Enhanced transcriptome maps from multiple mouse tissues reveal evolutionary constraint in gene expression

Dmitri D. Pervouchine, Sarah Djebali, Alessandra Breschi, Carrie Davis, Pablo Prieto, Alexander Dobin, Andrea Tanzer, Julien Lagarde, Chris Zaleski, Lei Hoon See, Meagan Fastuca, Jörg Drenkow, Huaien Wang, Giovanni Bussotti, Baikang Pei, Suganthi Balasubramanian, Jean Monlong, Arif Harmanci, Mark Gerstein, M Beer, Cédric Notredame, Roderic Guigó, T GingerasNature Communications 2015 · DOI

Abstract

Mice are the primary model for human biology, yet the extent of shared transcriptional programming remains partially understood. This study utilizes RNA sequencing across a diverse array of mouse tissues to augment the mouse transcriptome with thousands of novel transcript candidates. By comparing these profiles with human cell lines, researchers identified a distinct class of "constrained" genes with expression levels that have remained stable since early vertebrate evolution. These genes, primarily involved in housekeeping functions, account for a significant portion of cellular transcriptional output. Their stability is maintained not just by DNA sequence conservation, but through a regulatory framework involving epigenetic marking and alternative splicing.

Introduction

While the mouse has long served as the gold standard for modeling human disease, a fundamental question persists: how much of the transcriptional "instruction manual" is truly shared between the two species? Previous research often focused on DNA sequence conservation, which does not always correlate with gene behavior. This study addresses gaps in mouse genome annotation and identifies a "mammalian core"—a set of approximately 6,600 genes whose expression levels are strictly "volume-controlled" across different tissues and species.

Methodology and Datasets

The researchers conducted a massive comparative analysis integrating several high-throughput datasets:

  • RNA-Sequencing (RNA-seq): Profiles were generated for 30 mouse tissue samples (fetal and adult) and 18 human cell lines from the ENCODE project.
  • Transcriptome Augmentation: RNA-seq data was combined with Cap Analysis of Gene Expression (CAGE) data from the FANTOM project to identify previously unknown transcripts.
  • Dynamic Range (DNR) Analysis: Constrained genes were identified by calculating the ratio between a gene's highest and lowest expression across all samples; low DNR indicates stable expression.
  • Integrative Omics: Expression data was layered with ChIP-seq data for histone marks (e.g., H3K4me3, H3K27ac) and sub-cellular localization data.

Key Insights: The "Fixed Volume" Genes

The study revealed that roughly 40% of human-mouse orthologs (6,636 genes) are "constrained," maintaining nearly constant expression levels regardless of tissue type. Interestingly, high expression stability does not always require perfect DNA sequence conservation, suggesting that evolution prioritizes transcriptional output over the exact genetic code.

Comparative transcriptome maps showing novel transcripts and conserved expression patterns across human (hg19) and mouse (mm9) genomes for several loci.

These constrained genes act as the cell's "housekeepers." They are marked by specific active epigenetic signatures and are frequently regulated by alternative splicing, which serves as a fine-tuning mechanism to maintain steady protein levels. Perturbation of these genes is strongly associated with severe phenotypes, including embryonic lethality and cancer.

Major Results

  • Annotation Expansion: The study added approximately 200,000 novel transcripts to the mouse annotation, significantly narrowing the gap in detail between the human and mouse genomes.
  • Transcriptional Dominance: Although constrained genes represent only 17% of annotated genes, they account for 39-41% of total cellular RNA.
  • Cross-Species Correlation: The expression levels of constrained genes showed a high correlation (0.82) between humans and mice, far exceeding the 0.32 correlation observed for unconstrained genes.
  • Splicing as a Regulatory Hub: In constrained genes, variation in transcript abundance is more frequently driven by alternative splicing than by changes in overall gene expression volume.

Discussion and Future Directions

While the study provides a robust map of the mammalian core transcriptome, it relies on human cell lines rather than primary tissues for comparison. Future work should expand tissue-to-tissue comparisons across more species. Additionally, the focus on polyadenylated (polyA+) RNAs leaves the evolutionary rules for non-polyadenylated or small RNAs unexplored. Understanding the precise mechanisms by which epigenetic marks enforce this millenia-long "volume control" remains a key challenge. These constrained patterns may eventually serve as a predictive tool for identifying pathogenic genetic mutations in humans.

Summary updated

Abstract

Mice have been a long-standing model for human biology and disease. Here we characterize, by RNA sequencing, the transcriptional profiles of a large and heterogeneous collection of mouse tissues, augmenting the mouse transcriptome with thousands of novel transcript candidates. Comparison with transcriptome profiles in human cell lines reveals substantial conservation of transcriptional programmes, and uncovers a distinct class of genes with levels of expression that have been constrained early in vertebrate evolution. This core set of genes captures a substantial fraction of the transcriptional output of mammalian cells, and participates in basic functional and structural housekeeping processes common to all cell types. Perturbation of these constrained genes is associated with significant phenotypes including embryonic lethality and cancer. Evolutionary constraint in gene expression levels is not reflected in the conservation of the genomic sequences, but is associated with conserved epigenetic marking, as well as with characteristic post-transcriptional regulatory programme, in which sub-cellular localization and alternative splicing play comparatively large roles.
BibTeX
@inproceedings{pervouchine2015enhanced,
  author = {Dmitri D. Pervouchine and Sarah Djebali and Alessandra Breschi and Carrie Davis and Pablo Prieto and Alexander Dobin and Andrea Tanzer and Julien Lagarde and Chris Zaleski and Lei Hoon See and Meagan Fastuca and Jörg Drenkow and Huaien Wang and Giovanni Bussotti and Baikang Pei and Suganthi Balasubramanian and Jean Monlong and Arif Harmanci and Mark Gerstein and M Beer and Cédric Notredame and Roderic Guigó and T Gingeras},
  title = {{Enhanced transcriptome maps from multiple mouse tissues reveal evolutionary constraint in gene expression}},
  booktitle = {Nature Communications},
  year = {2015},
  doi = {10.1038/ncomms6903},
}
Figure