3D synaptic organization of layer III of the human anterior cingulate and temporopolar cortex

Abstract

The human anterior cingulate and temporopolar cortices have been proposed as highly connected nodes involved in high-order cognitive functions, but their synaptic organization is still basically unknown due to the difficulties involved in studying the human brain. Using Focused Ion Beam/Scanning Electron Microscopy (FIB/SEM) to study the synaptic organization of the human brain obtained with a short post-mortem delay allows excellent results to be obtained. We have used this technology to analyze layer III of the anterior cingulate cortex (Brodmann area 24) and the temporopolar cortex, including the temporal pole (Brodmann area 38 ventral and dorsal) and anterior middle temporal gyrus (Brodmann area 21). Our results, based on 6695 synaptic junctions fully reconstructed in 3D, revealed that Brodmann areas 24, 21 and ventral area 38 showed similar synaptic density and synaptic size, whereas dorsal area 38 displayed the highest synaptic density and the smallest synaptic size. However, the proportion of the different types of synapses (excitatory and inhibitory), the postsynaptic targets, and the shapes of excitatory and inhibitory synapses were similar, regardless of the region examined. These observations indicate that certain aspects of the synaptic organization are rather homogeneous, whereas others show specific variations across cortical regions.

Keywords: FIB-SEM; autopsy; cerebral cortex; postsynaptic targets; ultrastructure.

Fig. 1

Cortical and FIB/SEM sampling regions. (A–D) Nissl-stained sections to illustrate the cytoarchitectonic differences between regions from an autopsy case. Cortical layer delimitation based on Palomero-Gallagher et al. (2008) for (A), Alonso-Nanclares et al. (2008) in (B), and Ding et al. (2009) for (C) and (D). Illustration of the analyzed FIB/SEM sampling regions (superimposed as dark trapezoids in A–D). (E–G) Correlative light/electron microscopy analyses of layer III neuropil. (E) 1-μm-thick semithin section stained with toluidine blue, which is adjacent to the block for FIB/SEM imaging (F). (F) SEM image illustrating the block surface. (G) Higher magnification of (F), showing the trenches made in the neuropil to acquire the FIB/SEM stacks of images. White arrowheads in (E), (F), and (G) point to the same blood vessel, allowing the exact location of the region of interest to be identified. (H) SEM image at higher magnification showing the front of a trench (white asterisk in G) made to acquire an FIB/SEM stack of images. (I) FIB/SEM image from a stack of images. Two synapses are indicated (arrows). Scale bar (in D) indicates 320 μm for (A)–(D); 200 μm for (E)–(F); 80 μm in (G); 15 μm in (H); 1.20 μm in (I). BA: Brodmann area; d: dorsal; v: ventral.

Fig. 2

Stereological estimation of the volume fraction occupied by different cortical elements in BA24, vBA38, dBA38, and BA21. (A) Low-power micrograph of a toluidine blue-stained 1.5-μm-thick semithin section of vBA38. (B) High-power magnification of the boxed area in (A) showing a 50 μm × 50 μm grid superimposed on the original micrograph, where points hitting the different cortical elements were counted. Some blood vessels (arrowheads) and cell bodies (arrows) are indicated. (C) Plot of the volume fraction occupied by each analyzed cortical element. BA24 shows a higher volume fraction occupied by cells than the other cortical regions (χ², P < 0.001). Scale bar (in B) indicates 40 μm in (A), and 25 μm in (B). BA: Brodmann area; d: dorsal; v: ventral.

Fig. 3

Images obtained by FIB/SEM showing neuropil of vBA38 (A), and BA21 (B) from an autopsy case (AB3). Some examples of synapses are indicated in all regions (arrowheads). Scale bar (in B) indicates 1 μm in (A-B). BA: Brodmann area; d: dorsal; v: ventral.

Fig. 4

Three-dimensional analysis of synapses. Identification and annotation of synapses. (A–D) Screenshots of the EspINA software user interface. (A) In the main window, the sections are viewed through the xy plane (as obtained by FIB/SEM microscopy). The other two orthogonal planes, yz and xz, are also shown in adjacent windows (on the right). (B) The 3D window shows the three orthogonal planes and the 3D reconstruction of asymmetric (green) and symmetric (red) synaptic junctions. (C) 3D reconstructed synaptic junction displayed by colors. (D) Computed synaptic apposition surface for each reconstructed synaptic junction (yellow). Scale bar (in D) indicates 2.6 μm, for (B)–(D).

Fig. 5

Study of the morphology of the synapses. 3D synaptic morphology representation. (A, B) Representative examples of SAS of asymmetric synapses (A, green) and symmetric synapses (B, red). Analyses of SAS were distributed into 20 bins of equal size (an example of each bin has been illustrated). (C) Schematic representation of the shape of the synaptic junctions: macular synapses, with continuous disk-shaped PSD; perforated synapses, with holes in the PSD; horseshoe-shaped synapses, with tortuous horseshoe-shaped perimeter with an indentation; and fragmented synapses, with two or more PSDs with no connections between them. (D) Proportion of macular, perforated, horseshoe-shaped, and fragmented AS in BA24, vBA38, dBA38, and BA21. BA21 showed a higher proportion of complex-shaped synapses (including perforated, horseshoe, and fragmented synapses) than BA24, vBA38, and dBA38 (χ²; P < 0.0001). Scale bar (in B) indicates 500 nm in (A) and (B). BA: Brodmann area; d: dorsal; SAS: synaptic apposition surface; v: ventral.

Fig. 6

Synaptic density. Mean synaptic density of the neuropil from layer III of BA21, BA24, vBA38, and dBA38. Each dot represents a single autopsy case according to the colored key on the right. dBA38 shows a significantly higher synaptic density than the other analyzed regions (ANOVA; P < 0.05). BA: Brodmann area; d: dorsal; v: ventral.

Fig. 7

Serial images obtained by FIB/SEM showing a dendritic segment with a spine (blue). Asymmetric synapses established on the spine head (green arrow; A–L) and the dendritic shaft (green arrowhead; B–G) are indicated. Scale bar (in O) = 1 μm.

Fig. 8

Study of the postsynaptic targets. Schematic representation of the asymmetric and symmetric synapses on the different postsynaptic targets (spine head, spine neck and dendritic shaft) from analyzed synapses in layer III of BA24 (A), vBA38 (B), dBA38 (C), and BA21 (D). Spine head percentages include both complete and incomplete spines. Dendritic shaft percentages include dendritic shafts with and without spines. No significant differences (χ²; P > 0.05) were found between cortical regions. AS have been represented in green and SS in red. BA: Brodmann area; d: dorsal; v: ventral.

Fig. 9

Study of the postsynaptic targets-multisynaptic spines. Schematic representation of spine heads (including both complete and incomplete spines) receiving single and multiple synapses in BA24, vBA38, dBA38, and BA21. Percentages of each type are indicated (absolute numbers of synapses are in parentheses). AS have been represented in green and SS in red. Spines with more than one AS (including all combinations) were more frequent in BA24 than in the other regions (χ², P < 0.05). AS: asymmetric synapses; BA: Brodmann area; d: dorsal; SS: symmetric synapses; v: ventral.