HMN 2025: How Comprehensive map reveals neuronal dendrites in the mouse brain in greater detail

Understanding the shape or morphology of neurons and mapping the tree-like branches via which they receive signals from other cells (i.e., dendrites) is a long-standing objective of neuroscience research. Ultimately, this can help to shed light on how information flows through the brain and pin-point differences associated with specific neurological or psychiatric disorders.

The X. William Yang Lab at the Jane and Terry Semel Institute and the Department of Psychiatry and Biobehavioral Sciences at University of California, Los Angeles (UCLA) have devised new sophisticated methods to map neuronal dendrites in the mouse brain, which combine large-scale data collection with genetic labeling techniques and computational tools.

Their research approach, outlined in a paper published in Nature Neuroscience, allowed them to create a comprehensive map of two genetic types of neurons in the mouse brain, known as D1- and D2-type striatal medium spiny neurons (MSNs).

“The primary objective of our study was to develop a new systems biology approach to study the complex shape (i.e. morphology) of individual neurons in the mouse brain,” Dr. Chang Sin (Chris) Park, the first author of this study, told Medical Xpress.

“The foundation of modern neuroscience was established by Ramón y Cajal who studied the shape of neurons using Golgi staining over a century ago. Cajal’s Nobel-prize-winning work demonstrated that a fundamental understanding of the brain requires in-depth knowledge of its basic building block, individual neurons.”

Why mapping neuron morphology is challenging

While technological advances enabled significant progress within the field of neuroscience over the past decades, mapping the morphology of neurons remains a key challenge. Typically, studying the shape of neurons requires advanced technical expertise, equipment and intensive work on behalf of researchers.

Due to these challenges, morphological analyses of individual neurons in the brains of mammals have so far only been performed on a small scale by most labs, for instance, studying dozens of neurons at once. Moreover, neurons are typically only analyzed in 2D brain tissue slices, instead of in 3D within thick and intact tissues.

“Despite the vast number of neurons in the human brain (?86 billion) and the mouse brain (100 million), only about 200k neurons have ever been imaged and digitally reconstructed for their shape (called dendrites) according to an online database,” said Dr. X. William Yang, senior author of the paper and principal investigator of the study.

“To address this challenge, our Lab obtained three US NIH BRAIN Initiative grants. We sought to develop a series of genetic and bioinformatic tools to study morphology of genetically-labeled single neurons in the mouse brain at unprecedented resolution and scale.”

New tools to study the morphology of neurons and dendrites

In their earlier research, the researchers at UCLA introduced a new genetic sparse labeling reporter mouse model called MORF mice. This is a model that can be used to brightly and sparsely label individual genetically-defined neurons in the mouse brain.

“MORF mice capture at high resolution the shape of single neurons of a genetic type by simply breeding the mice, which is advantageous over the prior labor intensive neuronal labeling methods (e.g. dye or viral injections),” said Yang and Park.

“In this study, we employed MORF mice to further develop a new systems biology approach—called dendritome mapping—to systematically map the dendritic shape of thousands of neurons of a given genetic type in the mouse brain.”

As part of their recent study, Yang, Park and their colleagues set out to specifically map the shape of dendrites belonging to two genetically distinct types of neurons in the mouse striatum, known as D1- and D2-MSNs. These neurons are known to play a key role in the control of voluntary behavior, including movements and the learning of habits and motor skills, as well as higher mental functions, such as decision-making and language.

“The dysfunction or degeneration of D1- and D2-MSNs is implicated in brain disorders ranging from movement disorders (e.g. Parkinson’s disease or Huntington’s disease) to psychiatric disorders and drug addiction,” explained Yang.

“The D1- and D2-MSNs are the principal neurons in the striatum that are highly responsive to dopamine through the differential expression of dopamine D1- and D2-receptors, hence their namesake.”

There are approximately 1.7 to 1.9 million MSNs within the mouse brain. Yet previous neuroscience studies have cumulatively mapped the morphology of only about 500 D1- and D2- MSNs.

“Since the BRAIN Initiative has recently identified thousands of neuronal types in the mouse and human brains based on the expression of molecular markers, there remains a major knowledge gap,” said Yang and Park.

“Specifically, very little is known about the morphological shapes of any of these molecularly or genetically defined neurons. Our dendritome mapping provides the first proof-of-concept for studying the brain-wide dendritic morphology of two genetically defined neuronal types: the striatal D1- and D2-MSNs.”

A large-scale exploration of mouse neuronal dendrites

To study neuronal dendrites in the mouse striatum on a large scale, Yang, Park and his colleagues developed a new scalable approach that they dubbed “dendritome mapping.” Firstly, they analyzed the morphology of over 3,700 D1- and D2- type MSNs in the brains of twelve mice that were either healthy or expressing disease-causing mutant gene for Huntington’s disease in humans.

“The neurons were sparsely and brightly labeled with our MORF3 genetic sparse labeling reporter mice and imaged in thick iDISCO-cleared brain sections to capture their full 3D morphology,” said Park.

“Together with our colleague Drs. Daniel Tward and Hong-Wei Dong’s labs, we developed an integrated computational pipeline with a novel algorithm to register the thick brain sections to the 3D mouse brain reference atlas (called CCFv3), a streamlined digital neuronal reconstruction process, and comprehensive statistical analysis of neuronal shape with 31 morphometric features per neuron.”

To analyze the shape of neurons in even greater detail, Yang, Park and colleagues devised a highly innovative voxel-based analysis tool. This is an advanced method to study the spatial variation of neuron morphologies with a significantly finer resolution than that achieved by other existing approaches.

“Working closely with Dr. Tward, we divided the entire digital mouse brain (CCFv3) into 7,020 cubic boxes—each box being 500 µm per side—which resulted in 210 latticed cubic boxes in the striatum,” explained Yang.

“We then summarized the morphology of all MSNs in each box (up to dozens per box) with a morphometric representation (termed ‘eigen-morph’)—inspired by the weighted gene-co-expression network analysis method (i.e WGCNA)—and clustered boxes to identify those with shared morphometry.”

The team’s key findings and their implications

Overall, the results of the team’s analysis suggest that D1- and D2-MSNs have very similar morphologies, yet the former are slightly larger and more complex than the latter. Notably, the researchers found that these differences between the two types of MSNs are only robust in specific subregions of the striatum.

“Using the ‘box analysis, we found that the striatal MSNs are organized into groups of MSNs with shared ‘dendritic modules,” which have strikingly similar dendritic features within a module and quite divergent features across modules,” said Yang and Park.

“This finding was not previously known—as many neuroscientists (including ourselves) believed that the vast majority of MSNs in the striatum share similar morphologies.”

The UCLA team also found that MSNs in a given dendritic module are located in adjacent “boxes,” forming what they refer to as “morphological territories” that receive different corticostriatal inputs via axons (i.e., long projections of nerve cells that transmit information). This suggests that the different dendritic architectures of MSNs are linked to differences in their connections to parts of the cortex.

“Our study also showed that dendritome mapping can be used to study neuronal pathologies at exquisite resolution,” said Yang. “We found that aging uniformly causes shrinkage of both D1- and D2-MSNs, a phenomenon referred to as dendritic atrophy. While the Huntington’s disease mouse model showed more subtle and differential dendritic deficits in D1- and D2-MSNs.”

Paving the way for further studies

The results of this recent study highlight the potential of the system-biology approach for studying the shape of neurons devised by Yang’s lab and collaborators at UCLA.

In the future, the team’s approach could be improved and used by other neuroscience labs to investigate the link between the morphology of specific neuronal populations and different neurological or neuropsychiatric disorders.

“Our work provides the first striatum-wide D1- and D2-MSN dendritic morphological atlas in the adult mouse brain,” said the authors. “This database can be used as a reference for future studies of the development, biology, and pathology of these important neuronal types in the mammalian brain.”

Using their newly devised methods, the researchers have performed the first unbiased 3D brain-atlas and box-based dendritic morphological analysis at such refined resolutions. While their study focused on D1- and D-2 MSNs, their methods can be used to study the spatial organization of dendrites belonging to other genetically defined neurons in the brain.

“The results we gathered suggest that circuit-specific neuronal communications could be a source of dendritic variation in MSNs and possibly in other genetically defined neuronal cell types,” said Yang and Park. “This hypothesis should be further investigated in future studies.”

Based on their initial findings, the authors believe that dendritome mapping is a highly sensitive technique that could help to identify patterns in the morphology of single neurons that are linked to specific diseases or disorders. In the future, they hope that their approach will help to better understand specific neurological or neuropsychiatric disorders, potentially informing the development of new therapeutic interventions.

“We are currently extending our study in several directions,” said Yang.

“First, we are adapting our MORF approach to not only confer sparse and bright labeling of genetically defined neurons in the mouse brain, but also enable genetic perturbations (e.g. gain- or loss-of-function studies). The latter may allow us to investigate the causal relationships between genes and neuronal morphology and connectivity, and to identify novel modifier genes for disease pathologies.”

Drawing from their initial results, Yang, Park and their colleagues are now further expanding their MORF imaging technique and analysis toolbox. Their enriched toolkit could soon allow researchers to also image MORF-labeled neurons in whole tissue-cleared mouse brains via a technique known as light-sheet microscopy, and to analyze large neuronal image datasets using novel bioinformatic tools.

“We are also applying the MORF technologies and dendritome mapping to study pathogenesis of neurodegenerative disorders, including Alzheimer’s disease and Huntington’s disease,” added Yang.

The content is provided for information purposes only.