Date: October 23, 2024
Source: Jackson Laboratory
In an innovative leap for gene therapy and biotechnology, researchers have successfully harnessed artificial intelligence to design specialized DNA switches that can precisely regulate gene expression in specific tissues. This development, led by The Jackson Laboratory in collaboration with the Broad Institute of MIT and Harvard and Yale University, introduces a new method for activating or repressing genes with unprecedented accuracy across different cell types. The study was published in the journal Nature, showing promising implications for future medical therapies and biotechnological advancements.
“What makes these AI-designed DNA elements remarkable is their precision in targeting specific cell types,” explained Dr. Ryan Tewhey, senior author of the study and associate professor at The Jackson Laboratory. “Our approach now allows us to regulate genes in just one tissue without impacting others, potentially revolutionizing the way we apply gene therapy.”
In recent years, gene editing and gene therapy have made remarkable strides in modifying genes in living organisms. However, targeting gene expression to specific cells or tissues has been challenging. This is largely because of the intricate nature of cis-regulatory elements (CREs), the DNA sequences that control gene expression, which function as biological “switches” by determining when, where, and how much a gene is expressed.
To overcome these limitations, Tewhey and his team developed a machine learning model that decodes the “grammar” of these DNA sequences. By analyzing hundreds of thousands of DNA sequences and their activity levels across three types of cells—blood, liver, and brain—the AI model was able to recognize complex patterns that influence gene activation. With this data, the researchers developed CODA (Computational Optimization of DNA Activity), a platform capable of designing novel CREs with high specificity for particular cell types.
The team tested the AI-designed CREs by introducing them into cells and evaluating how selectively they activated genes in their target cells. The synthetic CREs demonstrated a higher level of specificity than their natural counterparts, in some cases activating genes only in the desired cell type. For example, in trials on zebrafish and mice, certain synthetic CREs succeeded in activating fluorescent proteins specifically in zebrafish liver cells while leaving other tissues unaffected.
“This technology brings us closer to a future where we can engineer regulatory DNA sequences with specific functions,” Tewhey said. “Not only could these synthetic CREs be useful for fundamental research, but they also hold promise for precise therapeutic interventions, where we could control gene activity only in targeted cell types.”
By using synthetic CREs, researchers could soon develop therapeutic applications that can deliver targeted treatments without unintended effects on non-targeted organs or cells, potentially enhancing the safety and effectiveness of future therapies.
Journal Reference:
Sager J. Gosai, Rodrigo I. Castro, Natalia Fuentes, John C. Butts, et al. “Machine-guided design of cell-type-targeting cis-regulatory elements.” Nature, 2024; DOI: 10.1038/s41586-024-08070-z