Principles for Dendritic Spine Size and Density in Human and Mouse Cortical Pyramidal Neurons

Abstract

Dendritic spines of pyramidal neurons are the targets of most excitatory synapses in the cerebral cortex, and dendritic spine morphology directly reflects their function. However, there are scarce data available regarding both the detailed morphology of these structures for the human cerebral cortex and the extent to which they differ in comparison with other species. Thus, in the present study, we used intracellular injections of Lucifer yellow to reconstruct-in three dimensions-the morphology of basal dendritic spines from pyramidal cells in the human and mouse CA1 hippocampal region and compared these spines with those of the human temporal and cingular cortex. We found that human hippocampal dendrites show lower spine density, larger volume, and longer length of dendritic spines than mouse CA1 spines. Furthermore, human hippocampal dendrites show higher spine density, smaller spine volume, and shorter length compared to dendritic spines from the human temporal and cingular cortex. This morphological diversity suggests an equally large variability of synaptic strength and learning rules across these brain regions in humans and between humans and mice. Nevertheless, a balance between size and density was found in all cases, which may be a cortical rule maintained across cortical areas and species.

Keywords: 3D reconstructions; F factor; JEL Classification: RRID:AB_1062582; RRID:SCR_002798; RRID:SCR_007370; RRID:SCR_016788; RRID:SCR_017348; RRID:SCR_020233; cerebral cortex; dendrites; intracellular injections; morphology; pyramidal cells.

FIGURE 1

(A–D) Confocal microscopy z projection images of LY‐labeled pyramidal neurons in the mouse (A) and human (B) CA1 field of the hippocampal formation and the human temporal (C) and cingular (D) cortex. (E–H) 3D reconstruction of the same cells shown in A–D. (I–L) High‐magnification confocal microscopy images showing labeled basal dendritic segments in the mouse (I) and human (J) CA1 field of the hippocampal formation and the human temporal (K) and cingular (L) cortex. (M–P) Each spine and shaft was 3D reconstructed by selecting a particular threshold that represented a solid surface that matched the contour of the spine (red) and shaft (white). Scale bar (in P) = 50 µm in A–H and 6 µm in I–P.

FIGURE 2

(A–F) High‐magnification confocal microscopy images showing an example of a labeled cingular basal dendritic segment in (A) to illustrate the 3D reconstruction of (B) spines (red) and shaft (white), (C) spines color coded by size (volume: blue‐red: 0.01 ‐ 0.7 µm³) and shaft (white), (D and E) selected spine heads (red) and neck lengths (white lines), and (F) selected spine heads shown in D color coded by size (area: blue‐red: 0.5 ‐ 4.0 µm²) and neck lengths shown in E. (G and H) Higher magnification images to illustrate neck diameter measurements. (I) Schematic of spine variables analyzed. Scale bar (in H) = 2 µm in A–F and 0.7 µm in G and H.

FIGURE 3

Graphs showing spine density (A–C), spine volume (D–F), and spine length (G–I) as mean values (A, D, G), frequency distribution (B, E, H), and as a function of the distance from soma (C, F, I) for the basal dendrites from mouse and human CA1 region and human temporal and cingular cortical areas. Measurements are reported as mean ± SEM. The statistical significance of the differences is shown in Tables S1–S3. Source: (A–C) Spine density values from CA1 regions were taken from Benavides‐Piccione et al. (2024), and values from the Hcing region were taken from Benavides‐Piccione et al. (2013). (D–I) Spine volume and length values from the Hcing region were taken from Benavides‐Piccione et al. (2013).

FIGURE 4

Graphs showing spine head area (A and B), spine neck length (C and D), and spine neck diameter (E and F) as mean values (A, C, E) and frequency distribution (B, D, F) for the basal dendrites from mouse and human CA1 region and human temporal and cingular cortical areas. Measurements are reported as mean ± SEM. The statistical significance of the differences is shown in Tables S1 and S2. Source: Spine values from the Htemp region were taken from Eyal et al. (2018).