Pyramidal cell types drive functionally distinct cortical activity patterns during decision-making

Simon Musall, Xiaonan Richard Sun, Hemanth Mohan, Xu An, Steven Gluf, Shu-Jing Li, Rhonda Drewes, Emma Cravo, Irene Lenzi, Chaoqun Yin, Björn M. Kampa, Anne K. ChurchlandbioRxiv 2021 · DOI

TL;DR

This study reveals that different types of pyramidal neurons (PyNs) in the cortex exhibit distinct patterns of activity and play specialized roles in auditory decision-making. By combining advanced imaging techniques with optogenetic manipulation, the researchers demonstrate that these PyN subtypes form functionally segregated subnetworks that contribute uniquely to sensory processing and choice formation.

What Problem Does This Paper Address?

The brain's complex behavior arises from the coordinated activity of its constituent cells. While inhibitory neurons have been extensively studied, the functional diversity of excitatory pyramidal neurons (PyNs) – which make up ~80% of cortical neurons and are crucial for long-range communication – remains less understood. It's unclear whether functional differences observed within a single cortical area extend across the entire cortex and how these differences manifest in large-scale neural dynamics during complex tasks like decision-making. This paper aims to bridge this gap by investigating the cortex-wide activity patterns and functional roles of distinct PyN subtypes during an auditory decision task.

Methods

The researchers employed a multi-pronged approach:

  • Genetic and Retrograde Labeling: They used novel knock-in mouse lines (Fezf2-CreER for Pyramidal Tract (PT) neurons and PlexinD1-CreER for Intratelencephalic (IT) neurons) to specifically express calcium indicators (GCaMP6s) in these PyN populations. They also used a retrograde viral approach to label corticostriatal (CStr) projection neurons.
  • Widefield Calcium Imaging: This technique allowed them to measure neural activity across large areas of the dorsal cortex in awake, behaving mice performing an auditory decision task.
  • Dimensionality Reduction Techniques: Semi-nonnegative matrix factorization (sNMF) and localized sNMF (LocaNMF) were used to extract key spatial and temporal components of the cortex-wide activity, revealing underlying dynamics and complexity.
  • Optogenetic Inactivation: To causally assess the role of different PyN types, they selectively inhibited specific PyN populations in different cortical areas using the inhibitory opsin stGtACR2.
  • Encoding and Decoding Models: These models were used to analyze how neural activity relates to task variables (stimulus, choice) and to predict animal choices from neural activity, respectively.

Datasets

The study utilized data from genetically engineered mice. No external datasets were explicitly mentioned for training or evaluation.

Key Insights and Ideas

  • PyN Subtypes Have Unique Cortex-Wide Dynamics: Even at a large scale, different PyN types exhibit distinct patterns of neural activity, suggesting specialized subnetworks throughout the cortex.
  • PT Neurons Play a Surprisingly Large Causal Role in Sensory Processing: While all PyNs were recruited for auditory perception, PT neurons, which project to subcortical areas, were found to have the largest causal role in parietal cortex.
  • Frontal PyNs are Crucial for Decision-Making, with Distinct Choice Tuning: All PyN types in the frontal cortex are necessary for accurate choices, but they exhibit different patterns of activity related to the decision itself.
  • CStr Neurons Drive Ipsilateral Choice Preference in Specific Frontal Areas: A subset of corticostriatal neurons showed a preference for ipsilateral choices, particularly in the anterior lateral motor cortex (ALM), suggesting a specific role in motor output.
  • Functional Specialization Extends to Localized Subregions: PyN-specific activity patterns were often confined to smaller subregions within broader cortical areas, hinting at a finer-grained organization than traditionally defined areas.

Key Results

  • Dimensionality: PT neurons showed lower dimensionality in their cortex-wide activity compared to IT neurons, suggesting they are more functionally homogeneous.
  • Sensory Responses: PyN types differed in the location and lateralization of their sensory responses in parietal and frontal cortices. For example, PT neurons showed more posteromedial responses in parietal cortex compared to IT neurons.
  • Choice Signals: Frontal cortex exhibited strong choice-related activity across all PyN types, with PT and EMX mice showing more prominent signals than IT mice.
  • Optogenetic Inactivation: Inactivation of auditory and parietal cortex strongly impaired performance in the auditory decision task, with PT neurons in parietal cortex showing the most significant impact. Frontal cortex inactivation broadly impaired decision accuracy across all PyN types.
  • CStr Neuron Specificity: CStr neurons in ALM showed a higher fraction of ipsilateral-selective cells compared to unlabeled neurons, supporting their role in ipsilateral choice preference.

Limitations and Future Work

  • Movement Artifacts: The authors acknowledge that movement can obscure neural signals and employed strategies to mitigate this, but it remains a challenge in interpreting brain-wide activity.
  • Granularity of PyN Subtypes: While the study differentiates between PT, IT, and CStr neurons, the cortex contains many more PyN subtypes. Future work could explore these finer distinctions.
  • Area-Specific Roles: The study focused on parietal and frontal cortices. Future research could investigate the roles of other cortical areas and their interactions with these PyN types.
  • Causal Role of IT Neurons: While IT neurons were found to be necessary for accurate choices, their specific causal role in sensory processing and choice formation warrants further investigation, especially considering the heterogeneity within IT populations.
  • Interactions Between PyN Types: The study highlights distinct dynamics but future work could focus on how these PyN subnetworks interact to perform computations.

Summary updated

Abstract

Abstract Understanding how cortical circuits generate complex behavior requires investigating the cell types that comprise them. Functional differences across pyramidal neuron (PyN) types have been observed within cortical areas, but it is not known whether these local differences extend throughout the cortex, nor whether additional differences emerge when larger-scale dynamics are considered. We used genetic and retrograde labeling to target pyramidal tract (PT), intratelencephalic (IT) and corticostriatal projection neurons and measured their cortex-wide activity. Each PyN type drove unique neural dynamics, both at the local and cortex-wide scale. Cortical activity and optogenetic inactivation during an auditory decision task also revealed distinct functional roles: all PyNs in parietal cortex were recruited during perception of the auditory stimulus, but, surprisingly, PT neurons had the largest causal role. In frontal cortex, all PyNs were required for accurate choices but showed distinct choice-tuning. Our results reveal that rich, cell-type-specific cortical dynamics shape perceptual decisions.
BibTeX
@misc{musall2021pyramidal,
  author = {Simon Musall and Xiaonan Richard Sun and Hemanth Mohan and Xu An and Steven Gluf and Shu-Jing Li and Rhonda Drewes and Emma Cravo and Irene Lenzi and Chaoqun Yin and Björn M. Kampa and Anne K. Churchland},
  title = {{Pyramidal cell types drive functionally distinct cortical activity patterns during decision-making}},
  year = {2021},
  doi = {10.1101/2021.09.27.461599},
  howpublished = {bioRxiv},
}
Figure