Genome-Wide Single-Cell and Single-Molecule Footprinting of Transcription Factors with Deaminase

Academic Background

In humans and other mammals, although the genome of each somatic cell is essentially the same, the functions of different cell types vary significantly. This diversity is primarily determined by the binding of transcription factors (TFs) to gene regulatory regions, which control the transcription of DNA into RNA. Understanding how TFs bind to the genome is a central issue in functional genomics research. However, existing research methods have certain limitations. Traditional “bottom-up” approaches (such as atomic-resolution structures and single-molecule imaging) and “top-down” approaches (such as classical genetics and molecular biology) have provided valuable information but cannot comprehensively reveal TF binding patterns at the single-cell and single-molecule levels.

To overcome these limitations, this study developed a new technique called “Footprinting with Deaminase” (FOODIE), which enables precise genome-wide detection of TF binding sites at the single-cell and single-molecule levels. This technology not only reveals TF binding sites but also detects the cooperative interactions between adjacent TFs, providing new insights into gene regulatory networks.

Source of the Paper

This paper was co-authored by Runsheng He, Wenyang Dong, Zhi Wang, and others from Changping Laboratory, Peking University, and other institutions, and was published in PNAS on December 17, 2024. The corresponding author is Xiaoliang Sunney Xie, whose team has a strong research background in single-molecule genomics and transcriptional regulation.

Research Process

1. Development and Optimization of FOODIE Technology

The core idea of FOODIE technology is to modify DNA using double-stranded DNA deaminase. The binding of TFs protects the cytosine at their binding sites from deamination, leaving footprints in single-molecule sequencing. The research team first tested various deaminases and ultimately selected DDDB and MGYPDA829 because they can efficiently convert cytosine to uracil in a wide range of sequence contexts.

2. Experimental Workflow

The FOODIE experimental workflow includes the following key steps:

1. Cell Permeabilization and Open Chromatin Enrichment: Cells are permeabilized using Tn5 transposase to enrich open chromatin regions, which are the primary sites of TF binding.
2. Deamination Reaction: After Tn5 treatment, deaminase is added to the reaction. The binding sites of TFs are protected from deamination due to steric hindrance, while unbound regions are modified.
3. Single-Molecule Sequencing: DNA that has undergone deamination is subjected to single-molecule sequencing, and TF binding sites are identified by analyzing the conversion rate of cytosine.
4. Data Analysis: Specialized algorithms and a web server (http://foodie.sunneyxielab.org) were developed for data visualization and analysis.

3. Application of Single-Cell FOODIE (scFOODIE)

To perform cell-type-specific TF footprinting in heterogeneous tissues, the research team developed single-cell FOODIE technology. By analyzing approximately 11,200 cells from the mouse hippocampus, they successfully identified eight major cell types and detected TF binding patterns in different cell types. For example, footprints of AP-1 were detected in hippocampal pyramidal cells, consistent with the known role of AP-1 in synaptic activity and plasticity.

Main Results

1. Genome-Wide Distribution of TF Footprints

Using FOODIE technology, the research team mapped TF binding sites across the genome and found that TF binding sites are primarily located in gene promoters and enhancers. For example, CTCF binding sites are mainly located upstream of the transcription start site (TSS), while YY1 binding sites are located downstream.

2. Detection of TF Cooperativity

FOODIE technology can also detect cooperative interactions between adjacent TFs. The study found that some TFs exhibit positive cooperativity (e.g., RFX and CREB), while others exhibit negative cooperativity (e.g., two NRF1 binding sites). This cooperativity may be achieved through protein-protein interactions or competitive binding.

3. Shared TFs in Gene Modules

The research team also found that gene modules performing specific biological functions (Correlated Gene Modules, CGMs) are often regulated by shared TFs. For example, E2F factors are significantly enriched in the promoters of genes in cell cycle-related modules, while RelB is significantly enriched in the enhancers of genes in immune system regulation modules. This suggests that different gene modules have distinct regulatory patterns.

Conclusions and Significance

FOODIE technology provides a powerful tool for studying TF binding patterns at the single-cell and single-molecule levels. This technology not only precisely detects TF binding sites but also reveals cooperative interactions between adjacent TFs, offering new perspectives for understanding gene regulatory networks. Additionally, FOODIE technology is high-throughput and cost-effective, making it suitable for a wide range of applications in clinical samples and human tissues.

Research Highlights

1. High-Precision Detection: FOODIE technology enables precise genome-wide detection of TF binding sites at the single-cell and single-molecule levels.
2. Cooperativity Analysis: The technology can detect cooperative interactions between adjacent TFs, providing new insights into gene regulatory networks.
3. Broad Application Prospects: FOODIE technology is high-throughput and cost-effective, making it suitable for various cell types and tissue samples.

Other Valuable Information

The research team also developed an open database (http://foodie.sunneyxielab.org) for storing and sharing data generated by FOODIE technology. This database provides valuable research resources for other scientists and is expected to advance the field of transcriptional regulation.

Through this study, we have gained a deeper understanding of the role of TFs in gene regulation and provided new technological tools for future functional genomics research.