suppression on target sensitivity. Similarly, CRISPRa can be used to mimic gain of function and overexpression mutations, and the results studied.