Integrative single cell multiomics analysis of human retina indicates a role for hierarchical transcription factors collaboration in genetic effects on gene regulation

Jun Wang, Xuesen Cheng, Qingnan Liang, Leah A. Owen, Meng Wang, Margaret M. DeAngelis, Yumei Li, Rui ChenbioRxiv 2022 · DOI

TL;DR

This study uses single-cell multiomics to dissect how genetic variations influence gene regulation in specific retinal cell types. They found that cell-type specific genetic effects are largely driven by the complex interplay of transcription factors, rather than just the accessibility of DNA regions, offering a new model for understanding genetic contributions to eye diseases.

What Problem Does This Paper Address?

Understanding how genetic variations lead to differences in gene regulation is crucial for deciphering complex traits and diseases. However, previous studies using bulk tissue data often masked cell-type specific regulatory effects. This paper tackles the challenge of pinpointing how genetic variants impact gene expression and chromatin accessibility within individual cell types of the human retina, a critical step for understanding the genetic basis of eye diseases.

Methods

The researchers employed a powerful single-cell multiomics approach, integrating:

  • Whole Genome Sequencing (WGS): To identify genetic variations in the donors.
  • Single-Nuclei RNA Sequencing (snRNA-seq): To profile gene expression in individual cells.
  • Single-Nuclei Assay for Transposase-Accessible Chromatin Sequencing (snATAC-seq): To map open chromatin regions (indicating regulatory potential) in individual cells.

They then used these data to map:

  • sc-eQTLs (single-cell expression Quantitative Trait Loci): Variants associated with gene expression in specific cell types.
  • sc-caQTLs (single-cell chromatin accessibility Quantitative Trait Loci): Variants associated with chromatin accessibility in specific cell types.
  • sc-ASE (single-cell allelic specific expression): Differences in expression between the two alleles of a gene in a single cell.
  • sc-ASCA (single-cell allelic specific chromatin accessibility): Differences in chromatin accessibility between the two alleles in a single cell.

By integrating these analyses, they aimed to identify regulatory elements and variants that are active in specific retinal cell types and to understand the mechanisms by which they modulate gene regulation.

Datasets

The study utilized retinal tissues from 20 healthy human donors. The data generated included:

  • Whole genome sequencing data.
  • Single-nuclei RNA sequencing data.
  • Single-nuclei ATAC sequencing data.

These datasets were used to identify cell types, map genetic variants, and analyze their regulatory effects.

Key Insights and Ideas

  • Cell-type specificity is key: Genetic effects on gene regulation are not uniform across all cells; they are highly context-dependent on the specific cell type.
  • Transcription factors are the main drivers: For cell types with similar origins (like retinal cells), genetic variants primarily influence gene regulation by altering the binding of transcription factors (trans-factors), rather than solely by changing the accessibility of DNA regions (cis-elements).
  • Hierarchical collaboration: This suggests a model where transcription factors work together in a hierarchical manner to fine-tune gene expression and chromatin states in a cell-type specific way.
  • Beyond cis-elements: While cis-regulatory elements (like enhancers) are important, their accessibility alone doesn't fully explain cell-type specific genetic effects. The interaction with trans-acting factors is crucial.
  • GWAS loci fine-mapping: The single-cell data provides a powerful lens to pinpoint causal variants and target genes underlying genetic associations with eye diseases, revealing the specific cell types involved.

Key Results

  • The researchers mapped a significant number of sc-eQTLs (14,377) and sc-caQTLs (23,287), with a substantial proportion being cell-type specific (87-92% for sc-eQTLs).
  • Interestingly, many cell-type specific sc-eQTLs and sc-caQTLs were found within regulatory elements (OCRs) that are accessible in multiple cell types. This highlights the role of trans-factors in conferring cell-type specificity.
  • They found that transcription factors whose binding sites are affected by genetic variants tend to be highly expressed in the cell types where the variant has an effect.
  • Integration with GWAS data allowed them to identify enriched cell types for specific eye traits (e.g., astrocytes and Müller glia for glaucoma) and to fine-map candidate causal variants and genes underlying these associations.

Limitations and Future Work

  • Sample size: While 20 donors is a good starting point for single-cell multiomics, the limited number of individuals might affect the power to detect rare variants or subtle effects.
  • Cell type resolution: The study focused on major retinal cell types. Future work could explore finer distinctions within these cell types or investigate other tissues.
  • Mechanistic validation: While the study proposes a model involving transcription factor collaboration, further experimental validation would be needed to confirm these mechanisms directly.
  • Investigating different lineages: The study focused on cell types with similar lineages. Future research could explore how these mechanisms differ in cell types from distinct lineages.

Summary updated

Abstract

Abstract Background Systematic characterization of how genetic variation modulates gene regulation in a cell type specific context is essential for understanding complex traits. To address this question, we profiled gene expression and chromatin state of cells from healthy retinae of 20 human donors with a single-cell multiomics approach, and performed genomic sequencing. Results We mapped single-cell eQTL (sc-eQTLs), single-cell caQTL (sc-caQTL), single-cell allelic specific chromatin accessibility (sc-ASCA) and single-cell allelic specific expression (sc-ASE) in major retinal cell types. By integrating these results, we identified and characterized regulatory elements and genetic variants effective on gene regulation in individual cell types. Most of the sc-eQTLs and sc-caQTLs identified show cell type specific effects, while the cis-elements containing the genetic variants with cell type specific effects tend to be accessible in multiple cell types. Furthermore, the transcription factors with binding sites perturbed by genetic variants tend to have higher expression in the cell types, where the variants have effect, than the cell types where the variants do not have effect. Finally, we identified the enriched cell types, candidate causal variants and genes, and cell type specific regulatory mechanism underlying GWAS loci. Conclusions Overall, genetic effects on gene regulation are highly context dependent. Our results suggest that among cell types sharing a similar lineage, cell type dependent genetic effect is primarily driven by trans-factors rather than cell type specific chromatin state of cis-elements. Our findings indicate a role for hierarchical transcription factors collaboration in cell type specific effects of genetic variants on gene regulation.
BibTeX
@misc{wang2022integrative,
  author = {Jun Wang and Xuesen Cheng and Qingnan Liang and Leah A. Owen and Meng Wang and Margaret M. DeAngelis and Yumei Li and Rui Chen},
  title = {{Integrative single cell multiomics analysis of human retina indicates a role for hierarchical transcription factors collaboration in genetic effects on gene regulation}},
  year = {2022},
  doi = {10.1101/2022.11.16.516814},
  howpublished = {bioRxiv},
}
Figure