Astrocytic engagement of the corticostriatal synaptic cleft is disrupted in a mouse model of Huntington’s disease

Carlos Benitez Villanueva, Hans Jacob Teglbjærg Stephensen, Rajmund Mokso, Abdellatif Benraiss, Jon Sporring, Steven A. GoldmanProceedings of the National Academy of Sciences 2023 · DOI

TL;DR

This study reveals that in a mouse model of Huntington's Disease (HD), astrocytes exhibit a significant impairment in their engagement with mature synapses in the corticostriatal pathway. This structural pathology, characterized by reduced astrocytic coverage of mature dendritic spines, may contribute to the hyperexcitability observed in HD by allowing for the leakage of glutamate and potassium ions.

What Problem Does This Paper Address?

Huntington's Disease (HD) is a devastating neurodegenerative disorder linked to mutations in the huntingtin gene. While astrocytic dysfunction is known to play a role in HD pathogenesis, the precise structural relationship between these glial cells and synapses in the diseased state has remained unclear. Specifically, understanding how astrocytes interact with medium spiny neurons (MSNs) – the primary neurons affected in HD – and their synapses is crucial. MSNs receive glutamatergic inputs, and astrocytes normally regulate synaptic environments by clearing neurotransmitters like glutamate and potassium. This paper aims to precisely map and quantify the astrocytic engagement of corticostriatal synapses in an HD mouse model, seeking to uncover the structural basis for the observed neuronal hyperexcitability.

Methods

The researchers employed a sophisticated, multi-modal imaging and analysis pipeline to achieve nanometer-scale resolution of astrocyte-synapse interactions. Key techniques included:

  • Viral Tracing: Monosynaptic retrograde tracing using glycoprotein-deleted rabies virus was used to label specific corticostriatal projection neurons and their connected MSNs. Astrocytes were labeled with lentiviral GFAP-driven EGFP.
  • 2-Photon Imaging: This allowed for the initial identification and visualization of interacting astrocyte processes and MSN dendritic domains in live tissue.
  • Correlated Light Electron Microscopy (CLEM): This technique bridges light and electron microscopy, enabling the precise localization of regions of interest (ROIs) identified by 2-photon imaging within the electron microscope.
  • Serial Block-Face Scanning Electron Microscopy (SBF-SEM): This advanced electron microscopy technique allows for the acquisition of high-resolution 3D volumes of tissue, enabling nanometer-scale reconstruction of synaptic structures and their surrounding astrocytic elements.
  • Automated Structural Analysis: A novel analytical framework was developed to quantify astrocytic engagement. This involved modeling the synapse using a torus region around the postsynaptic density (PSD) and employing k-cross-section statistics to derive geometric measurements of astrocytic processes in relation to the PSD. This allowed for the assessment of astrocytic volume, surface area, and other geometric features as a function of distance from the PSD.

Datasets

Not discussed in the paper. The study utilized tissue from R6/2 HD and wild-type (WT) mice.

Key Insights and Ideas

  • Maturity Matters: The study revealed that astrocytic engagement is not uniform across all synapses but is significantly influenced by the maturity of the dendritic spine.
  • HD's Selective Impact: In HD, astrocytes appear to "retract" from mature, mushroom-shaped dendritic spines, while paradoxically increasing their engagement with immature, thin spines.
  • Structural Basis for Hyperexcitability: The reduced astrocytic coverage of mature synapses in HD mice suggests a compromised ability to sequester glutamate and potassium. This "leakage" could directly contribute to the hyperexcitability characteristic of HD.
  • Novel Analytical Framework: The development of a robust analytical pipeline combining advanced imaging and quantitative geometric analysis provides a powerful new tool for studying the intricate relationships between glial cells and synapses at the nanoscale.

Key Results

  • Reduced PSD Volume: R6/2 HD mice showed a significantly lower volume of the postsynaptic density (PSD) compared to WT mice.
  • Closer Astrocytic Proximity: Despite reduced coverage, astrocytic processes were found to be in closer proximity to the PSD in R6/2 mice.
  • Differential Astrocytic Volume: Mature (mushroom) spines in R6/2 mice exhibited a significant decrease in associated astrocytic volume, while thin spines showed a trend towards increased astrocytic volume.
  • Differential Astrocytic Surface Area: R6/2 mice showed increased astrocytic surface area engagement with thin spines but a trend towards decreased surface area with mature spines.

Limitations and Future Work

  • Immunodeficient Hosts: The use of Rag1-/- mice (which are immunodeficient) for some experiments, while necessary for future transplantation studies, could potentially influence synaptic structure due to the absence of T and B cells. However, the authors note that both WT and R6/2 mice shared this background, isolating the effect of mHTT expression.
  • Under-representation of Mature Spines: In late-stage R6/2 mice, mature synapses are less abundant, which might artifactually minimize the observed effects of astrocytic retraction from these spines in the overall dataset.
  • Future Directions: The study's methodology is presented as broadly applicable for investigating glial-synaptic relationships in various neurological disorders. Future work could explore the molecular mechanisms underlying these observed structural changes and investigate therapeutic strategies aimed at restoring normal astrocytic engagement.

Summary updated

Abstract

Astroglial dysfunction contributes to the pathogenesis of Huntington's disease (HD), and glial replacement can ameliorate the disease course. To establish the topographic relationship of diseased astrocytes to medium spiny neuron (MSN) synapses in HD, we used 2-photon imaging to map the relationship of turboRFP-tagged striatal astrocytes and rabies-traced, EGFP-tagged coupled neuronal pairs in R6/2 HD and wild-type (WT) mice. The tagged, prospectively identified corticostriatal synapses were then studied by correlated light electron microscopy followed by serial block-face scanning EM, allowing nanometer-scale assessment of synaptic structure in 3D. By this means, we compared the astrocytic engagement of single striatal synapses in HD and WT brains. R6/2 HD astrocytes exhibited constricted domains, with significantly less coverage of mature dendritic spines than WT astrocytes, despite enhanced engagement of immature, thin spines. These data suggest that disease-dependent changes in the astroglial engagement and sequestration of MSN synapses enable the high synaptic and extrasynaptic levels of glutamate and K<sup>+</sup> that underlie striatal hyperexcitability in HD. As such, these data suggest that astrocytic structural pathology may causally contribute to the synaptic dysfunction and disease phenotype of those neurodegenerative disorders characterized by network overexcitation.
BibTeX
@inproceedings{villanueva2023astrocytic,
  author = {Carlos Benitez Villanueva and Hans Jacob Teglbjærg Stephensen and Rajmund Mokso and Abdellatif Benraiss and Jon Sporring and Steven A. Goldman},
  title = {{Astrocytic engagement of the corticostriatal synaptic cleft is disrupted in a mouse model of Huntington’s disease}},
  booktitle = {Proceedings of the National Academy of Sciences},
  year = {2023},
  doi = {10.1073/pnas.2210719120},
}
Figure