Characterization of Neocortical Principal Cells and Interneurons by Network Interactions and Extracellular Features

Péter Barthó, Hajime Hirase, Lénaı̈c Monconduit, Michaël Zugaro, Kenneth D. Harris, György BuzsákiJournal of Neurophysiology 2004 · DOI

TL;DR

This paper introduces a novel method for distinguishing between excitatory principal neurons and inhibitory interneurons in the neocortex using high-density extracellular recordings and spike cross-correlation analysis. By identifying direct synaptic interactions, they create templates to classify a larger population of neurons based on spike duration, offering a crucial tool for understanding local cortical circuit function.

What Problem Does This Paper Address?

Understanding how neural circuits compute requires knowing the identity and function of individual neurons. In the neocortex, principal cells (largely excitatory) and interneurons (largely inhibitory) play distinct roles, but experimentally distinguishing them based solely on extracellular recordings has been challenging. While methods exist for the hippocampus, a reliable approach for the neocortex, especially for classifying neurons based on their network interactions, was lacking. This paper aims to bridge that gap by developing a method to identify and classify these fundamental cell types within local cortical circuits.

Methods

The researchers employed a 64-site, two-dimensional silicon probe for high-density extracellular recordings in layer 5 of the rat somatosensory and prefrontal cortices. This allowed for simultaneous recording of multiple neurons and, crucially, the determination of their approximate two-dimensional spatial positions.

The core of their method relies on spike cross-correlation analysis. By analyzing the temporal relationship between the spikes of simultaneously recorded neuron pairs, they identified short-latency (<3 ms) interactions.

  • Putative excitatory (pyramidal) cells were identified by significant, short-latency peaks in their cross-correlograms, indicating they directly excited another neuron.
  • Putative inhibitory interneurons were identified by significant suppression of spiking in their partners, indicating they inhibited other neurons.

These physiologically identified neurons served as templates. The researchers then used these templates to classify the broader population of recorded neurons. They tested various extracellular features for classification and found that spike duration (specifically, the duration of the unfiltered spike from trough to peak) was the most reliable parameter for distinguishing between the two main cell types.

Datasets

The study utilized data from acute and chronic recordings in Sprague-Dawley rats.

  • Acute recordings: Performed under urethane anesthesia.
  • Chronic recordings: Performed in drug-free, behaving rats using wire tetrodes.

The paper does not introduce new benchmark datasets but rather analyzes data collected using their novel recording and analysis techniques.

Key Insights and Ideas

  • Direct synaptic interactions are key identifiers: The paper demonstrates that identifying direct, short-latency monosynaptic connections is a powerful way to functionally classify neurons in vivo.
  • Spike duration as a reliable classifier: Contrary to some previous assumptions, spike duration alone, when measured carefully, proved to be a surprisingly effective and reliable feature for distinguishing between putative pyramidal cells and interneurons in the neocortex.
  • Interneurons are crucial network hubs: The study revealed that a significant proportion of identified interneurons suppressed the activity of multiple target cells, highlighting their critical role in shaping network dynamics.
  • Urethane anesthesia preserves fast synaptic interactions: The findings in anesthetized rats were largely consistent with those in freely moving animals, suggesting that urethane anesthesia does not dramatically alter the fast monosynaptic connections crucial for this classification method.
  • Local circuits are highly interconnected: The analysis of functional connectivity provided insights into the dense local wiring within layer 5, showing both uni- and bidirectional connections between principal cells and interneurons.

Key Results

  • Out of approximately 60,000 recorded cell pairs, only 0.2% showed robust short-term interactions.
  • Putative excitatory (pyramidal) cells had significantly wider spike durations (0.86 ± 0.17 ms) compared to putative interneurons (0.43 ± 0.27 ms).
  • Approximately 40% of narrow-spike neurons (putative interneurons) were involved in inhibitory interactions (either inhibiting others or being excited by a partner), suggesting a high degree of functional connectivity for this group.
  • Classification based on spike duration resulted in a bimodal distribution, with approximately 7% of units classified as interneurons.
  • The study identified both excitatory and inhibitory connections between neurons recorded on the same shank and between neurons on different shanks, providing a detailed map of local circuit interactions.

Limitations and Future Work

  • Anesthesia effects: While the study found anesthesia didn't drastically alter fast synaptic interactions, it acknowledges that firing patterns might be influenced, and future work in behaving animals is crucial.
  • Classification errors: The authors admit that spike duration-based classification is not perfect and can lead to commission and omission errors, particularly for neurons with unusual firing dynamics or those not fitting neatly into the "narrow" or "wide" spike categories.
  • Limited scope: The study focused on layer 5 of the somatosensory and prefrontal cortices. Extending this methodology to other cortical layers and brain regions is a natural next step.
  • Further characterization: The paper suggests that combining spike duration with other parameters like firing rate, spike waveform, and auto-correlogram features could further refine classification and identify subclasses of neurons.
  • Functional circuit analysis: The authors emphasize that this method provides the tools for studying functional connectivity in local circuits and how these interactions are modified by behavioral states and experience.

Summary updated

Abstract

Most neuronal interactions in the cortex occur within local circuits. Because principal cells and GABAergic interneurons contribute differently to cortical operations, their experimental identification and separation is of utmost important. We used 64-site two-dimensional silicon probes for high-density recording of local neurons in layer 5 of the somatosensory and prefrontal cortices of the rat. Multiple-site monitoring of units allowed for the determination of their two-dimensional spatial position in the brain. Of the approximately 60,000 cell pairs recorded, 0.2% showed robust short-term interactions. Units with significant, short-latency (<3 ms) peaks following their action potentials in their cross-correlograms were characterized as putative excitatory (pyramidal) cells. Units with significant suppression of spiking of their partners were regarded as putative GABAergic interneurons. A portion of the putative interneurons was reciprocally connected with pyramidal cells. Neurons physiologically identified as inhibitory and excitatory cells were used as templates for classification of all recorded neurons. Of the several parameters tested, the duration of the unfiltered (1 Hz to 5 kHz) spike provided the most reliable clustering of the population. High-density parallel recordings of neuronal activity, determination of their physical location and their classification into pyramidal and interneuron classes provide the necessary tools for local circuit analysis.
BibTeX
@article{bartho2004characterization,
  author = {Péter Barthó and Hajime Hirase and Lénaı̈c Monconduit and Michaël Zugaro and Kenneth D. Harris and György Buzsáki},
  title = {{Characterization of Neocortical Principal Cells and Interneurons by Network Interactions and Extracellular Features}},
  journal = {Journal of Neurophysiology},
  year = {2004},
  doi = {10.1152/jn.01170.2003},
}
Figure