The functional organization of excitatory synaptic input to place cells

Michael D. Adoff, Jason R. Climer, Heydar Davoudi, Jonathan S. Marvin, Loren L. Looger, Daniel A. DombeckNature Communications 2021 · DOI

TL;DR

This study reveals how excitatory synaptic inputs are organized on the dendrites of hippocampal place cells, showing that increased and clustered inputs within a place field are crucial for its formation. This work provides a synaptic-scale explanation for how these cells encode location, suggesting that place cell representations are largely inherited from upstream brain regions.

What Problem Does This Paper Address?

Hippocampal place cells are fundamental to spatial navigation and memory, firing specifically when an animal is in a particular location. While we know that they fire in specific "place fields," the how at the synaptic level has been a mystery. Specifically, how the excitatory inputs to these cells are organized across their dendrites and how this organization contributes to location-specific firing remained unknown. Existing models proposed that the pattern of inputs and their plasticity were key, but direct evidence was lacking. This paper aims to bridge this gap by visualizing and analyzing excitatory synaptic input at a fine spatial scale within CA1 place cells of behaving mice.

Methods

The researchers employed resonant scanning two-photon microscopy to visualize excitatory synaptic input in real-time. A key innovation was the use of iGluSnFR, a genetically encoded glutamate sensor that fluoresces in response to glutamate release, allowing them to optically record synaptic activity. This was combined with jRGECO1a, a red calcium indicator, to simultaneously monitor somatic action potential firing in CA1 pyramidal neurons.

Mice navigated a virtual reality linear track while their hippocampal CA1 neurons were imaged. The iGluSnFR signal was analyzed at the micron-scale (1 µm regions of interest or ROIs) along dendritic segments. They developed sophisticated analysis pipelines to:

  • Identify and filter significant iGluSnFR transients from noise, establishing a threshold for reliable detection.
  • Characterize the spatial tuning of individual ROIs, classifying them as "place-ROIs" (spatially tuned to the animal's location) or "active-nonplace-ROIs" (active but not specifically tuned to a location).
  • Quantify total excitatory input by integrating iGluSnFR transients.
  • Analyze the functional organization of inputs by examining spatial and temporal correlations between ROIs on the same dendrite.
  • Identify "functional clusters" of nearby ROIs that were co-active in time, using independent component analysis.

Control experiments using GFP instead of iGluSnFR were performed to validate the specificity of the observed transients. Slice experiments were also conducted to calibrate the iGluSnFR response to known synaptic events.

Datasets

The study utilized data collected from 11 behaving mice navigating a virtual reality linear track. They recorded from 109 basal and proximal oblique dendritic segments across these mice. For control experiments, data were collected from 4 mice expressing GFP. The analysis involved identifying and characterizing thousands of 1-µm dendritic ROIs. No new benchmark datasets were introduced; rather, the study generated novel data by imaging neural activity in vivo.

Key Insights and Ideas

  • Place fields are not just about somatic firing: The study demonstrates that the dendritic input to place cells is highly organized and plays a critical role in defining their location-specific firing.
  • Input inheritance: Many CA1 place cells seem to inherit their place fields from upstream synaptic partners (e.g., in CA2/3) that already have spatially tuned inputs. This suggests that CA1 acts as an integrator of pre-formed representations rather than solely generating them from scratch.
  • Clustering matters: Excitatory inputs within a place field are not randomly distributed but are functionally clustered on the dendrites at the ~10 µm scale. This clustering, along with increased overall input, is implicated in driving place cell firing.
  • iGluSnFR as a powerful tool: The study showcases the utility of iGluSnFR for real-time, high-resolution imaging of synaptic glutamate release in behaving animals, providing a synaptic-scale view previously unattainable.

Key Results

  • Increased excitatory input in place fields: Total excitatory input to CA1 dendrites was significantly greater within the somatic place field compared to outside. This increase primarily originated from place-ROIs whose spatial tuning overlapped with the somatic field.
  • Spatial tuning of inputs: Dendrites received a mix of spatially tuned (place-ROIs) and less-tuned (active-nonplace-ROIs) inputs. Place cells had a higher proportion of place-ROIs compared to nonplace cells.
  • Functional clustering of inputs: Excitatory inputs were found to be temporally co-active and anatomically clustered on the dendrites at the ~10 µm scale, and this clustering was more pronounced within the somatic place field.
  • Place-ROIs drive the in-field increase: The increase in total excitatory input within a place field was largely due to a greater number of place-ROIs active in-field, rather than increased input strength to individual place-ROIs.
  • Inheritance from upstream: The findings suggest that many CA1 place cells are part of larger multi-brain-region assemblies, inheriting their spatial representations from upstream synaptic partners.

Limitations and Future Work

  • Resolution limits: The iGluSnFR signal has a spatial resolution of ~1 µm and a temporal resolution of hundreds of milliseconds. This means individual synapses cannot be resolved, and input synchrony is measured at a coarse temporal scale. The authors acknowledge that increased presynaptic strength might contribute to excitation but could not be directly resolved by their method.
  • Proxy for synaptic release: iGluSnFR reports glutamate release, not postsynaptic potential or strength. While a valuable proxy, it doesn't capture the full picture of synaptic integration.
  • Future directions: The authors suggest future studies should investigate how these synaptic input properties change during place field formation and explore the role of other inputs (e.g., inhibitory) and mechanisms (e.g., distal inputs, cellular properties) in place cell firing. They also highlight the need to understand how context-specificity and global remapping affect these synaptic input patterns.

Summary updated

Abstract

Hippocampal place cells contribute to mammalian spatial navigation and memory formation. Numerous models have been proposed to explain the location-specific firing of this cognitive representation, but the pattern of excitatory synaptic input leading to place firing is unknown, leaving no synaptic-scale explanation of place coding. Here we used resonant scanning two-photon microscopy to establish the pattern of synaptic glutamate input received by CA1 place cells in behaving mice. During traversals of the somatic place field, we found increased excitatory dendritic input, mainly arising from inputs with spatial tuning overlapping the somatic field, and functional clustering of this input along the dendrites over ~10 µm. These results implicate increases in total excitatory input and co-activation of anatomically clustered synaptic input in place firing. Since they largely inherit their fields from upstream synaptic partners with similar fields, many CA1 place cells appear to be part of multi-brain-region cell assemblies forming representations of specific locations.
BibTeX
@inproceedings{adoff2021functional,
  author = {Michael D. Adoff and Jason R. Climer and Heydar Davoudi and Jonathan S. Marvin and Loren L. Looger and Daniel A. Dombeck},
  title = {{The functional organization of excitatory synaptic input to place cells}},
  booktitle = {Nature Communications},
  year = {2021},
  doi = {10.1038/s41467-021-23829-y},
}

Loading related content...

Figure