Three-dimensional autoradiographic localization of quench-corrected glycine receptor specific activity in the mouse brain using 3H- strychnine as the ligand

White Wf, S O'Gorman, Anna Wang RoeJournal of Neuroscience 1990 · DOI

Abstract

The autoradiographic analysis of neurotransmitter receptor distribution is a powerful technique that provides extensive information on the localization of neurotransmitter systems. Computer methodologies are described for the analysis of autoradiographic material which include quench correction, 3-dimensional display, and quantification based on anatomical boundaries determined from the tissue sections. These methodologies are applied to the problem of the distribution of glycine receptors measured by 3H-strychnine binding in the mouse CNS. The most distinctive feature of this distribution is its marked caudorostral gradient. The highest densities of binding sites within this gradient were seen in somatic motor and sensory areas; high densities of binding were seen in branchial efferent and special sensory areas. Moderate levels were seen in nuclei related to visceral function. Densities within the reticular formation paralleled the overall gradient with high to moderate levels of binding. The colliculi had low and the diencephalon had very low levels of binding. No binding was seen in the cerebellum or the telencephalon with the exception of the amygdala, which had very low levels of specific binding. This distribution of glycine receptors correlates well with the known functional distribution of glycine synaptic function. These data are illustrated in 3 dimensions and discussed in terms of the significance of the analysis techniques on this type of data as well as the functional significance of the distribution of glycine receptors.

Introduction

In 1990, researchers published a landmark study in the Journal of Neuroscience that advanced neuroanatomical mapping by combining early computer-assisted imaging with rigorous biochemical correction. The goal was to visualize the 3D landscape of glycine receptors in the mouse brain using $^3H$-strychnine as a ligand.

At the time, receptor autoradiography—applying radioactive tags to brain slices—was a standard technique. However, it faced two major hurdles:

  1. The Quenching Problem: Tritium ($^3H$) emits low-energy particles easily absorbed (quenched) by lipids. Consequently, lipid-rich white matter appeared to have fewer receptors than gray matter, even if densities were identical.
  2. The 2D Limitation: Brain structures are complex 3D shapes. Single 2D slices can miss peak densities or fail to represent the overall volume and boundaries of a nucleus.

Methodology

The authors developed a quench-correction algorithm and a method for 3D reconstruction from serial brain sections to address these issues.

Quench Correction and Quantification

To account for signal loss, a "quench map" was created by labeling sections with $^3H$-succinimidyl propionate, which binds uniformly to proteins. By comparing this signal to a non-quenched standard ($^{125}I$), a correction factor was calculated for every pixel. Specific activity was converted from optical density using Amersham step-wedge and rat whole-brain paste standards.

Color-coded maps of glycine receptor density and quench correction factors.

3D Reconstruction and Anatomical Registration

Serial sections (120 $\mu m$ apart) were digitized. The researchers used "isodensity contours"—topographic maps of receptor density—to build 3D models. Anatomical boundaries from stained tissue were overlaid onto these maps, allowing for the calculation of average receptor density for specific nuclei across their entire 3D volume.

Key Insights and Results

The study revealed a dramatic caudorostral (tail-to-head) gradient of glycine receptors. While GABA is the primary inhibitory neurotransmitter in the forebrain, glycine dominates the "lower" brain.

Functional Mapping

The highest receptor densities were found in somatic motor and sensory areas, such as the ventral layers of the spinal cord, the hypoglossal nucleus, and the facial nucleus. This aligns with glycine's role in mediating fast inhibition for precise movement and sensory processing.

3D isodensity contours illustrating the spatial distribution of glycine receptors.

Regional Distribution

  • Spinal Cord and Medulla: Contained the highest densities. The cervical spinal cord had nearly three times the binding sites of more rostral trigeminal nerve nuclei.
  • Forebrain: Binding was virtually absent in the cerebellum and telencephalon (cortex/striatum).
  • The Amygdala Exception: A small but specific population of receptors was found in the amygdala, suggesting a unique role for glycinergic inhibition in limbic processing.

Impact of Correction

Quench correction was shown to be critical. Tritium detection efficiency varied from 85% in gray matter to as low as 20% in white matter. Without this correction, uncorrected autoradiography could lead to incorrect biological conclusions regarding relative receptor densities.

Discussion and Future Directions

The authors noted that 3D data is inherently complex, making it difficult to summarize without losing anatomical nuance. They predicted that future electronic storage methods would be essential for handling such high-volume datasets.

While the 3D models were limited by the 120 $\mu m$ sampling interval, the study laid the groundwork for comparing different receptor systems (e.g., Glycine vs. GABA) or comparing receptor density with uptake sites within the same 3D space. This work remains a foundational example of how computational methods can refine our understanding of neurotransmitter systems.

Summary updated

Abstract

The autoradiographic analysis of neurotransmitter receptor distribution is a powerful technique that provides extensive information on the localization of neurotransmitter systems. Computer methodologies are described for the analysis of autoradiographic material which include quench correction, 3-dimensional display, and quantification based on anatomical boundaries determined from the tissue sections. These methodologies are applied to the problem of the distribution of glycine receptors measured by 3H-strychnine binding in the mouse CNS. The most distinctive feature of this distribution is its marked caudorostral gradient. The highest densities of binding sites within this gradient were seen in somatic motor and sensory areas; high densities of binding were seen in branchial efferent and special sensory areas. Moderate levels were seen in nuclei related to visceral function. Densities within the reticular formation paralleled the overall gradient with high to moderate levels of binding. The colliculi had low and the diencephalon had very low levels of binding. No binding was seen in the cerebellum or the telencephalon with the exception of the amygdala, which had very low levels of specific binding. This distribution of glycine receptors correlates well with the known functional distribution of glycine synaptic function. These data are illustrated in 3 dimensions and discussed in terms of the significance of the analysis techniques on this type of data as well as the functional significance of the distribution of glycine receptors.
BibTeX
@article{wf1990three,
  author = {White Wf and S O'Gorman and Anna Wang Roe},
  title = {{Three-dimensional autoradiographic localization of quench-corrected glycine receptor specific activity in the mouse brain using 3H- strychnine as the ligand}},
  journal = {Journal of Neuroscience},
  year = {1990},
  doi = {10.1523/jneurosci.10-03-00795.1990},
}
Figure