Deep mutational scanning (DMS) is a powerful method for studying the functional consequences of various genetic variants within a specific gene or genomic region. This technique combines high-throughput DNA sequencing with systematic mutagenesis to create and assess the impact of many different mutations simultaneously (Araya & Fowler, 2011). It has particular relevance in the field of drug discovery, offering transformative potential across various stages of the pipeline, from target identification and validation to lead optimisation.
Target Identification and Validation
In drug discovery, the identification of robust drug targets is critical. DMS can be used to comprehensively assess the functional impact of mutations in a candidate protein, helping to validate whether drugs can effectively target it. By systematically mutating every amino acid in the protein and assessing the resulting phenotype, researchers can identify essential domains and residues crucial for the protein's biological function. This high-resolution mapping can confirm the protein’s role in disease pathology and highlight allosteric sites that might be more amenable to drug targeting than the active site, providing a solid rationale for prioritising it as a drug target. For example, a DMS approach has been conducted to reveal the functional importance of each residue in proton recognition in GPR68. The result suggested that, in contrast to other proton-sensitive channels and receptors, GPR68 did not have a single essential site for proton recognition. Instead, a collection of titratable residues spans from the extracellular surface to the transmembrane area, linking with canonical class A GPCR activation motifs to initiate proton-sensing GPCRs. Specifically, the study revealed that the protonation of key residues surrounding an extracellular facing cavity resulted in conformational rearrangements with TM3 as a central conduit (Howard et al., 2024).
Lead Optimisation
Once potential lead compounds are identified, DMS can be employed to refine these molecules for enhanced potency, efficacy, and specificity. By examining how mutations in the target protein affect its interaction with the drug, researchers can identify which protein regions are crucial for binding the drug and which mutations affect its potency and efficacy. This data can guide the modification of the chemical structure of the lead compounds to improve potency and selectivity, thereby optimising the drug's design.
Predicting and Overcoming Drug Resistance
A significant challenge in drug development, particularly in the treatment of infectious diseases and cancer, is resistance. Mutations in target proteins can lead to decreased drug affinity and/or potency, rendering treatments ineffective. DMS provides a proactive approach to this problem by predicting potential resistance mutations before they are clinically observed (Pines et al., 2020). Furthermore, understanding these resistance mechanisms enables the design of next-generation drugs that either avoid these mutations or remain effective against them. It also helps in combination therapy strategies, where drugs are designed to target multiple sites or pathways simultaneously, reducing the likelihood of resistance development.
Precision Medicine
Finally, DMS aids in the advent of precision medicine by enabling a more detailed understanding of how genetic variation within human populations affects drug response. By evaluating a broad spectrum of mutations, DMS can help predict which patient subpopulations will respond to a drug and which might suffer adverse effects. This can guide the development of more personalised therapies tailored to the genetic makeup of individual patients, enhancing therapeutic outcomes and minimising harm.
Limitations
Deep mutational scanning encounters several constraints that affect its broad applicability and the interpretation of its results. The choice of model systems, such as yeast, bacteria, or mammalian cells, can heavily influence the method's effectiveness. Certain systems might not replicate the natural environment or post-translational modifications observed in humans, potentially leading to inconsistencies in how mutations influence protein function in real-world scenarios. Moreover, DMS requires considerable resources, including time, financial investment, and specialised expertise, limiting accessibility for some research laboratories. Additionally, DMS is not universally applicable to all proteins; it struggles with proteins that lack a clear function or perform multiple complex roles. Finally, current sequencing technology restricts the mutagenizable region to about 300 amino acids, although this limitation has eased from an initial 25 amino acids and is expected to further diminish as sequencing technology advances (Fowler et al., 2014).
Future directions
As deep mutational scanning continues to evolve, its future directions will likely focus on enhancing its scalability, accuracy, and applicability across broader biological contexts. An essential advancement will be the integration of DMS with emerging sequencing technologies that allow for longer reads and more accurate mapping of complex mutations. This could expand the method's capability to explore larger genomic regions and more intricate genetic variations. Additionally, combining DMS with computational modelling and machine learning algorithms promises to improve the predictive power of the technique, enabling more precise interpretations of how mutations affect protein function in diverse cellular environments. Another promising development is the application of DMS to a wider array of biological systems, including multicellular organisms and human cells, to better mimic physiological conditions and disease states. This would help bridge the gap between in-vitro findings and clinical outcomes.
In conclusion, deep mutational scanning is a versatile tool in drug discovery, providing detailed insights that can drive the early stages of target and lead identification, combat drug resistance, and refine therapeutic indices to produce safer and more effective drugs.
Reference
Araya, C. L., & Fowler, D. M. (2011). Deep mutational scanning: assessing protein function on a massive scale. Trends in Biotechnology, 29(9), 435–442. https://doi.org/10.1016/j.tibtech.2011.04.003
Fowler, D. M., Stephany, J. J., & Fields, S. (2014). Measuring the activity of protein variants on a large scale using deep mutational scanning. Nature Protocols, 9(9), 2267–2284. https://doi.org/10.1038/nprot.2014.153
Howard, M. K., Hoppe, N., Huang, X.-P., Macdonald, C. B., Mehrotra, E., Patrick Rockefeller Grimes, Zahm, A. M., Trinidad, D. D., English, J. G., Coyote-Maestas, W., & Aashish Manglik. (2024). Molecular basis of proton-sensing by G protein-coupled receptors. BioRxiv (Cold Spring Harbor Laboratory). https://doi.org/10.1101/2024.04.17.590000
Pines, G., Fankhauser, R. G., & Eckert, C. A. (2020). Predicting Drug Resistance Using Deep Mutational Scanning. Molecules, 25(9), 2265. https://doi.org/10.3390/molecules25092265
Comments