Twin Prime Editing Swap-seq Deciphers Control Mechanisms of Enhancers and Silencers in Noncoding Genomes
Background
The noncoding regions, which account for 98% of the human genome, serve as a critical regulatory landscape governing gene expression. This region contains various regulatory sequences, such as enhancers that promote transcription, silencers that suppress it, and insulators that compartmentalize 3D chromatin structures. Most disease-associated genetic variants are located in noncoding regions. However, determining the quantitative impact of each regulatory element on target gene expression has been a long-standing challenge.
Researchers have primarily relied on Massively Parallel Reporter Assays (MPRA). However, exogenous plasmids have clear limitations in failing to reflect the actual 3D chromatin folding or epigenetic environment. Subsequently, CRISPR interference (CRISPRi) screening was introduced, but it faced constraints in precisely measuring the individual effects of elements due to the spread of inhibitory signals. Thus, there is a need for a new platform to precisely delete multiple regulatory elements within their native chromatin context and track changes in expression.
Key Findings
A research team led by Professor Jesse Engreitz at Stanford University developed 'Swap-seq,' a large-scale screening platform that combines twin prime editing (twinPE) with flow cytometry. The strategy applied twinPE, which deletes hundreds of base pairs without double-strand breaks by combining two prime editing guide RNAs (pegRNAs). The team deleted hundreds of candidate regulatory sequences in the endogenous genome of the human monocytic leukemia cell line (THP-1). They then separated cells using fluorescence-activated cell sorting (FACS) and quantitatively measured the deletion effects through sequencing.
As a first target, the peptidyl-prolyl cis-trans isomerase F (PPIF) gene locus was analyzed. We successfully precisely measured the quantitative activity of distal regions surrounding PPIF and intronic enhancers. The functional boundary of the CCCTC-binding factor (CTCF) insulator, which forms the three-dimensional structure of chromatin, is also clearly delineated.
Notably, the team identified a novel silencer that suppresses distal gene expression. This silencer revealed a mechanism that inhibits the transcription of distant genes through a histone deacetylase (HDAC)-mediated pathway acting on nearby enhancers. Hundreds of silencers with similar epigenetic characteristics were also confirmed to be distributed across the genome.
Furthermore, the team utilized Swap-seq data to validate sequence-to-function predictive deep learning models. While deep neural network models captured the local chromatin activity of individual elements, they revealed significant predictive limitations in the task of linking long-range interactions with distal target genes and subsequent expression changes.
Significance and Outlook
Swap-seq has opened the way to systematically evaluate various noncoding elements, such as enhancers, silencers, and insulators, within the same endogenous genomic context. This represents a new turning point in functional genomics research, enabling the elucidation of pathogenic mechanisms for complex disease variants and establishing causal links with target genes. By identifying the causes of long-range interaction prediction failures using empirical data, it has provided essential benchmark data to advance next-generation genomic deep learning models.
From the perspective of whole-genome expansion, technical challenges also remain. The research team explained that twinPE editing efficiency may vary depending on chromatin accessibility and sequence characteristics. To expand beyond single-gene loci to regulatory elements across the entire genome, both the scale of screening and sequencing costs must be reduced. Follow-up research is needed to complete cell-type-specific regulatory maps by expanding the platform to various disease cell models and organoids.
Noncoding regulatory elements in the human genome can control gene expression through distinct mechanisms, including acting as enhancers, silencers, or topological elements. However, we lack tools to identify all such classes systematically and quantify their effects on gene expression in an endogenous genomic context. Here we develop Swap-seq, a method that combines pooled twin prime editing with flow-sorting to delete hundreds of endogenous regulatory sequences and measure their quantitative effects on the expression of a nearby gene. We apply Swap-seq to dissect the regulatory elements that control
This achievement is expected to accelerate the discovery of targets in noncoding regions, which account for the majority of disease-causing genetic variants but have been difficult to characterize functionally. This is because it can clearly demonstrate how specific noncoding variants identified via Genome-Wide Association Studies (GWAS)—such as those for autoimmune diseases or cancer susceptibility—disrupt key genes in specific cell types. In the field of new drug development, it becomes possible to design strategies that precisely target abnormally hyperactivated disease enhancers or silencers of inhibitory mechanisms using antisense oligonucleotides (ASOs) or small-molecule epigenetic control drugs. It can also serve as a key empirical metric for optimizing artificial synthetic promoters and enhancers to fine-tune the expression levels of therapeutic genes in a patient-specific manner within the field of cell and gene therapies.