Simultaneous 3D Cellular Positioning and Apical Dendritic Morphology of Transgenic Fluorescent Mouse CA3 Hippocampal Pyramidal Neurons

Abstract

Background: Pyramidal neurons of the hippocampus are not homogeneous in their structure or function, yet the structure and function of individual neurons are well-correlated. Despite these well-known facts, few structural studies of pyramidal neuronal dendrites in the hippocampus have controlled for precise cellular position within the hippocampus along its dorsoventral (longitudinal), tangential (proximodistal), or radial axes. We believe this is largely due to technical limitations that limit positional information about the neuron, limit throughput, and limit the control of confounding variables. New method: Here, we present a simple approach to address these limitations effectively in mouse CA3, adapted from our prior work in the mouse cerebral cortex. Our approach tracks the dorsoventral, tangential, and radial positions of neurons within the hippocampus, increases the number of available neurons for analysis, and provides several new controls. Results: We demonstrate how topographic and morphological data are related to one another among mouse CA3 pyramidal neurons, in a pilot data set providing proof-of-concept. Comparison with existing methods: Other common methods of labeling neurons, such as biocytin fills or Golgi-Cox stains, are labor- or time-intensive. Here, we validate the use of the transgenic fluorescent Thy1-GFP-M line, so the labeling step is practically eliminated. Other methods section the hippocampus coronally, which distorts the dorsoventral, tangential, and radial axes. Here, we preserve all three axes. Some existing methods do not collect all sections in order, instead focusing on "more dorsal" or "more ventral" sections, for example. We preserve much more dorsoventral information. Recent work has shown that CA2 is better defined by immunohistochemical rather than neuroanatomical markers. We use that immunohistochemistry here to increase precision in defining tangential position. Other studies have not measured other, critical variables and relationships that could confound the results. Here, we assess many more of these variables and relationships to increase the rigor of our work. Conclusions: We developed a more rigorous method for preserving precise cellular positioning information among reconstructed mouse hippocampal pyramidal neurons. Although we provide several innovations in our method, each innovation identified here could independently improve the rigor and reproducibility of a wide variety of approaches for understanding the morphological heterogeneity of pyramidal neurons in the brain.

Integrin β3 regulates apical dendritic morphology of pyramidal neurons throughout hippocampal CA3

“…This allowed us to make exact transverse sections of hippocampus at 100 μm with a VT1000S Vibratome (Leica Microsystems), taking care to preserve the dorsoventral position of each section. We then immunostained the sections for GFP (to enhance GFP signal) and PCP4 (to label CA2 pyramidal neurons), and imaged as described previously (Handwerk et al, 2022 ). We collected information from four WT hippocampi (two left and two right) and five cKO hippocampi (three left and two right).…”

“…We collected information from four WT hippocampi (two left and two right) and five cKO hippocampi (three left and two right). To determine dorsoventral topographic information, we imaged 166 total sections: 80 sections from WT (18, 16, 24, and 22 sections from WT hippocampi 1–4, respectively; some WT2 data were reported in Handwerk et al, 2022 ) and 86 sections from cKO (25, 12, 23, 8, and 18 sections from cKO hippocampi 1–5, respectively).…”

“…We 3D reconstructed neurons using neuTube (Feng et al, 2015 ) and Fiji software (Schindelin et al, 2012 ) with the Simple Neurite Tracer plugin (Longair et al, 2011 ), as described previously (Handwerk et al, 2022 ). To determine radial and proximodistal topographic information, and to determine morphological information, we imaged, reconstructed, and analyzed a total of 121 neurons from 62 sections (54 WT neurons from 26 WT sections and 67 cKO neurons from 36 cKO sections), creating .swc files.…”

“…Knowing that there may be considerable morphological variation along the dorsoventral, proximodistal, and radial axes of the hippocampus (Handwerk et al, 2022;Hunt et al, 2018;Sun et al, 2020), we also controlled for the precise cellular position of each CA3 neuron we analyzed. We also controlled for potential clustering of data collected from multiple hippocampi.…”

“…We also compared the apical dendritic morphology of Thy1 ‐GFP‐M‐labeled CA3 hippocampal pyramidal neurons in conditional knockouts (“cKO” where Itgb3 is deleted from chiefly hippocampal and cortical excitatory neurons and astrocytes) to Thy1 ‐GFP‐M‐labeled CA3 hippocampal pyramidal neurons in controls. Knowing that there may be considerable morphological variation along the dorsoventral, proximodistal, and radial axes of the hippocampus (Handwerk et al, 2022 ; Hunt et al, 2018 ; Sun et al, 2020 ), we also controlled for the precise cellular position of each CA3 neuron we analyzed. We also controlled for potential clustering of data collected from multiple hippocampi.…”