Welcome explorers of the epigenome! Are you looking for new tools to help answer the toughest of epigenetic enigmas? Then check out this suite of single-cell methods papers from technologically-talented teams led by Aviv Regev (Broad Institute of MIT and Harvard) and collaborators.
inCITE into How Transcription Factors Shape In Vivo Gene Expression
A collaboration with Hattie Chung (Broad Institute of Harvard and MIT) sought to provide “insight” into how nuclear proteins shape gene expression within tissue samples through a new epigenetic tool called inCITE-seq – intranuclear cellular indexing of transcriptomes and epitopes by sequencing. Implementing DNA-conjugated antibodies and single-nucleus RNA-seq across thousands of single nuclei on a droplet-based profiling platform allows for parallel quantitative measurements of nuclear protein levels and the transcriptome, which provides deep insight into how transcription factor levels affect gene expression in vivo.
Let’s hear all the details of this new study from Chung and colleagues to understand what new insight inCITE-seq brings:
- First, as a proof-of-concept, they showed that inCITE-seq accurately quantifies the nuclear translocation of the p65 (NF-κB transcription factor component) and RNA levels after stimulation of HeLa cells with tumor necrosis factor-α
- Integration of single-nucleus data identified 142 genes that positively-associate with p65 levels
- Then, they turned to an in vivo application of inCITE-seq to simultaneously evaluate p65 (NF-κB), c-Fos (AP-1), NeuN (pan-neural marker), and PU.1 (microglial marker) levels in single nuclei isolated from the mouse hippocampus after a pharmacologically induced seizure
- Nuclear protein levels reflect the diverse levels of activity-regulated transcription factors across key hippocampal cell types before and after seizure
- In this way, inCITE-seq allows the accurate quantification of associations between transcription factors and gene expression modules/pathways in excitatory neurons and describes the contributions of transcription factor combinations to improve the interpretation of gene expression programs
Overall, inCITE-seq efficiently describes how nuclear protein combinations can dynamically mold gene expression patterns in native tissue samples and help decipher complex phenotypes and regulatory mechanisms. Moving forward, the authors anticipate that integrating inCITE-seq with other single-cell profiling tools may provide for an even deeper insight into the epigenetic complexities of single-cell activities.
PHAGE-ATAC Assays Protein Levels and Chromatin Accessibility in Single Cells
A partnership with Evgenij Fiskin (Broad Institute of Harvard and MIT) led to the development of an epigenetic tool known as PHAGE-ATAC – a massively parallel droplet-based method that throws nanobody-displaying phages into the assay for transposase-accessible chromatin. PHAGE-ATAC takes advantage of recombinant phage display technology to simultaneously measure extracellular/intracellular protein levels, evaluate chromatin accessibility, and allow mitochondrial DNA-based tracing in single cells to understand the regulatory mechanisms that mediate single-cell states.
PHAGE-ATAC represents an exciting new addition to the single-cell profiling toolbox, so let’s hear from Fiskin and Colleagues about what it can do:
- Expression of an engineered nanobody-encoding phagemid (DNA-based cloning vector with both bacteriophage and plasmid properties) enables nanobody-displaying phage-based recognition of cell-surface antigens, simultaneous indexing of phagemids and ATAC fragments, and the separate generation of phage-derived tag and ATAC-seq libraries
- Development/validation/scaling used nanobodies against surface-exposed GFP in HEK293T cells or various cell surface markers in single peripheral blood mononuclear cells, with the latter results reflecting the expected cell populations while capturing accessible chromatin and mitochondrial genotypes
- Extensive benchmarking demonstrated high specificity and sensitive monitoring of protein expression
- As a proof-of-concept, PHAGE-ATAC supports sample multiplexing, intracellular protein analysis, and the detection of the severe acute respiratory syndrome (SARS)-CoV-2 spike protein in peripheral blood mononuclear cells
- Overall, PHAGE-ATAC shows promise for the single-cell detection of host and viral antigens
This exciting new epigenetic study introduces phages as renewable reagents for high-throughput single-cell protein profiling and demonstrates how to leverage phage libraries to select antigen-specific antibodies. Indeed, the description of a synthetic high-complexity phage library supports the selection of nanobodies that bind specific cells and allow for protein detection, cell characterization, and screening with single-cell genomics. Overall, PHAGE-ATAC represents a handy new addition to the single-cell profiling toolbox.
Tangram – A Deep Learning Tool Spatially Dissects Single-cell Transcriptome Puzzles
To solve the spatial puzzles present within our cells, a collaboration with Tommaso Biancalani (Genentech, San Francisco) led to the development of a novel deep learning framework called “Tangram,” which aligns single-cell and single-nucleus transcriptomics to various forms of spatial data (in situ, histological, and anatomical data from the same specimens) to create high-resolution integrated atlases.
Let’s hear how Biancalani, Scalia, and colleagues employed this deep learning tool that should help to spatially “resolve” more than a few single-cell transcriptome puzzles:
- Tangram learns a spatial alignment of single-cell/single-nucleus RNA-seq data from diverse types of spatial data so that it can be applied to a non-spatial dataset
- Tangram supports the spatial mapping of snRNA-seq data from the adult healthy mouse brain (including data from the BRAIN Initiative Cell Census Network) using data from various spatial techniques of different resolutions and gene coverage
- A few hundred marker genes stratified across cell types allows the transcriptome-wide mapping of the mouse brain cortex, which demonstrated consistent cell-type patterns in all cases
- Tangram helps to overcome limitations in throughput or resolution by correcting low-quality genes (even in high-resolution methods), providing single-cell resolution for low-resolution methods, and supporting genome-wide coverage for targeted methods
- Mapping simultaneous high-throughput ATAC and RNA expression with sequencing (SHARE-seq) from the mouse brain with spatial support from Tangram allows the prediction of spatial patterns of chromatin accessibility and transcription factor motif scores at single-cell resolution
While the authors have clearly demonstrated how Tangram can spatially resolve single-cell transcriptomic and epigenetic puzzles in their fascinating study, they also look to the future by suggesting uses beyond the mouse brain to other organs and the application of their new method to study diseased tissues.
Three Tools for One Cell – A Biological Bonanza!
Overall, this suite of single-cell tools – InCITE-seq, PHAGE-ATAC, and Tangram – should enable us to solve even the trickiest of epigenetic enigmas and support more than a few biological breakthroughs by supporting the efficient generation and resolution of single-cell data sets.
For further insight into inCITE-seq, see Nature Methods, October 2021; for more on the novelties of PHAGE-ATAC, see Nature Biotechnology, October 2021; and for how Tangram can spatially resolve single-cell transcriptional data, see Nature Methods, November 2021.