A spatially-resolved transcriptional atlas of the murine dorsal pons at single-cell resolution

Stefano Nardone, Roberto De Luca, Antonino Zito, Nataliya Klymko, Dimitris Nicoloutsopoulos, Oren Amsalem, Cory Brannigan, Jon M. Resch, Christopher Jacobs, Deepti Pant, Molly Veregge, Harini Srinivasan, Ryan M. Grippo, Zongfang Yang, Mark L. Zeidel, Mark L. Andermann, Kenneth D. Harris, Linus Tsai, Elda Arrigoni, Anne M.j. Verstegen, Clifford B. Saper, Bradford B. Lowell

TL;DR

Researchers have created a high-resolution, spatially-resolved transcriptional atlas of the murine dorsal pons, identifying over 120 distinct neuronal subtypes. By combining single-nucleus RNA sequencing (snRNA-seq) with MERFISH spatial imaging, the study reveals an unexpected level of cellular diversity in this critical brainstem region and demonstrates significant transcriptional conservation between mice and humans.

What Problem Does This Paper Address?

The dorsal pons (or dorsal pontine tegmentum, dPnTg) is a "strategic hub" in the brainstem, packed with nuclei that regulate life-sustaining functions like breathing, arousal, sleep-wake cycles, and pain processing. Despite its importance, our understanding of its cellular makeup has been limited.

Historically, brain regions were defined by anatomical boundaries (nuclei). However, these nuclei are often heterogeneous, containing various types of neurons with different functions. Previous maps lacked the resolution to distinguish these subtypes or pinpoint their exact spatial locations. This paper addresses the need for a "parts list" of the dPnTg, providing the molecular and spatial coordinates necessary to understand how this region controls complex physiological processes.

Methods

The authors employed a powerful two-step multimodal approach to map approximately 1 million cells:

  1. Single-nucleus RNA-seq (snRNA-seq): Using both DroNc-seq and 10X Genomics platforms, they profiled the transcriptomes of over 222,000 nuclei. This allowed them to categorize cells into clusters based on unique gene expression profiles, effectively identifying "what" the cell types are.
  2. MERFISH (Multiplexed Error-Robust Fluorescence In Situ Hybridization): They used a panel of 315 informative marker genes (identified from the snRNA-seq data) to spatially map ~685,000 cells across 46 coronal brain sections. This revealed "where" each transcriptionally defined subtype resides.
  3. Computational Integration: The team used the MetaNeighbor framework to correlate the sequencing clusters with the spatial data, ensuring that the cell types identified in a tube matched the cells imaged in the tissue.

Datasets

  • New Murine dPnTg Atlas: A massive dataset comprising snRNA-seq and MERFISH data from the mouse dorsal pons.
  • Allen Brain Atlas (ABA): Used as a reference for 10X Genomics data of the entire pons.
  • Human Pons snRNA-seq: The authors integrated their mouse data with a recently published human brainstem dataset (Siletti et al., 2023) to perform cross-species comparisons.
  • Interactive Dashboard: The data is hosted on a publicly accessible, GPU-accelerated SQL dashboard (http://harvard.heavy.ai:6273/).

Key Insights and Ideas

  • Unprecedented Diversity: The study identified over 120 neuronal subtypes, a far higher number than previously predicted. This suggests that the "functional units" of the brainstem are much more granular than traditional anatomical nuclei.
  • Spatial Logic: Subtypes are not randomly distributed; they follow precise spatial patterns. For example, within the Kölliker Fuse (a key respiratory center), they identified specific Calca+ clusters that represent a rostral continuation of the parabrachial nucleus.
  • Evolutionary Conservation: A major "aha moment" was the high degree of transcriptional similarity between mouse and human neuronal subtypes. This validates the mouse as a highly relevant model for studying human brainstem disorders and autonomic dysfunction.
  • Genetic Handles: The study pinpointed unique marker genes (e.g., Onecut3, Ebf2, Chst9) for specific subtypes. These markers act as "genetic handles," allowing future researchers to create transgenic mice to target and manipulate specific neurons to see how they affect behavior or physiology.

Key Results

  • Cellular Breakdown: The analysis identified 63 initial clusters across 12 major cell types (e.g., neurons, astrocytes, oligodendrocytes). Neurons accounted for ~40% of the total nuclei.
  • Subregion Granularity: By re-clustering neurons within specific subregions (like the Parabrachial nucleus or Locus Coeruleus), they resolved the 120+ subtypes.
  • High Fidelity: The correspondence between snRNA-seq and MERFISH was exceptionally high (AUROC > 0.85), proving the robustness of their classification.
  • Translational Value: 50 out of 67 human snRNA-seq clusters had direct mouse counterparts, highlighting the conserved molecular architecture of the brainstem across species.

Limitations and Future Work

  • Connectivity Gap: While the atlas tells us what and where the cells are, it doesn't fully map who they talk to. Future work will need to integrate this atlas with viral tracing to map the long-range axonal projections of these specific subtypes.
  • Functional Validation: The study provides a molecular map, but the specific physiological roles of most of these 120+ subtypes remain unknown. The identified marker genes now provide the roadmap for functional studies using optogenetics or chemogenetics.
  • Dynamic States: The atlas represents a "snapshot" of the healthy adult brain. Future research could explore how these transcriptional profiles change during disease, aging, or different physiological states (like hunger or sleep deprivation).

Summary updated

Abstract

The "dorsal pons", or "dorsal pontine tegmentum" (dPnTg), is part of the brainstem. It is a complex, densely packed region whose nuclei are involved in regulating many vital functions. Notable among them are the parabrachial nucleus, the Kölliker Fuse, the Barrington nucleus, the locus coeruleus, and the dorsal, laterodorsal, and ventral tegmental nuclei. In this study, we applied single-nucleus RNA-seq (snRNA-seq) to resolve neuronal subtypes based on their unique transcriptional profiles and then used multiplexed error robust fluorescence in situ hybridization (MERFISH) to map them spatially. We sampled ~1 million cells across the dPnTg and defined the spatial distribution of over 120 neuronal subtypes. Our analysis identified an unpredicted high transcriptional diversity in this region and pinpointed the unique marker genes of many neuronal subtypes. We also demonstrated that many neuronal subtypes are transcriptionally similar between humans and mice, enhancing this study's translational value. Finally, we developed a freely accessible, GPU and CPU-powered dashboard ( http://harvard.heavy.ai:6273/ ) that combines interactive visual analytics and hardware-accelerated SQL into a data science framework to allow the scientific community to query and gain insights into the data.
BibTeX
@inproceedings{nardone2024spatially,
  author = {Stefano Nardone and Roberto De Luca and Antonino Zito and Nataliya Klymko and Dimitris Nicoloutsopoulos and Oren Amsalem and Cory Brannigan and Jon M. Resch and Christopher Jacobs and Deepti Pant and Molly Veregge and Harini Srinivasan and Ryan M. Grippo and Zongfang Yang and Mark L. Zeidel and Mark L. Andermann and Kenneth D. Harris and Linus Tsai and Elda Arrigoni and Anne M.J. Verstegen and Clifford B. Saper and Bradford B. Lowell},
  title = {{A spatially-resolved transcriptional atlas of the murine dorsal pons at single-cell resolution}},
  booktitle = {Nature Communications},
  year = {2024},
  doi = {10.1038/s41467-024-45907-7},
}
Figure