TY - JOUR
AU - Lübke, Joachim H, R
AB - Abstract Studies of synapses are available for different brain regions of several animal species including non-human primates, but comparatively little is known about their quantitative morphology in humans. Here, synaptic boutons in Layer 5 (L5) of the human temporal lobe (TL) neocortex were investigated in biopsy tissue, using fine-scale electron microscopy, and quantitative three-dimensional reconstructions. The size and organization of the presynaptic active zones (PreAZs), postsynaptic densities (PSDs), and that of the 3 distinct pools of synaptic vesicles (SVs) were particularly analyzed. L5 synaptic boutons were medium-sized (~6 μm2) with a single but relatively large PreAZ (~0.3 μm2). They contained a total of ~1500 SVs/bouton, ~20 constituting the putative readily releasable pool (RRP), ~180 the recycling pool (RP), and the remainder, the resting pool. The PreAZs, PSDs, and vesicle pools are ~3-fold larger than those of CNS synapses in other species. Astrocytic processes reached the synaptic cleft and may regulate the glutamate concentration. Profound differences exist between synapses in human TL neocortex and those described in various species, particularly in the size and geometry of PreAZs and PSDs, the large RRP/RP, and the astrocytic ensheathment suggesting high synaptic efficacy, strength, and modulation of synaptic transmission at human synapses. electron microscopy, human neocortex, pools of synaptic vesicles, quantitative 3D reconstructions, synaptic boutons Introduction Synapses are key elements in the establishment, maintenance, and termination of synaptic transmission between neurons in any network of the brain. Numerous studies have compared dendritic morphology of human (DeFelipe 2011; Mohan et al. 2015), non-human primates (NHPs) (Elston et al. 2011), and rodent cortices (Elston and Manger 2014), but much less information is available regarding the presynaptic boutons, in particular at the fine-scale ultrastructural level. Quantitative studies of synaptic geometry are available for different brain regions of several animal species (for example, Harris and Sultan 1995; Spirou et al. 1998; Xu-Friedman et al. 2001; Sätzler et al. 2002; Xu-Friedman and Regehr 2003; Marrone et al. 2005; Rollenhagen et al. 2007; Popov and Stewart 2009; Wilke et al. 2013; Rollenhagen et al. 2015; Rodriguez-Moreno et al. 2017; reviewed by Rollenhagen and Lübke 2006) including NHPs (Bopp et al. 2017; Hsu et al. 2017), but are comparatively lacking for humans (but see for example, Cragg 1976; Gibson 1983; Kirkpatrick et al. 2006; Alonso-Nanclares et al. 2008; Blazquez-Llorca et al. 2013; Kay et al. 2013; Bernhardt et al. 2013; Liu and Schumann 2014). In particular, no comprehensive and coherent studies about their quantitative geometry exist. This can be partially attributed to the limited availability of human brain tissue with excellent ultrastructural preservation that can be achieved only with biopsy material but not post-mortem brains. However, such detailed quantitative and structural descriptions are required to fully understand the functional dynamics underlying the computational properties of synapses. One possible way to adequately address these issues is the use of high-resolution fine-scale electron microscopy (EM) and computer-assisted quantitative three-dimensional (3D) reconstructions of synaptic structures, based on serial digital electron micrographs, taking advantage of non-affected (non-epileptic) neocortical access tissue provided from temporal lobe (TL) epilepsy surgery. Interestingly, TL represents ~17% of the cerebral cortex in humans (Kiernan 2012), and exhibits a more complex cytoarchitectoral organization than that observed in rodents and NHPs (Elston et al. 2001; DeFelipe 2011), as regards its thickness, larger size of neurons (Mohan et al. 2015), density of synaptic inputs per neuron (DeFelipe et al. 2002; DeFelipe 2011), receptor architectonic (Zilles 2015), and the composition of the astrocytic network. TL is also a typical and thus representative example of a more homotypic granular, 6-layered associative neocortex as described in humans (von Economo and Koskinas 1925; Zilles et al. 2015), and NHPs (Elston et al. 2001; DeFelipe 2011; Kiernan 2012). This is in contrast to the motor cortex characterized by its heterotypic agranular cytoarchitecture, a common feature in human, and NHPs (Zilles et al. 1996). The primary sensory cortices, including somatosensory cortex, are also granular but heterotypic in humans (Roland and Zilles 1996). Here, we provided, for the first time, a comprehensive quantitative EM analysis of the structural characteristics of excitatory synaptic boutons terminating on dendritic segments, the majority of which originate from Layer 5 (L5) pyramidal neurons in the human TL neocortex. We focused on structural parameters underlying synaptic transmission and plasticity, such as the size, shape, and number of presynaptic active zones (PreAZs), PSDs as well as the organization and size of the 3 functional pools of synaptic vesicles (SVs), namely the readily releasable pool (RRP), the recycling pool (RP), and the resting pool. A quantitative morphological analysis of these structural parameters is a necessary pre-requisite to directly compare human synapses with those in various animal species including NHPs where data are already available. Moreover, quantitative 3D models of synapses provide the basis for any type of simulation of various aspects of synaptic function that remain only partially accessible to experiment. Human synapses display profound differences in their structural composition compared with their counterparts in different animal species and NHPs, namely in the shape and size of the PreAZs, PSDs and in the 3 functionally defined pools of SVs. Hence, the present study contributes to an improved understanding of synapses embedded in different networks of the brain and across species. Material and Methods All experimental procedures were approved by the Ethical Committees of the Rheinische Friedrich-Wilhelms-University/University Hospital Bonn (ethic votum of the Medical Faculty to Prof. Dr med. Johannes Schramm and Prof. Dr rer. nat. Joachim Lübke, Nr. 146/11), the University of Bochum (ethic votum of the Medical Faculty to PD Dr med. Marec von Lehe and Prof. Dr rer. nat. Joachim Lübke), and the Research Committee of the Research Centre Jülich GmbH. This study complied with the guidelines laid out in the EU directive (2004/23/EC) regarding the work with human tissue used for experimental and scientific purposes. Fixation and Tissue Processing for 3D Volume Reconstructions Neocortical brain tissue from 3 male and 4 female patients aged between 20 and 50 years who underwent epilepsy surgery was used in this study (see Supplementary Table 1 for clinical details of the patients). All patients suffered from drug-resistant TL epilepsy. The pre-surgical work-up comprised at least high-resolution MRI together with long-term video-EEG-monitoring. In all cases, the circumscribed epileptic focus was located in the hippocampus proper, but not in the surrounding neocortical regions of the TL. During surgery, blocks of neocortical tissue from the temporo-lateral or temporo-basal regions were sampled from the approach, remote from the epileptogenic area that was resected to control the seizures (Supplementary Fig. 1A) and for histological inspection by neuropathologists. The tissue was taken always far from the epileptic focus and may thus be regarded as non-affected (non-epileptic) as also demonstrated by other studies using the same experimental approach (Alonso-Nanclares et al. 2008, 2011; Navarrete et al. 2013; Mohan et al. 2015; Molnár et al. 2016). It has to be noted though that the only possible ‘healthy’ tissue that may serve as controls is post-mortem material from subjects without any brain traumata or pathologies. However, the ultrastructural quality (preservation) of such material is not suitable for our study, namely the analysis of the quantitative geometry of cortical synapses, as it shows severe distortions of relevant structural features (e.g. membranes of synaptic boutons, PreAZs and PSDs and SVs) for the quantitative 3D synaptic models (Lübke, personal observation). Moreover, as no significant interindividual differences were found in the synaptic parameters analyzed (see Supplementary Fig. 3), one might speculate that our results were not affected by the disease, although patients had different clinical backgrounds and in the onset of epilepsy (Supplementary Table 1). Other recent studies using the same experimental approach also came to the same conclusion and discarded the effect of the treatments and disease condition (Alonso-Nanclares et al. 2008, 2011; Testa-Silva et al. 2010, 2014; Navarrete et al. 2013; Mohan et al. 2015; Molnár et al. 2016). Immediately after their removal, tissue blocks were immersion-fixed in ice-cold 4% paraformaldehyde (PFA) and 2.5% glutaraldehyde (GA) diluted in 0.1 M phosphate buffer (PB, pH 7.4) for 24–48 h at 4°C. The fixative was changed twice during this period. After extensive washing in PB, serial vibratome sections of 150 or 200 μm in thickness were cut in the frontal (coronal) plane through the human TL neocortex. Sections were then post-fixed for 30–60 min in 0.5 or 1% osmium tetroxide (Sigma, Munich, Germany) diluted in PB-buffered sucrose (300 mOsm, pH 7.4) at room temperature in the dark. After several washing steps in PB (10 min for each step), they were dehydrated in an ascending series of ethanol starting at 20% to absolute ethanol (10 min for each step and absolute ethanol, 30 min twice), followed by a brief incubation (2 min twice) in propylene oxide (Fluka, Neu-Ulm, Germany). Sections were then transferred into a mixture of propylene oxide and DurcupanTM resin (2:1, 1:1 for 1 h each; Fluka, Neu-Ulm, Germany) and stored overnight in pure resin. The next day, sections were flat-embedded on coated glass slides in fresh DurcupanTM, coverslipped and polymerized at 60°C for 2 days. After light microscopic (LM) inspection, a tissue block containing the region of interest (ROI; Supplementary Fig. 1B) was glued on a pre-polymerized block and trimmed. Semithin sections were cut with a Leica UltracutS ultramicrotome (Leica Microsystems, Vienna, Austria), stained with methylene-blue to identify the cortical layers (Supplementary Fig. 1C), in particular L4 and L5 (Supplementary Fig. 1D), and examined with LM and photographed using a motorized Olympus BX61 microscope equipped with the Olympus CellSense analysis hard- and software (Olympus GmbH, Hamburg, Germany). Then, serial ultrathin sections were cut with a Leica UltracutS (Leica Microsystems, Wetzlar, Germany) through the determined ROI of L5, and photographed using the EM. Glutamine Synthetase Immunohistochemistry To examine the astrocytic coverage at L5 synaptic boutons and their target structures, 2 human tissue blocks were immersion-fixed as described above (identified by asterisks in Supplementary Table 1). After post-fixation, 100-μm thick vibratome sections were cut in the frontal (coronal) plane. Sections were then cryoprotected in PB-buffered 10% (30 min), 20% (30 min), and 30% sucrose overnight. They were then freeze-thawed in liquid nitrogen (1 min), rinsed in PB, blocked in PB-buffered saline (PBS) containing 0.5% goat serum albumin (1.5 h), and finally incubated in a monoclonal mouse antiglutamine synthetase antibody (1:2000; Chemicon Europe, Hampshire, UK) overnight at 4°C. The next day, sections were thoroughly washed in PBS and then incubated in biotinylated antimouse secondary antibody for 2 h (1:100; Vector, Linaris, Wertheim, Germany). Several washing steps in PBS were followed by incubation in PBS-buffered ABC-elite kit solution for 2 h (1:100; Vector, Linaris, Wertheim, Germany). Thereafter, sections were transferred to 0.05 M Tris-buffered saline (TBS) and reacted in 3-3′-diaminobenzidine (DAB; 0.5 mg/ml) diluted in TBS containing 0.03% H2O2 for 10 min. Reaction was stopped by transferring them to TBS and back to PBS. After LM inspection, sections were post-fixed in sucrose-PBS-buffered 0.5% osmium tetroxide (30 min), dehydrated through an ascending series of ethanol, propylene oxide, and finally flat-embedded in DurcupanTM as described above. Ultrathin Sectioning and Data Acquisition Serial ultrathin sections (50 ± 5 nm in thickness, silver to light gray interference contrast appearance) were cut on a Leica UltracutS ultramicrotome and collected on Formvar or Pioloform-coated slot copper grids. Individual series comprising 65–165 ultrathin sections were prepared to allow the reconstruction of numerous synaptic boutons and their target structures. Prior to EM examination, sections were stained with 5% aqueous uranyl acetate for 15–20 min and lead citrate for 3–5 min according to Reynolds (1963) to enhance the contrast of biological membranes at the EM level. One or 2 ROIs, within a series of ultrathin sections through L5, were selected choosing an area containing a maximum of well-preserved synaptic complexes (a presynaptic bouton and its postsynaptic target structure). ROIs in serial sections were then photographed with a Zeiss Libra 120 (Fa. Zeiss, Oberkochen, Germany) equipped with a bottom-mounted Proscan 2 K digital camera using the SIS Multi Images Acquisition software (Olympus Soft Imaging System, Münster, Germany) at a primary magnification of 8000×. In addition, other digital images were taken at various magnifications for further documentation and illustration. All images were stored in a database until further use. Selected LM and EM images were processed using Adobe PhotoshopTM and Adobe IllustratorTM software packages. 3D Volume Reconstructions and Quantitative Analysis of L5 Synaptic Boutons Electron micrographs composing each series were imported, stacked, and aligned in the reconstruction software OpenCAR (Contour Alignment Reconstruction; for details see Sätzler et al. 2002). The main goal of this study was to quantify several morphological parameters representing structural correlates of synaptic transmission and plasticity in L5. Excitatory synaptic boutons were characterized by large round SVs and prominent PreAZs and PSDs in contrast to putative GABAergic terminals that have smaller, more oval-shaped SVs, and thin postsynaptic densities. The following structural parameters were analyzed: (1) surface area and volume of synaptic boutons; (2) volume of mitochondria; (3) surface area of the PreAZs; see also Dufour et al. 2016 and the PSDs; 2 apposed membrane specializations separated by the synaptic cleft; (4) number and diameter of clear synaptic and dense-core vesicles; and (5) distance of individual SVs from the PreAZ for the structural definition of the RRP, RP, and resting pool. Presynaptic boutons, their mitochondria as well as their postsynaptic target structures were outlined on the outer edge of their membranes, using closed contour lines throughout the entire stack within a series (Supplementary Fig. 2). A synaptic bouton was considered completely captured, when it was possible to follow the axon in both directions through the entire series (en passant bouton) or the enlargement of the axon leading to an end terminal bouton (Fig. 1B, C). The PreAZs and PSDs were regarded as complete when their perimeters were entirely reconstructed in a series of ultrathin sections. The surface areas of the PreAZ and PSD were computed separately by first generating a 3D surface model of the synaptic bouton. The PreAZ was then measured by extracting this area from the reconstructed presynaptic bouton membrane that was covered by this membrane specialization (i.e. where the contour line coincided with <30 nm distance from the presynaptic membrane). Hence, the length (L) of the PreAZ (L PreAZ) and the surface area (SA) of the PreAZ (SA PreAZ) is already known. The size of the PSD opposing the PreAZ was estimated under the following assumptions: (1) both membrane specializations, PreAZ and PSD run parallel to each other at the pre- and postsynaptic apposition zone; (2) for both membrane specializations a contour line was drawn determining their actual length (L PreAZ and L PSD). Hence, the surface area of the PSD (SA PSD) is estimated by the following equation: SAPSD=SAPre∗LPSD/LPreAZ which is the perimeter ratio between the outlines of the PSD to that of the synaptic contact. Figure 1 View largeDownload slide Postsynaptic innervation patterns of synaptic boutons in L5 of the human TL neocortex. (A) Large dendritic shaft (sh) with a dense innervation of synaptic boutons (b1–b3, b6–b8) at different dendritic locations and 2 boutons (b4, b5) terminating on the same spine (asterisk). Scale bar 1 μm. (B) Typical example of an en passant synaptic bouton (b) terminating on a spine with a prominent spine apparatus (sp) with 2 PreAZs and PSDs (marked by red arrowheads) on the swelling and along the length of the axon. Scale bar 0.4 μm. (C), Two end terminal boutons (b1; b2) as an enlargement of the axon forming terminal synaptic contacts (red arrowheads) on 2 different caliber dendrites (d1; d2). Scale bar 0.3 μm. Figure 1 View largeDownload slide Postsynaptic innervation patterns of synaptic boutons in L5 of the human TL neocortex. (A) Large dendritic shaft (sh) with a dense innervation of synaptic boutons (b1–b3, b6–b8) at different dendritic locations and 2 boutons (b4, b5) terminating on the same spine (asterisk). Scale bar 1 μm. (B) Typical example of an en passant synaptic bouton (b) terminating on a spine with a prominent spine apparatus (sp) with 2 PreAZs and PSDs (marked by red arrowheads) on the swelling and along the length of the axon. Scale bar 0.4 μm. (C), Two end terminal boutons (b1; b2) as an enlargement of the axon forming terminal synaptic contacts (red arrowheads) on 2 different caliber dendrites (d1; d2). Scale bar 0.3 μm. Measurements of the width of the synaptic cleft were performed on random EM images taken from the series using the SIS Analysis software. Only synapses cut perpendicular to the PreAZ and PSD were included in the sample. The distance between the outer edge of the pre- and postsynaptic membranes was measured at the 2 lateral edges and at the center of the synapse; the 2 values of the lateral edges were averaged for each synapse and a mean ± SD was calculated for each PreAZ and PSDs and the data were pooled to obtain a total mean ± SD over all PreAZs and PSDs. To estimate the number and size of the clear synaptic and dense-core vesicles (DCVs), all vesicles were marked throughout each synaptic bouton and their diameters were measured individually. To determine the distribution profile of the vesicles, the minimal distance between each vesicle membrane to the contour lines of the PreAZ on the boutons membrane was measured throughout the bouton in every image of the series. To avoid double counts, only clear ring-like structures were counted as vesicles. Furthermore, vesicles might be missed in densely packed regions, because ring-like traces partly overlap. This effect may counteract any double counts. Based on the small extent and the partially counteracting nature of this effect, the numbers of small clear vesicles reported in this study remained uncorrected (Rollenhagen et al. 2015). For large DCVs, double-counts were excluded by careful examination of adjacent images and were only counted in the image where they appeared largest. Cluster Analysis of Synaptic Parameters Cluster analyses were performed using MATLAB and Statistics Toolbox Release 2016b (The MathWorks, Inc., Natick, MA, United States of America), Python and Python package Scikit-learn (Pedregosa et al. 2011). All synaptic boutons reconstructed in this study (n = 147) were initially analyzed using hierarchical cluster analysis (HCA; Rokach and Maimon 2005) and K-means clustering (Hartigan and Wong 1979). Twenty structural parameters were tested: bouton volume, bouton surface area, number of PreAZs and PSDs/bouton, number of mitochondria, volume of mitochondria, percentage of mitochondrial volume from bouton volume, number of SVs, total vesicular volume, percentage of vesicular volume from bouton volume, SV diameter, number of SVs at distances (perimeters) 10/20/30/40/60/70/80/90/100/200/300/400/500 nm from the PreAZs and PreAZ and PSD surface area. Subsequently, a principal component analysis (PCA) (Pearson 1901; Hotelling 1933) was performed on this dataset to simplify it by applying a rank reduction, in order to detect the principal parameters which characterize the synaptic boutons (see Fig. 7). PCA is a statistical procedure that uses an orthogonal transformation to convert a set of observations of possibly correlated variables into a set of values of linearly uncorrelated variables called principal components (Abdi and Williams 2010). The PCA showed a first component predominantly defined by the PreAZs and PSDs surface area and a second component predominantly defined by the vesicle pools at different perimeters from the PreAZs. Then, we used the simplified dataset composed by these 2 principle components to perform the cluster analysis using independent K-means clustering and HCA, which are methods for unsupervised learning, since the original dataset was not labeled. All Matlab codes can be retrieved under the following link: https://github.com/YachaoShao/SynapseClustering. Statistical Analysis For each parameter, the data were expressed by the median with the 1st and 3rd quartile, a mean value ± SD with the correlation coefficient (R2) and the respective P-value <0.05 was considered significant. From the numerous 3D volume reconstructions and spreadsheets computed by OpenCAR, statistical summaries and graphs were generated. To look for interindividual differences for each structural parameter, data distributions were analyzed using box plots (see Supplementary Fig. 3). As some of the structural parameters analyzed were not normally distributed as indicated by the skewness values (see Tables 1 and 2), the non-parametric Kruskal–Wallis H-test analysis was computed, using InStat (GraphPad Software Inc., USA) to look for differences between the human tissue blocks for the structural parameters analyzed. Correlation graphs between several parameters were also generated (see Fig. 6). Table 1 Quantitative analysis of structural elements of L5 synaptic boutons in the human TL neocortex Patient identity Bouton Synaptic contact Mitochondria PreAZ PSD Cleft width (nm) ± SD No. of boutons Surface (μm2) ± SD Volume (μm3) ± SD Number ± SD Surface area (μm2) ± SD Surface area (μm2) ± SD Lateral Central Volume (μm3) ± SD % of volume Hu_040 211_I 35 5.59 ± 2.80 0.51 ± 0.38 1.04 ± 0.20 0.19 ± 0.12 0.21 ± 0.19 12.79 ± 3.29 17.17 ± 3.35 0.06 ± 0.04 10.85 ± 4.52 CV — 0.50 0.75 0.19 0.75 0.75 0.25 0.19 0.75 0.75 Hu_110 217_I 9 5.85 ± 8.21 0.94 ± 1.76 1 0.22 ± 0.21 0.20 ± 0.20 18.46 ± 2.82 20.12 ± 3.81 0.31 ± 0.67 13.88 ± 9.17 CV — 1.85 1.85 0 0.95 1.01 0.15 0.18 2.14 0.66 Hu_110 505_I 16 4.65 ± 2.85 0.34 ± 0.23 1 0.20 ± 0.21 0.24 ± 0.23 15.57 ± 5.05 15.72 ± 1.87 0.05 ± 0.03 11.13 ± 6.33 CV — 0.61 0.66 0 1.01 0.95 0.32 0.11 0.63 0.56 Hu_120 301_II 9 6.05 ± 5.86 0.72 ± 0.87 1 — — 17.51 ± 4.58 24.84 ± 9.26 — — CV — 0.97 1.21 0 — — 0.26 0.37 — — Hu_120 706_I 9 6.46 ± 3.78 0.60 ± 0.51 1 — — 17.41 ± 7.91 17.53 ± 3.45 — — CV — 0.58 0.86 0 — — 0.45 0.19 — — Hu_150 626_I 17 6.37 ± 3.17 0.63 ± 0.34 1 0.32 ± 0.22 0.50 ± 0.41 19.31 ± 2.23 18.85 ± 5.79 0.07 ± 0.03 11.89 ± 3.74 CV — 0.49 0.53 0 0.69 0.83 0.11 0.30 0.48 0.31 Hu_200 511_II 52 7.67 ± 5.79 0.67 ± 0.63 1.06 ± 0.24 0.22 ± 0.13 0.23 ± 0.13 19.65 ± 2.57 19.12 ± 3.31 0.13 ± 0.12 12.46 ± 4.82 CV — 0.75 0.93 0.23 0.58 0.58 0.13 0.17 0.92 0.38 Mean; Median; IQR — 6.09; 6.05; 0.87 0.63; 0.63; 0.21 1.01; 1; 0.04 0.23; 0.22; 0.07 0.27; 0.23; 0.16 17.24; 17.51; 3.74 19.05; 18.85; 2.95 0.12; 0.07; 0.16 12.04; 11.89; 2.18 Variance — 23.04 0.46 0.11 0.03 0.06 20.73 30.84 50.30 23.04 Skewness — 4.49a 2.05 5.59a 1.86 2.77 1.14 1.82 8.22* 0.57 Patient identity Bouton Synaptic contact Mitochondria PreAZ PSD Cleft width (nm) ± SD No. of boutons Surface (μm2) ± SD Volume (μm3) ± SD Number ± SD Surface area (μm2) ± SD Surface area (μm2) ± SD Lateral Central Volume (μm3) ± SD % of volume Hu_040 211_I 35 5.59 ± 2.80 0.51 ± 0.38 1.04 ± 0.20 0.19 ± 0.12 0.21 ± 0.19 12.79 ± 3.29 17.17 ± 3.35 0.06 ± 0.04 10.85 ± 4.52 CV — 0.50 0.75 0.19 0.75 0.75 0.25 0.19 0.75 0.75 Hu_110 217_I 9 5.85 ± 8.21 0.94 ± 1.76 1 0.22 ± 0.21 0.20 ± 0.20 18.46 ± 2.82 20.12 ± 3.81 0.31 ± 0.67 13.88 ± 9.17 CV — 1.85 1.85 0 0.95 1.01 0.15 0.18 2.14 0.66 Hu_110 505_I 16 4.65 ± 2.85 0.34 ± 0.23 1 0.20 ± 0.21 0.24 ± 0.23 15.57 ± 5.05 15.72 ± 1.87 0.05 ± 0.03 11.13 ± 6.33 CV — 0.61 0.66 0 1.01 0.95 0.32 0.11 0.63 0.56 Hu_120 301_II 9 6.05 ± 5.86 0.72 ± 0.87 1 — — 17.51 ± 4.58 24.84 ± 9.26 — — CV — 0.97 1.21 0 — — 0.26 0.37 — — Hu_120 706_I 9 6.46 ± 3.78 0.60 ± 0.51 1 — — 17.41 ± 7.91 17.53 ± 3.45 — — CV — 0.58 0.86 0 — — 0.45 0.19 — — Hu_150 626_I 17 6.37 ± 3.17 0.63 ± 0.34 1 0.32 ± 0.22 0.50 ± 0.41 19.31 ± 2.23 18.85 ± 5.79 0.07 ± 0.03 11.89 ± 3.74 CV — 0.49 0.53 0 0.69 0.83 0.11 0.30 0.48 0.31 Hu_200 511_II 52 7.67 ± 5.79 0.67 ± 0.63 1.06 ± 0.24 0.22 ± 0.13 0.23 ± 0.13 19.65 ± 2.57 19.12 ± 3.31 0.13 ± 0.12 12.46 ± 4.82 CV — 0.75 0.93 0.23 0.58 0.58 0.13 0.17 0.92 0.38 Mean; Median; IQR — 6.09; 6.05; 0.87 0.63; 0.63; 0.21 1.01; 1; 0.04 0.23; 0.22; 0.07 0.27; 0.23; 0.16 17.24; 17.51; 3.74 19.05; 18.85; 2.95 0.12; 0.07; 0.16 12.04; 11.89; 2.18 Variance — 23.04 0.46 0.11 0.03 0.06 20.73 30.84 50.30 23.04 Skewness — 4.49a 2.05 5.59a 1.86 2.77 1.14 1.82 8.22* 0.57 Summary of structural parameters that have been extracted from the detailed 3D-reconstructions of L5 synaptic boutons in the human TL neocortex. Means ± SD (upper row in each box), Median (1st value lower row) and Interquartile Range (IQR; 2nd value lower row) are given for individual patients from which the average mean ± SD, median and Interquartile Range for each parameter was calculated. CVs were calculated for individual patients and were also given for the pooled data to explain the large variability of some of the structural parameters investigated. aMeasures with a skew >3, indicating non-normal distributions. Lateral: for cleft width measurements were taken at the 2 lateral edges of the PreAZs and PSDs. A mean ± SD was calculated from the 2 values. Central: cleft width at the central region under the PreAZs and PSDs. View Large Table 1 Quantitative analysis of structural elements of L5 synaptic boutons in the human TL neocortex Patient identity Bouton Synaptic contact Mitochondria PreAZ PSD Cleft width (nm) ± SD No. of boutons Surface (μm2) ± SD Volume (μm3) ± SD Number ± SD Surface area (μm2) ± SD Surface area (μm2) ± SD Lateral Central Volume (μm3) ± SD % of volume Hu_040 211_I 35 5.59 ± 2.80 0.51 ± 0.38 1.04 ± 0.20 0.19 ± 0.12 0.21 ± 0.19 12.79 ± 3.29 17.17 ± 3.35 0.06 ± 0.04 10.85 ± 4.52 CV — 0.50 0.75 0.19 0.75 0.75 0.25 0.19 0.75 0.75 Hu_110 217_I 9 5.85 ± 8.21 0.94 ± 1.76 1 0.22 ± 0.21 0.20 ± 0.20 18.46 ± 2.82 20.12 ± 3.81 0.31 ± 0.67 13.88 ± 9.17 CV — 1.85 1.85 0 0.95 1.01 0.15 0.18 2.14 0.66 Hu_110 505_I 16 4.65 ± 2.85 0.34 ± 0.23 1 0.20 ± 0.21 0.24 ± 0.23 15.57 ± 5.05 15.72 ± 1.87 0.05 ± 0.03 11.13 ± 6.33 CV — 0.61 0.66 0 1.01 0.95 0.32 0.11 0.63 0.56 Hu_120 301_II 9 6.05 ± 5.86 0.72 ± 0.87 1 — — 17.51 ± 4.58 24.84 ± 9.26 — — CV — 0.97 1.21 0 — — 0.26 0.37 — — Hu_120 706_I 9 6.46 ± 3.78 0.60 ± 0.51 1 — — 17.41 ± 7.91 17.53 ± 3.45 — — CV — 0.58 0.86 0 — — 0.45 0.19 — — Hu_150 626_I 17 6.37 ± 3.17 0.63 ± 0.34 1 0.32 ± 0.22 0.50 ± 0.41 19.31 ± 2.23 18.85 ± 5.79 0.07 ± 0.03 11.89 ± 3.74 CV — 0.49 0.53 0 0.69 0.83 0.11 0.30 0.48 0.31 Hu_200 511_II 52 7.67 ± 5.79 0.67 ± 0.63 1.06 ± 0.24 0.22 ± 0.13 0.23 ± 0.13 19.65 ± 2.57 19.12 ± 3.31 0.13 ± 0.12 12.46 ± 4.82 CV — 0.75 0.93 0.23 0.58 0.58 0.13 0.17 0.92 0.38 Mean; Median; IQR — 6.09; 6.05; 0.87 0.63; 0.63; 0.21 1.01; 1; 0.04 0.23; 0.22; 0.07 0.27; 0.23; 0.16 17.24; 17.51; 3.74 19.05; 18.85; 2.95 0.12; 0.07; 0.16 12.04; 11.89; 2.18 Variance — 23.04 0.46 0.11 0.03 0.06 20.73 30.84 50.30 23.04 Skewness — 4.49a 2.05 5.59a 1.86 2.77 1.14 1.82 8.22* 0.57 Patient identity Bouton Synaptic contact Mitochondria PreAZ PSD Cleft width (nm) ± SD No. of boutons Surface (μm2) ± SD Volume (μm3) ± SD Number ± SD Surface area (μm2) ± SD Surface area (μm2) ± SD Lateral Central Volume (μm3) ± SD % of volume Hu_040 211_I 35 5.59 ± 2.80 0.51 ± 0.38 1.04 ± 0.20 0.19 ± 0.12 0.21 ± 0.19 12.79 ± 3.29 17.17 ± 3.35 0.06 ± 0.04 10.85 ± 4.52 CV — 0.50 0.75 0.19 0.75 0.75 0.25 0.19 0.75 0.75 Hu_110 217_I 9 5.85 ± 8.21 0.94 ± 1.76 1 0.22 ± 0.21 0.20 ± 0.20 18.46 ± 2.82 20.12 ± 3.81 0.31 ± 0.67 13.88 ± 9.17 CV — 1.85 1.85 0 0.95 1.01 0.15 0.18 2.14 0.66 Hu_110 505_I 16 4.65 ± 2.85 0.34 ± 0.23 1 0.20 ± 0.21 0.24 ± 0.23 15.57 ± 5.05 15.72 ± 1.87 0.05 ± 0.03 11.13 ± 6.33 CV — 0.61 0.66 0 1.01 0.95 0.32 0.11 0.63 0.56 Hu_120 301_II 9 6.05 ± 5.86 0.72 ± 0.87 1 — — 17.51 ± 4.58 24.84 ± 9.26 — — CV — 0.97 1.21 0 — — 0.26 0.37 — — Hu_120 706_I 9 6.46 ± 3.78 0.60 ± 0.51 1 — — 17.41 ± 7.91 17.53 ± 3.45 — — CV — 0.58 0.86 0 — — 0.45 0.19 — — Hu_150 626_I 17 6.37 ± 3.17 0.63 ± 0.34 1 0.32 ± 0.22 0.50 ± 0.41 19.31 ± 2.23 18.85 ± 5.79 0.07 ± 0.03 11.89 ± 3.74 CV — 0.49 0.53 0 0.69 0.83 0.11 0.30 0.48 0.31 Hu_200 511_II 52 7.67 ± 5.79 0.67 ± 0.63 1.06 ± 0.24 0.22 ± 0.13 0.23 ± 0.13 19.65 ± 2.57 19.12 ± 3.31 0.13 ± 0.12 12.46 ± 4.82 CV — 0.75 0.93 0.23 0.58 0.58 0.13 0.17 0.92 0.38 Mean; Median; IQR — 6.09; 6.05; 0.87 0.63; 0.63; 0.21 1.01; 1; 0.04 0.23; 0.22; 0.07 0.27; 0.23; 0.16 17.24; 17.51; 3.74 19.05; 18.85; 2.95 0.12; 0.07; 0.16 12.04; 11.89; 2.18 Variance — 23.04 0.46 0.11 0.03 0.06 20.73 30.84 50.30 23.04 Skewness — 4.49a 2.05 5.59a 1.86 2.77 1.14 1.82 8.22* 0.57 Summary of structural parameters that have been extracted from the detailed 3D-reconstructions of L5 synaptic boutons in the human TL neocortex. Means ± SD (upper row in each box), Median (1st value lower row) and Interquartile Range (IQR; 2nd value lower row) are given for individual patients from which the average mean ± SD, median and Interquartile Range for each parameter was calculated. CVs were calculated for individual patients and were also given for the pooled data to explain the large variability of some of the structural parameters investigated. aMeasures with a skew >3, indicating non-normal distributions. Lateral: for cleft width measurements were taken at the 2 lateral edges of the PreAZs and PSDs. A mean ± SD was calculated from the 2 values. Central: cleft width at the central region under the PreAZs and PSDs. View Large Table 2 Synaptic vesicle pools in L5 synaptic boutons in the human TL neocortex Patient identity Pool size Total volume Putative RRP/bouton Putative recycling pool 60–200 nm Putative reserve pool/bouton >200 nm Total number of synaptic vesicles Minimum of synaptic vesicles Maximum of synaptic vesicles Vesicle diameter (nm) (μm3) (%) p10 (%) p20 (%) Mean number/bouton (%) Mean number/bouton (%) Hu_040 211_I 1347.21 ± 1031.49 274 5769 35.68 ± 2.65 0.04 8.66 3.80 ± 3.50 0.28 10.37 ± 9.49 0.76 147.29 10.93 1150.76 85.41 Hu_110 217_I 1968.56 ± 2521.67 142 8437 38.16 ± 2.53 0.08 12.08 1 ± 2.12 0.05 4.55 ± 6.52 0.23 180.89 9.18 1728.89 87.82 Hu_110 505_I 1282.12 ± 713.93 226 2639 34.29 ± 2.75 0.03 10.09 12.11 ± 9.58 0.94 28.17 ± 21.34 2.19 191.12 14.90 1001.18 78.08 Hu_120 301_II — — — — — — — — — — — — — — Hu_120 706_I — — — — — — — — — — — — — — Hu_150 626_I 1696.13 ± 825.39 570 3080 38.33 ± 3.04 0.05 9.49 4.93 ± 4.93 0.29 19.43 ± 10.72 1.14 220.51 13.00 1396.25 82.31 Hu_200 511_II 1298.6 ± 1403.17 1082 5978 37.02 ± 4.59 0.05 9.60 5.28 ± 6.05 0.40 13.55 ± 12.05 1.04 169.50 13.05 1043.19 80.33 Mean; Median; IQR 1518.52 ± 303.18; 1347.21; 541.98 — — 36.69 ± 1.71; 37.02; 3.26 0.05 9.98 5.42 ± 4.09; 4.93; 6.29 0.39 15.21 ± 9.02; 13.55; 16.34 1.07 181.86 12.21 1264.07 82.79 CV 0.19 — — 0.04 — — 0.75 — 0.59 — 0.58 0.24 — Variance 1 655 452.24 — — 153.21 0.002 21.34 39.93 — 206.69 — 11 469.97 — 72 853.49 — Skewness 2.39 — — −2.07 3.60a −0.16 2.17 2.06 — 1.25 — 0.66 — Patient identity Pool size Total volume Putative RRP/bouton Putative recycling pool 60–200 nm Putative reserve pool/bouton >200 nm Total number of synaptic vesicles Minimum of synaptic vesicles Maximum of synaptic vesicles Vesicle diameter (nm) (μm3) (%) p10 (%) p20 (%) Mean number/bouton (%) Mean number/bouton (%) Hu_040 211_I 1347.21 ± 1031.49 274 5769 35.68 ± 2.65 0.04 8.66 3.80 ± 3.50 0.28 10.37 ± 9.49 0.76 147.29 10.93 1150.76 85.41 Hu_110 217_I 1968.56 ± 2521.67 142 8437 38.16 ± 2.53 0.08 12.08 1 ± 2.12 0.05 4.55 ± 6.52 0.23 180.89 9.18 1728.89 87.82 Hu_110 505_I 1282.12 ± 713.93 226 2639 34.29 ± 2.75 0.03 10.09 12.11 ± 9.58 0.94 28.17 ± 21.34 2.19 191.12 14.90 1001.18 78.08 Hu_120 301_II — — — — — — — — — — — — — — Hu_120 706_I — — — — — — — — — — — — — — Hu_150 626_I 1696.13 ± 825.39 570 3080 38.33 ± 3.04 0.05 9.49 4.93 ± 4.93 0.29 19.43 ± 10.72 1.14 220.51 13.00 1396.25 82.31 Hu_200 511_II 1298.6 ± 1403.17 1082 5978 37.02 ± 4.59 0.05 9.60 5.28 ± 6.05 0.40 13.55 ± 12.05 1.04 169.50 13.05 1043.19 80.33 Mean; Median; IQR 1518.52 ± 303.18; 1347.21; 541.98 — — 36.69 ± 1.71; 37.02; 3.26 0.05 9.98 5.42 ± 4.09; 4.93; 6.29 0.39 15.21 ± 9.02; 13.55; 16.34 1.07 181.86 12.21 1264.07 82.79 CV 0.19 — — 0.04 — — 0.75 — 0.59 — 0.58 0.24 — Variance 1 655 452.24 — — 153.21 0.002 21.34 39.93 — 206.69 — 11 469.97 — 72 853.49 — Skewness 2.39 — — −2.07 3.60a −0.16 2.17 2.06 — 1.25 — 0.66 — Summary of structural parameters relevant for release that have been extracted from the detailed 3D-reconstructions. The numbers in brackets behind the experimental reference number corresponds to the individual patient identification in the box plots of the Supplementary material. Means ± SD (upper row where appropriate) are given for individual patients from which the average mean value for each parameter was calculated. In addition, the Median (1st value lower row) and the Interquartile Range (IQR, 2nd value lower row) is given where appropriate. aMeasures with a skew >3, indicating non-normal distributions. View Large Table 2 Synaptic vesicle pools in L5 synaptic boutons in the human TL neocortex Patient identity Pool size Total volume Putative RRP/bouton Putative recycling pool 60–200 nm Putative reserve pool/bouton >200 nm Total number of synaptic vesicles Minimum of synaptic vesicles Maximum of synaptic vesicles Vesicle diameter (nm) (μm3) (%) p10 (%) p20 (%) Mean number/bouton (%) Mean number/bouton (%) Hu_040 211_I 1347.21 ± 1031.49 274 5769 35.68 ± 2.65 0.04 8.66 3.80 ± 3.50 0.28 10.37 ± 9.49 0.76 147.29 10.93 1150.76 85.41 Hu_110 217_I 1968.56 ± 2521.67 142 8437 38.16 ± 2.53 0.08 12.08 1 ± 2.12 0.05 4.55 ± 6.52 0.23 180.89 9.18 1728.89 87.82 Hu_110 505_I 1282.12 ± 713.93 226 2639 34.29 ± 2.75 0.03 10.09 12.11 ± 9.58 0.94 28.17 ± 21.34 2.19 191.12 14.90 1001.18 78.08 Hu_120 301_II — — — — — — — — — — — — — — Hu_120 706_I — — — — — — — — — — — — — — Hu_150 626_I 1696.13 ± 825.39 570 3080 38.33 ± 3.04 0.05 9.49 4.93 ± 4.93 0.29 19.43 ± 10.72 1.14 220.51 13.00 1396.25 82.31 Hu_200 511_II 1298.6 ± 1403.17 1082 5978 37.02 ± 4.59 0.05 9.60 5.28 ± 6.05 0.40 13.55 ± 12.05 1.04 169.50 13.05 1043.19 80.33 Mean; Median; IQR 1518.52 ± 303.18; 1347.21; 541.98 — — 36.69 ± 1.71; 37.02; 3.26 0.05 9.98 5.42 ± 4.09; 4.93; 6.29 0.39 15.21 ± 9.02; 13.55; 16.34 1.07 181.86 12.21 1264.07 82.79 CV 0.19 — — 0.04 — — 0.75 — 0.59 — 0.58 0.24 — Variance 1 655 452.24 — — 153.21 0.002 21.34 39.93 — 206.69 — 11 469.97 — 72 853.49 — Skewness 2.39 — — −2.07 3.60a −0.16 2.17 2.06 — 1.25 — 0.66 — Patient identity Pool size Total volume Putative RRP/bouton Putative recycling pool 60–200 nm Putative reserve pool/bouton >200 nm Total number of synaptic vesicles Minimum of synaptic vesicles Maximum of synaptic vesicles Vesicle diameter (nm) (μm3) (%) p10 (%) p20 (%) Mean number/bouton (%) Mean number/bouton (%) Hu_040 211_I 1347.21 ± 1031.49 274 5769 35.68 ± 2.65 0.04 8.66 3.80 ± 3.50 0.28 10.37 ± 9.49 0.76 147.29 10.93 1150.76 85.41 Hu_110 217_I 1968.56 ± 2521.67 142 8437 38.16 ± 2.53 0.08 12.08 1 ± 2.12 0.05 4.55 ± 6.52 0.23 180.89 9.18 1728.89 87.82 Hu_110 505_I 1282.12 ± 713.93 226 2639 34.29 ± 2.75 0.03 10.09 12.11 ± 9.58 0.94 28.17 ± 21.34 2.19 191.12 14.90 1001.18 78.08 Hu_120 301_II — — — — — — — — — — — — — — Hu_120 706_I — — — — — — — — — — — — — — Hu_150 626_I 1696.13 ± 825.39 570 3080 38.33 ± 3.04 0.05 9.49 4.93 ± 4.93 0.29 19.43 ± 10.72 1.14 220.51 13.00 1396.25 82.31 Hu_200 511_II 1298.6 ± 1403.17 1082 5978 37.02 ± 4.59 0.05 9.60 5.28 ± 6.05 0.40 13.55 ± 12.05 1.04 169.50 13.05 1043.19 80.33 Mean; Median; IQR 1518.52 ± 303.18; 1347.21; 541.98 — — 36.69 ± 1.71; 37.02; 3.26 0.05 9.98 5.42 ± 4.09; 4.93; 6.29 0.39 15.21 ± 9.02; 13.55; 16.34 1.07 181.86 12.21 1264.07 82.79 CV 0.19 — — 0.04 — — 0.75 — 0.59 — 0.58 0.24 — Variance 1 655 452.24 — — 153.21 0.002 21.34 39.93 — 206.69 — 11 469.97 — 72 853.49 — Skewness 2.39 — — −2.07 3.60a −0.16 2.17 2.06 — 1.25 — 0.66 — Summary of structural parameters relevant for release that have been extracted from the detailed 3D-reconstructions. The numbers in brackets behind the experimental reference number corresponds to the individual patient identification in the box plots of the Supplementary material. Means ± SD (upper row where appropriate) are given for individual patients from which the average mean value for each parameter was calculated. In addition, the Median (1st value lower row) and the Interquartile Range (IQR, 2nd value lower row) is given where appropriate. aMeasures with a skew >3, indicating non-normal distributions. View Large Results A total of 147 synaptic boutons were completely reconstructed out of 7 series of ultrathin sections (65–165 ultrathin sections/series) and analyzed regardless of their different target structures (dendritic segments or spines) in L5 of the TL neocortex using biopsy material, taken from patients that underwent epilepsy surgery (Supplementary Fig. 1A and Supplementary Table 1). In general, the neuronal organization of the human TL neocortex mirrors that of a typical granular neocortex as described for other cortical regions in humans (von Economo and Koskinas 1925; Zilles 2015). L5 is characterized by large pyramidal neurons representing the vast majority (~85%) of all neurons in this layer (Supplementary Fig. 1A and 1D). The remaining were GABAergic interneurons. It is noteworthy to mention that the biopsy samples contained several apoptotic neurons identifiable by their dark appearance, severe distortions of their cytoplasm, and the presence of microglia, indicative for cell death of these neurons (not shown). Also, a few degenerating boutons, characterized by their content of distorted organelles, were observed (Fig. 3G). Within the surrounding neuropil myelinated axons, often containing SVs were seen. Interestingly, numerous clusters of unmyelinated axons often associated with synaptic complexes were found (Fig. 3F). The majority of the dendritic neuropil (~80%) in L5 is composed of mainly basal dendrites of L5 pyramidal neurons (Markram, Lübke, Frotscher, Roth, et al. 1997; Markram, Lübke, Frotscher, Sakmann, et al. 1997; in humans: Mohan et al. 2015; reviewed by Ramaswamy and Markram 2015) that form a dense network in L5 (Fig. 1A). Another but much smaller fraction is ascending apical and apical oblique dendrites of L6 pyramidal neurons (Zhang and Deschênes 1997); the remainder represents GABAergic smooth dendrites which were excluded in our sample. Thousands of synaptic complexes (Figs 1, 2, 3, 8) were formed by either presynaptic en passant (Fig. 1B) or endterminal boutons (Fig. 1C) with their prospective postsynaptic target structures, which could be either basal or apical oblique dendrites of different calibers (Figs. 1A, C), spines of different sizes and types (mushroom, stubby or filopodial; Figs 1B, 2B, C, E, F, 3B–G, 4, 8B), or the cell body of a neuron (Fig. 3F). The majority of all spines (~65%) contained a spine apparatus, a specialized form of the endoplasmic reticulum (Figs 1B, 2E, 3B, D, E, 4A; see also discussion). Figure 2 View largeDownload slide Synaptic organization of L5 in human TL neocortex I. (A) A large dendritic spine (sp) with a non-perforated PreAZ and PSDs (arrowheads) that occupies approximately half of the pre- and postsynaptic apposition zone. Note the presence of comparable large clear vesicles (asterisks) as well as docked or fused vesicles at the PreAZ (arrowheads). Scale bar 0.5 μm. (B) A synaptic bouton establishing a putative synaptic contact with an astrocytic endfeed (as) as indicated by the large PreAZ (arrowheads). Scale bar 0.5 μm. (C) Synaptic bouton (b) terminating on a large spine (sp) with a perforated PreAZ and PSD (arrowheads) that nearly covers the entire pre- and postsynaptic apposition zone. The framed area marks the spine apparatus. Scale bar 0.5 μm. (D) Synaptic bouton (b) terminating on a large spine (sp). Note the presence of large SVs (asterisks), the fusion of a synaptic vesicle and the omega-shaped body at the PreAZ (framed area) Scale bar 0.5 μm. (E, F) High magnification of a synaptic bouton (transparent yellow) terminating on a mushroom spine (transparent blue) and its respective 3D-volume reconstruction (F). Note the protrusion of the membrane at the PreAZs and PSDs (red) forming a so-called coated pit shown at higher magnification in the inset. Scale bars 0.5 μm and 0.2 μm (inset). Figure 2 View largeDownload slide Synaptic organization of L5 in human TL neocortex I. (A) A large dendritic spine (sp) with a non-perforated PreAZ and PSDs (arrowheads) that occupies approximately half of the pre- and postsynaptic apposition zone. Note the presence of comparable large clear vesicles (asterisks) as well as docked or fused vesicles at the PreAZ (arrowheads). Scale bar 0.5 μm. (B) A synaptic bouton establishing a putative synaptic contact with an astrocytic endfeed (as) as indicated by the large PreAZ (arrowheads). Scale bar 0.5 μm. (C) Synaptic bouton (b) terminating on a large spine (sp) with a perforated PreAZ and PSD (arrowheads) that nearly covers the entire pre- and postsynaptic apposition zone. The framed area marks the spine apparatus. Scale bar 0.5 μm. (D) Synaptic bouton (b) terminating on a large spine (sp). Note the presence of large SVs (asterisks), the fusion of a synaptic vesicle and the omega-shaped body at the PreAZ (framed area) Scale bar 0.5 μm. (E, F) High magnification of a synaptic bouton (transparent yellow) terminating on a mushroom spine (transparent blue) and its respective 3D-volume reconstruction (F). Note the protrusion of the membrane at the PreAZs and PSDs (red) forming a so-called coated pit shown at higher magnification in the inset. Scale bars 0.5 μm and 0.2 μm (inset). Figure 3 View largeDownload slide Synaptic organization of L5 in the human TL neocortex II. (A), Dendro-dendritic synapse with 2 PreAZs and PSDs (asterisks) indicated by the pools of SVs between 2 neighboring dendrites (de1, de2). Scale bar 0.5 μm. (B), Synaptic bouton (b) terminating on the head of an elongated dendritic spine (sp; highlighted in transparent blue). Note the prominent spine apparatus (framed area) in the spine neck. The PreAZ and PSD are highlighted in red. Scale bar 0.5 μm. (C), 3D-volume reconstruction of the elongated spine (blue) and the PreAZ (red) shown in B. (D) Two synaptic boutons (b1, b2) terminating on 2 adjacent stubby spines (sp1, sp2). Note the occurrence of a spine apparatus (framed area) at sp1. PreAZs and PSDs are marked by arrowheads. Scale bar 0.5 μm. (E) Two synaptic boutons (b1, b2) terminating on the spine head (sp, b1) and shaft (sh) close to the spine neck of the same dendrite. The spine apparatus is marked by the framed area. Scale bar 0.25 μm. (F) Two adjacent synaptic boutons (b1, b2) one of which established a putative GABAergic synapse at the somatic region (so, b2) whereas the other is putative glutamatergic terminating on a neighboring spine head (sp, b1). Note that both synaptic boutons are isolated from each other by non-myelinated axons (asterisks). Scale bar 0.5 μm. (G) Typical example of a degenerating synaptic bouton (highlighted in transparent yellow) as identified by distorted internal organelles terminating on a dendrite (de). Scale bar 0.5 μm. Figure 3 View largeDownload slide Synaptic organization of L5 in the human TL neocortex II. (A), Dendro-dendritic synapse with 2 PreAZs and PSDs (asterisks) indicated by the pools of SVs between 2 neighboring dendrites (de1, de2). Scale bar 0.5 μm. (B), Synaptic bouton (b) terminating on the head of an elongated dendritic spine (sp; highlighted in transparent blue). Note the prominent spine apparatus (framed area) in the spine neck. The PreAZ and PSD are highlighted in red. Scale bar 0.5 μm. (C), 3D-volume reconstruction of the elongated spine (blue) and the PreAZ (red) shown in B. (D) Two synaptic boutons (b1, b2) terminating on 2 adjacent stubby spines (sp1, sp2). Note the occurrence of a spine apparatus (framed area) at sp1. PreAZs and PSDs are marked by arrowheads. Scale bar 0.5 μm. (E) Two synaptic boutons (b1, b2) terminating on the spine head (sp, b1) and shaft (sh) close to the spine neck of the same dendrite. The spine apparatus is marked by the framed area. Scale bar 0.25 μm. (F) Two adjacent synaptic boutons (b1, b2) one of which established a putative GABAergic synapse at the somatic region (so, b2) whereas the other is putative glutamatergic terminating on a neighboring spine head (sp, b1). Note that both synaptic boutons are isolated from each other by non-myelinated axons (asterisks). Scale bar 0.5 μm. (G) Typical example of a degenerating synaptic bouton (highlighted in transparent yellow) as identified by distorted internal organelles terminating on a dendrite (de). Scale bar 0.5 μm. Figure 4 View largeDownload slide Shape and size of PreAZs and PSDs. (A) Three synaptic boutons (b1–b3) terminating on 2 different mushroom-like (sp1–sp2) and a stubby (sp3) spine identified by the occurrence of a prominent spine apparatus (framed areas). Note the different sizes of the PreAZs and PSDs (arrowheads) at the contours of synaptic complexes and the surrounding neuropil. Scale bar 0.5 μm. (B) Corresponding 3D-volume reconstruction. Here, the outlines of the synaptic boutons are omitted to allow a direct view of the pool of SVs (green dots) and the PreAZs (red). (C1–C3) Higher magnification of the 3 PreAZs of the dendrite shown in B. Note the perforated nature of all PreAZs occupying ~25% of the total volume of the spine. Figure 4 View largeDownload slide Shape and size of PreAZs and PSDs. (A) Three synaptic boutons (b1–b3) terminating on 2 different mushroom-like (sp1–sp2) and a stubby (sp3) spine identified by the occurrence of a prominent spine apparatus (framed areas). Note the different sizes of the PreAZs and PSDs (arrowheads) at the contours of synaptic complexes and the surrounding neuropil. Scale bar 0.5 μm. (B) Corresponding 3D-volume reconstruction. Here, the outlines of the synaptic boutons are omitted to allow a direct view of the pool of SVs (green dots) and the PreAZs (red). (C1–C3) Higher magnification of the 3 PreAZs of the dendrite shown in B. Note the perforated nature of all PreAZs occupying ~25% of the total volume of the spine. Dendro-dendritic synapses, a highly specialized type of synapses, were also rarely observed (Fig. 3A). Furthermore, synaptic contacts between fine astrocytic processes and synaptic boutons were observed as described for the hippocampus and cerebellar climbing fibers (reviewed by Allen 2014; Papouin et al. 2017), but rarely (Fig. 2B). Quantitative Analysis of Synaptic Boutons in L5 of the Human TL Neocortex As already mentioned the majority of synaptic boutons in our samples (~85% of the total) were found on dendritic spines of different types (thin 12%; filopodial 18%; mushroom 65%; and stubby 5%), ~10% of which are boutons establishing more than one synaptic contacts on the same spine and numerous contacts with either the same or different dendrites (Fig. 1A). A huge variability was observed with respect to the shape (ovoid to round) and size of synaptic boutons, as indicated by the mean ± SD regardless of their target structures (see Table 1). Besides very large synaptic boutons (27.33 μm2; 5.89 μm3); also very small boutons (0.45 μm2; 0.02 μm3) existed with a mean surface area of ~6 μm2, ~0.6 μm3 in volume (Table 1). No significant difference in neither surface area nor volume of the boutons (P > 0.05) was found. A high correlation existed between the volume and the surface area of synaptic boutons as indicated by R2 = 0.74 (Fig. 6A). In larger synaptic boutons, several mitochondria (range 3–21; Table 1) of different shape and size (0.12 ± 0.09 μm3) were present, occupying ~12% of the total bouton volume, whereas smaller boutons contained no (Figs 1B and 2C) or only a single mitochondrion (Figs 1C and 4A). Interestingly, a strong correlation was found between the volume of the boutons with that of mitochondria (R2 = 0.83; Fig. 6B), suggesting an important role of mitochondria in the function of the bouton (see Discussion). Structural Composition of PreAZs and PSDs The number, size, and shape of the PreAZs and PSDs are key structural determinants in synaptic transmission and in the modulation of synaptic efficacy, strength, short-, and long-term plasticity (Südhof 2012; Matz et al. 2010; Wilhelm et al. 2014). The majority (~97%) of synaptic boutons in L5 contained only a single (Figs 1C, 3B–G, 2), at most 2 PreAZs and PSDs (Figs 1B). On average, PreAZs are 0.23 ± 0.05 μm2, and PSDs 0.29 ± 0.15 μm2 in surface area, ranging from 0.03 to 0.92 μm2 (PreAZs), and from 0.02 to 1.75 μm2 (PSDs), respectively. The variation in PreAZ surface area was not significant (P > 0.05), in contrast to PSDs, where a significant difference in surface area was found (P < 0.05; P < 0.001) in 1 out of 5 patients. Some of the PreAZs and PSDs were comparably large (4–6-fold larger than the mean), thus a huge variability in shape and size was observed (see Table 1). Interestingly, the PreAZ and PSD overlapped nearly perfectly in size as indicated by the ratio and the correlation factors (1.12 ± 0.35; R2 = 0.61; Fig. 6D). Besides very large PreAZs and PSDs (Figs 2E, F, 3F and 4), also quite small ones covering only a fraction of the apposition zone (Figs 2A and 3B) were found. The majority of synaptic boutons (~51%) showed either a perforation in their PreAZs, PSDs, or both (Figs 2E, 4C1–C3 and 5C, D, F); the remainder was non-perforated (Figs 2A, C, F, 5A, B, E). Interestingly, the mean surface area of the PreAZs on spines, regardless of spine type, was almost 2-fold larger when compared with those found on dendritic shafts (0.24 ± 0.05 vs. 0.14 ± 0.15 μm2); although this difference is not significant (P = 0.52 using Mann–Whitney U-test). Strikingly, only a weak correlation between the mean surface area of synaptic boutons and that of PreAZs (R2 = 0.31, Fig. 6C) and between the PreAZs and the total pool of SVs (R2 = 0.32; Fig. 6G) was found suggesting that the PreAZ and PSD size may be independently regulated from the size of the synaptic boutons and that of the total pool of SVs. Figure 5 View largeDownload slide Comparison of the total pool of SVs at individual L5 synaptic boutons. (A–F), 3D-volume reconstructions of individual total pools of SVs (green dots) at PreAZs and PSDs (red) with perforations either in the PreAZ, PSD or both. Note the large differences in the total pool size. Large dense-core vesicles (magenta dots), intermingled with the pool of clear SVs, were frequently observed intermingled with the pool of clear SVs. Figure 5 View largeDownload slide Comparison of the total pool of SVs at individual L5 synaptic boutons. (A–F), 3D-volume reconstructions of individual total pools of SVs (green dots) at PreAZs and PSDs (red) with perforations either in the PreAZ, PSD or both. Note the large differences in the total pool size. Large dense-core vesicles (magenta dots), intermingled with the pool of clear SVs, were frequently observed intermingled with the pool of clear SVs. Figure 6 View largeDownload slide View largeDownload slide Correlations between various structural parameters of synaptic boutons. (A) The surface area vs. the volume of the synaptic boutons. (B) The volume of the synaptic boutons vs. the volume of mitochondria. (C) The surface area of synaptic boutons vs. the surface area of PreAZs. (D) The surface area of PreAZs vs. PSD surface areas. (E) The surface area of synaptic boutons vs. the total pool of SVs. (F) The volume of synaptic boutons vs. the total pool of SVs. (G) The surface area of the PreAZs vs. the total pool of SVs. (H) The total pool of SVs vs. the RRP at p10. (I) The total pool of SVs vs. the RRP at p20. *Data points were fitted by linear regression and the R2 is given for each correlation. Figure 6 View largeDownload slide View largeDownload slide Correlations between various structural parameters of synaptic boutons. (A) The surface area vs. the volume of the synaptic boutons. (B) The volume of the synaptic boutons vs. the volume of mitochondria. (C) The surface area of synaptic boutons vs. the surface area of PreAZs. (D) The surface area of PreAZs vs. PSD surface areas. (E) The surface area of synaptic boutons vs. the total pool of SVs. (F) The volume of synaptic boutons vs. the total pool of SVs. (G) The surface area of the PreAZs vs. the total pool of SVs. (H) The total pool of SVs vs. the RRP at p10. (I) The total pool of SVs vs. the RRP at p20. *Data points were fitted by linear regression and the R2 is given for each correlation. The width of the synaptic cleft was 17.24 ± 2.38 nm for the lateral, and 19.05 ± 2.93 nm for the central region, respectively. However, no clear difference, e.g. the typical wide broadening of the synaptic cleft, was observed at PreAZs and PSDs in our sample as indicated by the coefficient of variation for both the lateral and central edges, respectively (0.13 vs. 0.15). Organization of the Pool of SVs Another key structural parameters regulating synaptic efficacy, plasticity, and mode of release are the size and the organization of the respective pool of SVs (Saviane and Silver 2006; Watanabe et al. 2013; Schikorski 2014; reviewed by Schneggenburger et al. 2002; Rizzoli and Betz 2005; Chamberland and Tóth 2016). In our study, the total pool comprised 1518.51 ± 303.18 SVs (ranging from 226 to 8437) and occupied ~10% (0.05 μm3) of the total volume of a synaptic bouton. However, a huge variability in total pool size was found as indicated by the standard deviation and the range (see also Supplementary Fig. 3). Interestingly, a strong correlation between the total pool of SVs and the total surface area (R2 = 0.69; Fig. 6E) as well as volume of synaptic boutons (R2 = 0.78; Fig. 6F) was found suggesting that larger synaptic boutons contain larger pools of SVs. Three different types of vesicles were found: (1) Small clear SVs with a mean diameter of 36.69 ± 1.71 nm, (2) large clear SVs with a mean diameter of 69.74 ± 12.26 (Fig. 2A, D) and (3) large DCVs with an average diameter of 65.08 ± 4.03 nm. DCVs were seen to either fuse with the presynaptic membrane or found at extrasynaptic locations and throughout the synaptic bouton (Figs 2A and 5B, D–F). SVs in individual boutons were distributed throughout the entire terminal (Figs 1B, C, 2B–F, 3B–F, 4A). However, except for the docked vesicles (Figs 2C, F) primed to the PreAZ, it is impossible to morphologically define the 3 functionally identified pools of SVs, namely the RRP, RP, and resting pool (reviewed by Rizzoli and Betz 2005; Denker and Rizzoli 2010; Chamberland and Tóth 2016). Thus, we performed a distance analysis that determined the exact location of each SVs from the PreAZ. We defined a distance (perimeter p) within ≤10 nm and ≤20 nm for vesicles located in close proximity of the PreAZ that may constitute the RRP, from which SVs could be easily and fast recruited upon stimulation. The second group may represent the RP maintaining release on moderate (physiological) stimulation and was located in proximity to the PreAZ (60–200 nm). All vesicles further than ≥200 nm away likely belonged to the resting pool which functions as a depot of SVs from which release is only triggered by intense stimulation. Using the same perimeter criteria at human L5 synaptic boutons, the RRP/PreAZ was 5.42 ± 4.09 at p10 and increased by nearly 3-fold (15.21 ± 9.02) at p20. The RP/PreAZ was also comparably large with 181.86 ± 27.05 SVs at 60–200 nm and the resting pool contained on average 1264.05 ± 301.77 SVs. However, no correlation was found between the RRP at p10 (R2 = 0.002, Fig. 6H) and p20 (R2 = 0.02, Fig. 6I) with the total pool of SVs – RRP at p10 and p20, respectively. Although small to medium-sized in surface area and volume, synaptic boutons in L5 of TL neocortex have comparably large RRPs, RPs, and resting pools when compared with even much larger CNS terminals (see Discussion). Cluster Analysis of Synaptic Boutons The broad distribution of some structural synaptic parameters and their correlation could be related to the presence of several types of synaptic boutons within our sample. To select the main features (parameters) of the original data, we performed data zero-mean normalization (Dodge 2003) as the parameters have different units, in order to eliminate the effect of big differences. Then, we run PCA and projected the original data to the new coordinate system composed by PC1 and PC2 (Fig. 7A) and got an explained variance ratio histogram (Fig. 7B). The PreAZs, PSDs, and the vesicle pools were the structural parameters which contribute most to principal components and best separated the synaptic boutons (Fig. 7A, B; see also Supplementary Table 2). We used the maximum ratio between intra-cluster distance against distance between clusters as a factor to decide the number of stable clusters identified by both methods HCA and K-means cluster analysis on the structural synaptic parameters. Dendrograms and scatter plots were generated for each analysis (Fig. 7C–F). Figure 7 View largeDownload slide Cluster analysis of synaptic parameters of L5 of the human TL neocortex. (A) The projection of the original data in the new coordinate system based on the first and second PCs. In this plot, the orthonormal PC coefficients for each variable and the PC scores for each observation in the new coordinate system are shown. The PreAZs and PSDs area as well as the vesicle pools are the main contributions to the first PC, indicating that these parameters are the main features of the original data. Abbreviations: bmito.vol: bouton’s mitochondria volume; bmito. num: Bouton’s mitochondria number; bmito.pvol: bouton’s mitochondria volume percentage; bsc. num: bouton’s synaptic contacts number; bvol: bouton’s volume; barea: Bouton’s area; bves.vol: bouton’s vesicles volume; bves.num: bouton’s vesicles number; bves.pvol: bouton’s vesicles volume percentage; bves. p10–p500: bouton’s vesicles number at 10–500 nm perimeter from the PreAZ. (B) Histogram of the explained variance ratio of PCs. The number of PCs considered to explain the original dataset is 2 (PC1 and PC2), whose total explained variance ratio is ~70 %. (C, D) Dendrogram (C) and scatter plot (D) of the PreAZs and PSDs using HCA, showing 2 major clusters of synaptic boutons with respect to the surface area of their PreAZs and PSDs. The difference between both clusters is indicated by the Euclidean distance. (E) Dendrogram of the 3 functionally defined vesicle pools, identifying 3 clusters although cluster 1 (putative RRP) and cluster 2 (putative RP) partially overlap. Cluster 3 representing the resting pool is clearly separated. (F) Scatter plot of the 3 functionally defined vesicle pools using the k-means method. Each PC is a linear combination (Pearson 1901) of the original columns in a matrix of dataset, where the columns corresponded to the number of synaptic vesicles at certain perimeter (p10–p500) from the PreAZ. The rows corresponded to the synaptic boutons analyzed. The PCs are given by an eigenvalue decomposition of the correlation matrix (Pearson 1901). Figure 7 View largeDownload slide Cluster analysis of synaptic parameters of L5 of the human TL neocortex. (A) The projection of the original data in the new coordinate system based on the first and second PCs. In this plot, the orthonormal PC coefficients for each variable and the PC scores for each observation in the new coordinate system are shown. The PreAZs and PSDs area as well as the vesicle pools are the main contributions to the first PC, indicating that these parameters are the main features of the original data. Abbreviations: bmito.vol: bouton’s mitochondria volume; bmito. num: Bouton’s mitochondria number; bmito.pvol: bouton’s mitochondria volume percentage; bsc. num: bouton’s synaptic contacts number; bvol: bouton’s volume; barea: Bouton’s area; bves.vol: bouton’s vesicles volume; bves.num: bouton’s vesicles number; bves.pvol: bouton’s vesicles volume percentage; bves. p10–p500: bouton’s vesicles number at 10–500 nm perimeter from the PreAZ. (B) Histogram of the explained variance ratio of PCs. The number of PCs considered to explain the original dataset is 2 (PC1 and PC2), whose total explained variance ratio is ~70 %. (C, D) Dendrogram (C) and scatter plot (D) of the PreAZs and PSDs using HCA, showing 2 major clusters of synaptic boutons with respect to the surface area of their PreAZs and PSDs. The difference between both clusters is indicated by the Euclidean distance. (E) Dendrogram of the 3 functionally defined vesicle pools, identifying 3 clusters although cluster 1 (putative RRP) and cluster 2 (putative RP) partially overlap. Cluster 3 representing the resting pool is clearly separated. (F) Scatter plot of the 3 functionally defined vesicle pools using the k-means method. Each PC is a linear combination (Pearson 1901) of the original columns in a matrix of dataset, where the columns corresponded to the number of synaptic vesicles at certain perimeter (p10–p500) from the PreAZ. The rows corresponded to the synaptic boutons analyzed. The PCs are given by an eigenvalue decomposition of the correlation matrix (Pearson 1901). First, hierarchical clustering on the PreAZs and PSDs (Fig. 7C) revealed 2 groups of synaptic boutons that remained stable after shuffling and sorting the dataset different random orders. The 2 groups were composed of large synaptic boutons (mean surface area ~8 μm2) with large PreAZ (~0.27 μm2) and PSD (~0.30 μm2); red cluster in Fig. 7C, D, and smaller ones belonging to the larger cluster (mean surface area ~6 μm2) with smaller PreAZ (~0.21 μm2) and PSD (~0.24 μm2) blue cluster in Fig. 7C, D. Interestingly, the PreAZ and PSD surface areas were equally important features as shown by the linear distribution of the points, which allowed clustering the dataset into 2 clusters as shown in Fig. 7D. Worth to mention that, there is a strong correlation between the PreAZ and PSD surface areas (R2 = 0.61, see Fig. 6D). However, there was no correlation between the surface area of the PreAZs and PSDs and that of the synaptic boutons in both groups (group 1: R2~0.10 and 0.24 for PSDs and PreAZs; group 2: R2~0.10 and 0.34 for PSDs and PreAZs, respectively). The other structural parameter for clustering the synaptic boutons was the vesicle pools, leading to 3 distinct clusters as shown in the dendrogram in Fig. 7E. Cluster 1 (putative RRP, red cluster in Fig. 7F) and cluster 2 (putative RP, green cluster in Fig. 7F) partially overlapped, while cluster 3 (resting pool, blue cluster in Fig. 7F), at distances greater than 200 nm from the PreAZ, is clearly separated thus. In contrast, the number of vesicles closer to the PreAZ (<20 nm) was identical and not significantly different between the clusters. Thus, combining the 2 strongest principal components of the analysis, the clustering algorithms used revealed 2 subtypes of L5 synaptic boutons in human TL neocortex, according to the size of their PreAZs and pools of SVs. Glial Coverage of L5 Synaptic Boutons EM examination revealed that astrocytes and their fine processes formed a dense network in L5 (Fig. 8). The majority of L5 synaptic boutons and their target structures (~80%) were tightly ensheathed by fine astrocytic processes (Fig. 8). Individual astrocytic processes, even of higher order, were seen to tightly wrap around multiple synaptic complexes, isolating them from the surrounding neuropil and from neighboring synaptic complexes. Astrocytic fingers reached as far as the synaptic cleft under both the PreAZs and PSDs (Fig. 8) suggesting that astrocytes are involved in the regulation of the spatial and temporal glutamate concentration profile at these synapses. Figure 8 View largeDownload slide Astrocytic coverage of synaptic complexes in L5 of the human TL neocortex. (A) Astrocytic processes (dark DAB reaction product) ensheating several synaptic complexes (synaptic boutons highlighted in transparent yellow, target structures in transparent blue, PreAZs and PSDs are marked by arrowheads). Scale bar 1 μm. (B) Representative example of 2 synaptic complexes (sp1–b1, sp2–b2) tightly ensheathed by fine astrocytic processes (highlighted in green) that could be followed as far as the synaptic cleft. (transparent red and arrowheads). Scale bar 0.5 μm. (C) 3D-volume reconstruction showing the tight ensheathment of astrocytes (green) around a synaptic complex (bouton in transparent yellow, target dendrite in blue). Note that fine astrocytic processes extend as far as to PreAZ and PSD (red). Figure 8 View largeDownload slide Astrocytic coverage of synaptic complexes in L5 of the human TL neocortex. (A) Astrocytic processes (dark DAB reaction product) ensheating several synaptic complexes (synaptic boutons highlighted in transparent yellow, target structures in transparent blue, PreAZs and PSDs are marked by arrowheads). Scale bar 1 μm. (B) Representative example of 2 synaptic complexes (sp1–b1, sp2–b2) tightly ensheathed by fine astrocytic processes (highlighted in green) that could be followed as far as the synaptic cleft. (transparent red and arrowheads). Scale bar 0.5 μm. (C) 3D-volume reconstruction showing the tight ensheathment of astrocytes (green) around a synaptic complex (bouton in transparent yellow, target dendrite in blue). Note that fine astrocytic processes extend as far as to PreAZ and PSD (red). Discussion The present study is the first comprehensive and coherent analysis of synaptic boutons in L5 of the human TL neocortex. Although synaptic boutons are built of nearly the same structural subelements, it is their individual and specific composition that makes them unique entities, perfectly adapted to their function in the microcircuit, in which they are embedded. In line with findings in cortical synapses in rodents and hippocampal CA1 synapses, the majority of human L5 excitatory synaptic boutons was found on spines, these findings are in line with that reported in NHPs (Bopp et al. 2017; Hsu et al. 2017). Approximately 65% of all spines contained a spine apparatus. The most striking difference is the size of the PreAZs and PSDs and that of the 3 pools of SVs, which are nearly 3-fold larger occupying ~10% of the total volume than that reported in different CNS synapses in rodents (Sätzler et al. 2002; Rollenhagen et al. 2007, 2015) and NHPs (~5% of the total volume; Bopp et al. 2017). Synaptic Organization in L5 Here, we describe the excitatory synaptic organization of L5 in the temporo-lateral and temporo-basal human TL neocortex. It has to be noted that L5 is heterogeneous with respect to its dendritic (see Results) and synaptic organization best described for rodent visual and somatosensory cortex (reviewed by Ramaswamy and Markram 2015) and humans (Mohan et al. 2015). However, as revealed by paired recordings combined with intracellular biocytin-fillings and/or EM analysis of morphologically identified neurons in rodents (Markram, Lübke, Frotscher, Roth, et al. 1997; Markram, Lübke, Frotscher, Sakmann, et al. 1997) the majority (~80%) of dendrites in L5 arise from L5 pyramidal neurons (for humans see Mohan et al. 2015), with only a smaller fraction from L6 pyramidal neurons and the remainder from GABAergic interneurons which were excluded in our study. The majority of excitatory synaptic boutons in rodent and human L5 are established by axonal collaterals of L5 pyramidal neurons (Markram et al. 1997a, b; Mohan et al. 2015; reviewed by Ramaswamy and Markram 2015) that form a dense vertical and horizontal axonal plexus in L5. It has been estimated that a single L5 pyramidal neuron could be potentially innervated by as many as 50 neighboring L5 pyramidal neurons (Song et al. 2005). Another strong synaptic input comes from ascending axons from L6A pyramidal neurons traversing or terminating in L5 and L4 (Zhang and Deschênes 1997). Other known intracortical inputs are descending axonal collaterals from L2/3 pyramidal neurons (Bruno et al. 2009) and L4 spiny stellate neurons, but terminate mainly on apical and apical oblique dendrites of L5A pyramidal neurons (Feldmeyer et al. 2005). Finally, L5 neurons receive various extracortical inputs of different origins (thalamus: Constantinople and Bruno 2013; Rodriguez-Moreno et al. 2017; amygdala: Freese and Amaral 2005, 2006) but due to their comparably low density in L5, the impact of these inputs versus intracortical input is still controversially discussed. Different GABAergic interneurons also innervate L5 pyramidal neurons by establishing synaptic contacts preferentially onto somata, proximal dendrites, axon initial segments, distal and terminal tuft dendrites. However, the input and contribution of GABAergic interneurons to the synaptic organization of the human TL neocortex was not part of our study. Important Structural Subelements of Synaptic Complexes in L5 of the Human TL Neocortex One important subelement in nerve terminals is mitochondria. Nearly all synaptic boutons contained mitochondria, often organized in clusters. Mitochondria were always closely associated with the pool of SVs, in line with observations in several other CNS synapses in various animal species (for example see Rollenhagen et al. 2007, 2015; Smith et al. 2016). In addition, mitochondria are reported to be highly mobile (Mironov 2006; Mironov and Symonchuk 2006) and are involved in the mobilization of SVs from the resting pool (Verstreken et al. 2005; Perkins et al. 2010; Smith et al. 2016) suggesting the ability of these synapses to prevent a rapid depletion of the RRP and RP by fast refilling from the resting pool. Furthermore, mitochondria act not only as energy suppliers, but also as internal calcium stores (Pozzan et al. 2000; Rizzuto et al. 2000) and are thought to regulate internal Ca2+ levels in nerve terminals (Perkins et al. 2010). Secondly, most synaptic boutons investigated were found on spines of different types, including filopodial, thin, stubby, and mushroom spines. Interestingly, the majority of all spines (~65%) featured a spine apparatus, a highly specialized structure involved in spine mobility. The high degree in the presence of a spine apparatus in human TL neocortex is in good agreement with findings in L4 and L5 synaptic boutons of the rat ‘barrel’ cortex (Rollenhagen et al. 2015; Rollenhagen, unpublished observations), but different to hippocampal CA1 synapses where only a smaller fraction (~20%) of spines contained a spine apparatus (Martone et al. 1997; Spacek and Harris 1997; Deller et al. 2003). It has been demonstrated that the abundance of a spine apparatus is directly linked to modulate short- and long-term plasticity (Holtmaat et al. 2005; Umeda et al. 2005). Altogether, the geometry of human L5 synaptic boutons, the organization and contribution of mitochondria, together with a preference for establishing synapses onto spines containing a spine apparatus may partially contribute to a high synaptic reliability, synaptic strength but also plasticity. These factors may convey the ability of these synapses to maintain high release rates during repetitive stimulation (see also below). Shape and Size of PreAZs and PSDs Two of the most important structural parameters determining pr, synaptic strength, efficacy, and plasticity are the shape and size of the PreAZs and PSDs (Matz et al. 2010; Holderith et al. 2012; Südhof 2012). The majority of L5 synaptic boutons had only one at most 2 PreAZs and PSDs in line with recent findings for other cortical synaptic boutons of similar size in rodents (Marrone et al. 2005; Nava et al. 2014; Rollenhagen et al. 2015; Hsu et al. 2017) and NHPs (Bopp et al. 2017; Hsu et al. 2017;). L5 PreAZs and PSDs in humans were comparable with those in neocortical L4 and L5 synaptic boutons in rats, but were ~2–3-fold larger than comparably sized CA1 synapses (Harris and Sultan 1995, Schikorski and Stevens 2001, Marrone et al. 2005; Nava et al. 2014) or even much larger CNS terminals such as the Calyx of Held (Spirou et al. 1998; Sätzler et al. 2002; Wimmer et al. 2006) the cerebellar (Xu-Friedman and Regehr 2003) and hippocampal MFB (Rollenhagen et al. 2007) although a large variability in both shape and size was observed. This may partially contribute to differences in the mode of release (uni- or multivesicular; uni- or multiquantal release) and quantal size, the size of the RRP, and pr as shown for other CNS synapses (Matz et al. 2010; Freche et al. 2011; Holderith et al. 2012; reviewed by Xu-Friedman and Regehr 2004). In addition, ~65% the synaptic boutons showed perforations in their PreAZ and/or PSD, which is higher than reported for L4 synaptic boutons (~35%) and comparable with values in L5 (~60%) in rats. A strong correlation between PreAZ area, perforated PSDs, the number of docked and resting pool vesicles was reported (Nava et al. 2014). Interestingly, only a weak correlation between the PreAZ surface area with that of the bouton was found in human and cortical synaptic boutons in rat (Rollenhagen et al. 2015), suggesting that the geometry of the PreAZs and PSDs is an independent structural parameter and may be regulated in an activity-dependent manner (Matz et al. 2010; Holderith et al. 2012). Altogether, the comparably large size, nearly perfect overlap and high number of perforated PreAZs and PSDs at human synaptic boutons may suggest a high pr, synaptic strength, and efficacy, but also plasticity. Size of the RRP and RP of SVs Besides the size of the AZ, the pool of releasable SVs determines pr and thus synaptic efficacy, strength, and plasticity (Rosenmund and Stevens 1996; Schikorski and Stevens 2001; Rizzoli and Betz 2004; Schikorski 2014; Watanabe et al. 2014; reviewed by Rizzoli and Betz 2005). However, it is still rather unclear whether functionally heterogeneous SV pools are structurally identifiable, and thus supports diverse forms of synaptic transmission and short-term synaptic plasticity (reviewed by Rizzoli and Betz 2005; Chamberland and Tóth 2016). In a presynaptic bouton synaptic strength is determined by the RRP and the probability of release of vesicles from this pool (Schikorski and Stevens 2001; Silver et al. 2003; Schikorski 2014; Watanabe et al. 2014). These parameters are controlled at the PreAZ but vary substantially across CNS synapses (reviewed by Rizzoli and Betz 2005). Whether a slow and/or fast RRP exist at all CNS synapses, and how the size of the RP contributes to synaptic dynamics and how such specific control is achieved at individual synapses remains largely unknown. Based on reported criteria (Rizzoli and Betz 2005), we subdivided the total pool of SVs into pools located at various perimeters from the PreAZ suggesting a putative RRP (p10 nm, p20 nm), an RP (>p20–<200 nm), and a resting pool (>p200 nm; see Table 2). The total pool size of SVs at L5 synaptic boutons in human TL neocortex was ~1500/bouton/PreAZ and was thus nearly 3-fold (~550/PreAZ; Rollenhagen et al. 2015) and 2-fold larger (~750/PreAZ; Rollenhagen, unpublished observations) when compared with values for rat L4 and L5 synaptic boutons. Comparison with even much larger nerve terminals, for example, adult hippocampal mossy fiber boutons (MFB) (~20-fold larger in size), revealed a total pool size of SVs/PreAZ of ~850 (Rollenhagen et al. 2007), ~600/PreAZ at cerebellar mossy fiber synapses (Saviane and Silver 2006), but a nearly 12-fold larger total pool (~125 SVs/PreAZ) when compared with that at the Calyx of Held giant terminal (Sätzler et al. 2002). The much larger total pool size at human L5 synaptic boutons most likely also suggests a comparably large RRP and RP. The putative RRP size at human L5 synaptic boutons was on average 5.4 ± 4.1 (p10) and 15.2 ± 9.0 (p20) vesicles/PreAZ, comparable to those in rats (p10 3.9 ± 3.4, p20 11.56 ± 4.2; Rollenhagen, unpublished observations), but 2–3-fold larger than that in rat L4 synaptic boutons (p10 2.0 ± 2.6, p20 6.3 ± 6.4; Rollenhagen et al. 2015), hippocampal MFBs (p10 1.6 ± 1.5, p20 6.2 ± 4.1; Rollenhagen et al. 2007) and Calyx of Held (p10 1.9 ± 2.0, p20 4.8 ± 3.8; Sätzler et al. 2002). Recently, it has been further demonstrated that the number of docked vesicles at perforated synapses significantly exceeds that of non-perforated ones (Nava et al. 2014). Similarly, the putative RP/PreAZ was ~200 vesicles at human L5 synaptic boutons, ~130 vesicles for rat L4, ~200 vesicles for rat L5, ~3700 vesicles for adult MFBs in rat, but nearly 4-fold larger than at the rat Calyx of Held (~60 vesicles). DCVs, the second population of vesicles, were frequently found at human synaptic boutons, and are thought to have several functions at synapses: besides being involved in endo- and exocytosis (Watanabe et al. 2013), DCVs also contribute to the build-up and organization of PreAZs either by releasing Piccolo and Bassoon (Schoch and Gundelfinger 2006), or by clustering SVs at the presynaptic density (Mukherjee et al. 2010, Watanabe et al. 2013). In addition, they may contain various co-transmitters, such as neuropeptides, ATP, noradrenalin, and dynorphin (Ghijsen and Leenders 2005). In summary, L5 synaptic boutons in the human TL neocortex were characterized by a comparably large RRP, RP, and resting pool although with a great variability between individual synaptic boutons, as also demonstrated for other CNS synapses in rodents. The large RRP and RP at human L5 synaptic boutons may thus prevent depletion during repetitive high-frequency stimulation and even more importantly, the relatively large size of the RP and resting pool could be used to rapidly refill the releasable pool if subject to depletion by long-lasting repetitive stimulation. If the refilling rates were activity-dependent, the large size of the RP could explain some forms of short-term synaptic plasticity, e.g. a substantial increase in synaptic strength during frequency facilitation and post-tetanic potentiation at these synapses. In terms of the function of TL neocortex a high pr, synaptic strength and efficacy controlled by the structural composition of L5 synaptic boutons may lead to the synchronous firing of neuronal ensembles within the TL neocortex involved in the induction and regulation of various computations underlying perception, executive control, learning and memory in which the TL neocortex plays an important role. During sensory stimulation and execution of complex behaviors as well as up-states (Zhou and Fuster 1996; Sanchez-Vives and McCormick 2000; Sakata and Harris 2009) cortical L5 pyramidal neurons in particular fire multiple action potentials at high frequencies. During such prolonged and intense activity, synaptic transmission could be modulated in various ways depending on the availability of SVs and on their recycling rates. Thus, human L5 synapses in the TL neocortex may well function as detonator synapses sufficiently strong to fire the target neuron by themselves (Engel and Jonas 2005). On the other hand, such prolonged synchronous firing of excitatory neurons could onset epileptic seizures of excitatory neurons in the TL as demonstrated recently by loss of constitutive functional γ-aminobutyric acid type A-B receptor crosstalk in L5 pyramidal neurons of human epileptic temporal cortex (Martinello et al. 2018). Glial Coverage of L5 Synaptic Boutons in the Human TL Neocortex Astrocytes have long been thought to act as nutrition suppliers and stabilizing corset for neurons in the brain. However, it is now well-established that astrocytes also play an important role in synaptic function, acting not only as physical barriers to glutamate diffusion, but also mediate transmitter uptake by glutamate transporters (Oliet et al. 2004; Min and Nevian 2012; Pannasch et al. 2014; for review see Allen 2014). Moreover, it has been hypothesized that the degree of astrocytic coverage correlates with the activity of a synapse (Min and Nevian 2012, reviewed by Allen 2014). A striking feature in human L5 was the tight ensheathment of synaptic complexes and the establishment of direct synaptic contacts with astrocytic processes, in contrast to the hippocampal (Rollenhagen et al. 2007), cerebellar MFBs (Xu-Friedman and Regehr 2003) and calyx of Held synapses (Müller et al. 2009), where astrocytic processes were never located close to individual synaptic clefts under the PreAZs and PSDs. This may explain the occurrence of glutamate spillover, synaptic crosstalk and the switch from asynchronous to synchronous release upon repetitive stimulation as shown for the MFB (Hallermann et al. 2003) and calyx of Held synapses (reviewed by von Gersdorff and Borst 2002). In addition, synapses with a higher number of ‘docked’ vesicles were almost wrapped-up in astrocytic processes indicating that they may act as physical barriers to neurotransmitter diffusion. Astrocytes can actively take-up excessive or ‘spilled’ glutamate; hence, they regulate its concentration in the synaptic cleft, thereby shaping the unitary EPSP amplitude, and accelerate the recovery from receptor desensitization (Danbolt 2001; Oliet et al. 2004). In addition, neuronal activity can trigger calcium signals (waves) in astrocytes, and in turn, such calcium signals can elicit responses in neurons (Verkhratsky and Kettenmann 1996). Astrocytes can also release glutamate and GABA, which can control and modulate transmitter release (Le Meur et al. 2012). The close structural relationship between synaptic complexes and fine astrocytic processes strongly suggests a role of astrocytic processes at L5 synaptic complexes in the human TL neocortex by regulating the concentration and direction of diffusion of glutamate in the synaptic cleft. Hence, they control the induction, maintenance, and termination of synaptic transmission but also modulate short-term synaptic plasticity (Min and Nevian 2012) at these synapses. Funding The funding of Rachida Yakoubi by the DAAD is very much acknowledged. Notes Some of the work has been done during the Master thesis of Rachida Yakoubi under the very well appreciated supervison of Dr Rachid Mosbah (University of M´Hamed Bougara, Boumerdes, Algeria). We would like to thank our technicians Brigitte Marshallsay and Ulrike Holz for their excellent technical assistance. Many thanks to Prof. Christian Stricker (Australian National University, Canberra, Australia) and Dr Alexander Peyser (Institute for Advanced Simulation, Jülich Supercomputing Centre and Simulation Lab Neuroscience) for their critical reading and constructive comments on a prefinal version of the manuscript. Special thanks to Sandra Diaz and Dr Alexander Peyser (Institute for Advanced Simulation, Jülich Supercomputing Centre and Simulation Lab Neuroscience) for the constant support and help in running the cluster analysis. Conflict of Interest: None declared. References Abdi H , Williams LJ . 2010 . Principal component analysis . Wiley Interdiscip Rev Comput Mol Sci . 2 : 433 – 459 . Google Scholar Crossref Search ADS WorldCat Allen NJ . 2014 . Astrocyte regulation of synaptic behavior . Annu Rev Cell Dev Biol . 30 : 439 – 463 . Google Scholar Crossref Search ADS PubMed WorldCat Alonso-Nanclares L , Gonzalez-Soriano J , Rodriguez JR , DeFelipe J . 2008 . Gender differences in human cortical synaptic density . Proc Natl Acad Sci USA . 105 : 14615 – 14619 . Google Scholar Crossref Search ADS PubMed WorldCat Alonso-Nanclares L , Kastanauskaite A , Rodriguez JR , Gonzalez-Soriano J , Defelipe J . 2011 . A stereological study of synapse number in the epileptic human hippocampus . Front Neuroanat . 5 : 8 . Google Scholar Crossref Search ADS PubMed WorldCat Bernhardt BC , Hong S , Bernasconi A , Bernasconi N . 2013 . Imaging structural and functional brain networks in temporal lobe epilepsy . Front Hum Neurosci . 7 : 624 . Google Scholar Crossref Search ADS PubMed WorldCat Blazquez-Llorca L , Merchán-Pérez A , Rodriguez JR , Gascón J , DeFelipe J . 2013 . FIB/SEM technology and Alzheimer’s disease: three-dimensional analysis of human cortical synapses . J Alzheimer’s Dis . 34 : 995 – 1013 . Google Scholar Crossref Search ADS WorldCat Bopp R , Holler-Rickauer S , Martin KA , Schuhknecht GFP . 2017 . An ultrastructural study of the thalamic input to layer 4 of primary motor and primary somatosensory cortex in the mouse . J Neurosci . 37 : 2435 – 2448 . Google Scholar Crossref Search ADS PubMed WorldCat Bruno RM , Hahn TT , Wallace DJ , de Kock CP , Sakmann B . 2009 . Sensory experience alters specific branches of individual corticocortical axons during development . J Neurosci . 29 : 3172 – 3181 . Google Scholar Crossref Search ADS PubMed WorldCat Chamberland S , Tóth K . 2016 . Functionally heterogeneous synaptic vesicle pools support diverse synaptic signalling . J Physiol . 594 : 825 – 835 . Google Scholar Crossref Search ADS PubMed WorldCat Constantinople CM , Bruno RM . 2013 . Deep cortical layers are activated directly by thalamus . Science . 340 : 1591 – 1594 . Google Scholar Crossref Search ADS PubMed WorldCat Cragg BG . 1976 . Ultrastructural features of human cerebral cortex . J Anat . 121 : 331 – 362 . Google Scholar PubMed WorldCat Danbolt NC . 2001 . Glutamate uptake . Prog Neurobiol . 65 : 1 – 105 . Google Scholar Crossref Search ADS PubMed WorldCat DeFelipe J . 2011 . The evolution of the brain, the human nature of cortical circuits, and intellectual creativity . Front Neuroanat . 5 : 29 . Google Scholar PubMed WorldCat DeFelipe J , Alonso-Nanclares L , Arellano JI . 2002 . Microstructure of the neocortex: comparative aspects . J Neurocytol . 31 : 299 – 316 . Google Scholar Crossref Search ADS PubMed WorldCat Deller T, Korte M, Chabanis S, Drakew A, Schwegler H, Stefani GG, Zuniga A, Schwarz K, Bonhoeffer T, Zeller R, Frotscher M, Mundel P. 2003. Synaptopodin-deficient mice lack a spine apparatus and show deficits in synaptic plasticity. Proc Natl Acad Sci USA. 100:10494–10499. Denker A , Rizzoli SO . 2010 . Synaptic vesicle pools: an update . Front Synaptic Neurosci . 2 : 135 . Google Scholar PubMed WorldCat Dodge Y 2003 . The Oxford dictionary of statistical terms, OUP. ISBN 0-19-920613-9 entry for normalization of scores. Dufour A, Rollenhagen A, Sätzler K, Lübke JH. 2016. Development of synaptic Boutons in Layer 4 of the Barrel Field of the Rat Somatosensory Cortex: A Quantitative Analysis. Cereb Cortex. 26:838–854. Elston GN , Benavides-Piccione R , DeFelipe J . 2001 . The pyramidal cell in cognition: a comparative study in human and monkey . J Neurosci . 21 : RC163 . Google Scholar Crossref Search ADS PubMed WorldCat Elston GN , Benavides-Piccione R , Elston A , Manger PR , DeFelipe J . 2011 . Pyramidal cells in prefrontal cortex of primates: marked differences in neuronal structure among species . Front Neuroanat . 5 : 2 . Google Scholar Crossref Search ADS PubMed WorldCat Elston GN , Manger PR . 2014 . Pyramidal cells inv1 of African rodents are bigger, more branched and more spiny than those in primates . Front Neuroanat . 8 : 4 . Google Scholar PubMed WorldCat Engel D , Jonas P . 2005 . Presynaptic action potential amplification by voltage-gated Na+ channels in hippocampal mossy fiber boutons . Neuron . 45 : 405 – 417 . Google Scholar Crossref Search ADS PubMed WorldCat Feldmeyer D , Roth A , Sakmann B . 2005 . Monosynaptic connections between pairs of spiny stellate cells in layer 4 and pyramidal cells in layer 5A indicate that lemniscal and paralemniscal afferent pathways converge in the infragranular somatosensory cortex . J Neurosci . 25 : 3423 – 3431 . Google Scholar Crossref Search ADS PubMed WorldCat Freche D , Pannasch U , Rouach N , Holcman D . 2011 . Synapse geometry and receptor dynamics modulate synaptic strength . PLoS One . 6 : e25122 . Google Scholar Crossref Search ADS PubMed WorldCat Freese JL , Amaral DG . 2005 . The organization of projections from the amygdala to visual cortical areas TE and V1 in the macaque monkey . J Comp Neurol . 486 : 295 – 317 . Google Scholar Crossref Search ADS PubMed WorldCat Freese JL, Amaral DG. 2006. Synaptic organization of projections from the amygdala to visual cortical areas TE and V1 in the macaque monkey. J Comp Neurol. 496:655–667. Ghijsen WE , Leenders AG . 2005 . Differential signaling in presynaptic neurotransmitter release . Cell Mol Life Sci . 62 : 937 – 954 . Google Scholar Crossref Search ADS PubMed WorldCat Gibson PH . 1983 . EM study of the numbers of cortical synapses in the brains of ageing people and people with Alzheimer-type dementia . Acta Neuropathol . 62 : 127 – 133 . Google Scholar Crossref Search ADS PubMed WorldCat Hallermann S , Pawlu C , Jonas P , Heckmann M . 2003 . A large pool of releasable vesicles in a cortical glutamatergic synapse . Proc Natl Acad Sci USA . 100 : 8975 – 8980 . Google Scholar Crossref Search ADS PubMed WorldCat Harris KM , Sultan P . 1995 . Variation in the number, location and size of synaptic vesicles provides an anatomical basis for the nonuniform probability of release at hippocampal CA1 synapses . Neuropharmacology . 34 : 1387 – 1395 . Google Scholar Crossref Search ADS PubMed WorldCat Hartigan JA , Wong MA . 1979 . Algorithm AS 136: a k-means clustering algorithm . J R Stat Soc Ser C Appl Stat . 28 : 100 – 108 . WorldCat Holderith N , Lorincz A , Katona G , Rozsa B , Kulik A , Watanabe M , Nusser Z . 2012 . Release probability of hippocampal glutamatergic terminals scales with the size of the active zone . Nat Neurosci . 15 : 988 – 997 . Google Scholar Crossref Search ADS PubMed WorldCat Holtmaat AJ , Trachtenberg JT , Wilbrecht L , Shepherd GM , Zhang X , Knott GW , Svoboda K . 2005 . Transient and persistent dendritic spines in the neocortex in vivo . Neuron . 45 : 279 – 291 . Google Scholar Crossref Search ADS PubMed WorldCat Hotelling H . 1933 . Analysis of a complex of statistical variables into principal components . J Educ Psychol . 24 ( 6 ): 417 – 441 . Google Scholar Crossref Search ADS WorldCat Hsu A , Luebke JI , Medalla M . 2017 . Comparative ultrastructural features of excitatory synapses in the visual and frontal cortices of the adult mouse and monkey . J Comp Neurol . 525 : 2175 – 2191 . Google Scholar Crossref Search ADS PubMed WorldCat Kay KR , Smith C , Wright AK , Serrano-Pozo A , Pooler AM , Koffie R , Bastin ME , Bak TH , Abrahams S , Kopeikina KJ , McGuone D , Frosch MP , Gillingwater TH , Hyman BT , Spires-Jones TL . 2013 . Studying synapses in human brain with array tomography and electron microscopy . Nat Proto . 8 : 1366 – 1380 . Google Scholar Crossref Search ADS WorldCat Kiernan JA . 2012 . Anatomy of the temporal lobe . Epilepsy Res Treat . 2012 : 176157 . Google Scholar PubMed WorldCat Kirkpatrick B , Xu L , Cascella N , Ozeki Y , Sawa A , Roberts RC . 2006 . DISC1 immunoreactivity at the light and ultrastructural level in the human neocortex . J Comp Neurol . 497 : 436 – 450 . Google Scholar Crossref Search ADS PubMed WorldCat Le Meur K , Mendizabal-Zubiaga J , Grandes P , Audinat E . 2012 . GABA release by hippocampal astrocytes . Front Comput Neurosci . 6 : 59 . Google Scholar Crossref Search ADS PubMed WorldCat Liu XB , Schumann CM . 2014 . Optimization of electron microscopy for human brains with long-term fixation and fixed-frozen sections . Acta Neuropathol Commun . 2 : 42 . Google Scholar Crossref Search ADS PubMed WorldCat Markram H , Lübke J , Frotscher M , Roth A , Sakmann B . 1997 a. Physiology and anatomy of synaptic connections between thick tufted pyramidal neurones in the developing rat neocortex . J Physiol . 500 : 409 – 440 . Google Scholar Crossref Search ADS PubMed WorldCat Markram H , Lübke J , Frotscher M , Sakmann B . 1997 b. Regulation of synaptic efficacy by coincidence of postsynaptic APs and EPSPs . Science . 275 : 213 – 215 . Google Scholar Crossref Search ADS PubMed WorldCat Marrone DF , LeBoutillier JC , Petit TL . 2005 . Modeling behavioral recovery following lesion induction in the rat dentate gyrus . Neurobiol Learn Mem . 83 : 196 – 205 . Google Scholar Crossref Search ADS PubMed WorldCat Martinello K , Sciaccaluga M , Morace R , Mascia A , Arcella A , Esposito V , Fucile S . 2018 . Loss of constitutive functional γ-aminobutyric acid type A-B receptor crosstalk in layer 5 pyramidal neurons of human epileptic temporal cortex . Epilepsia . 59 : 449 – 459 . Google Scholar Crossref Search ADS PubMed WorldCat Martone ME, Zhang Y, Simpliciano VM, Carragher BO, Ellisman MH. 1993. Three-dimensional visualization of the smooth endoplasmic reticulum in Purkinje cell dendrites. J Neurosci. 13:4636–4646. Matz J , Gilyan A , Kolar A , McCarvill T , Krueger SR . 2010 . Rapid structural alterations of the active zone lead to sustained changes in neurotransmitter release . Proc Natl Acad Sci USA . 107 : 8836 – 8841 . Google Scholar Crossref Search ADS PubMed WorldCat Min R , Nevian T . 2012 . Astrocyte signaling controls spike timing-dependent depression at neocortical synapses . Nat Neurosci . 15 : 746 – 753 . Google Scholar Crossref Search ADS PubMed WorldCat Mironov SL . 2006 . Spontaneous and evoked neuronal activities regulate movements of single neuronal mitochondria . Synapse . 59 : 403 – 411 . Google Scholar Crossref Search ADS PubMed WorldCat Mironov SL , Symonchuk N . 2006 . ER vesicles and mitochondria move and communicate at synapses . J Cell Sci . 119 : 4926 – 4934 . Google Scholar Crossref Search ADS PubMed WorldCat Mohan H , Verhoog MB , Doreswamy KK , Eyal G , Aardse R , Lodder BN , Goriounova NA , Asamoah B , B Brakspear AB , Groot C , van der Sluis S , Testa-Silva G , Obermayer J , Boudewijns ZS , Narayanan RT , Baayen JC , Segev I , Mansvelder HD , de Kock CP . 2015 . Dendritic and axonal architecture of individual pyramidal neurons across layers of adult human neocortex . Cereb Cortex . 25 : 4839 – 4853 . Google Scholar Crossref Search ADS PubMed WorldCat Molnár G , Rozsa M , Baka J , Holderith N , Barzo P , Nusser Z , Tamas G . 2016 . Human pyramidal to interneuron synapses are mediated by multi-vesicular release and multiple docked vesicles . eLife . 5 . doi: 10.7554/eLife.18167. WorldCat Mukherjee K , Yang X , Gerber SH , Kwon HB , Ho A , Castillo PE , Liu X , Südhof TC . 2010 . Piccolo and bassoon maintain synaptic vesicle clustering without directly participating in vesicle exocytosis . Proc Natl Acad Sci USA . 107 : 6504 – 6509 . Google Scholar Crossref Search ADS PubMed WorldCat Müller J , Reyes-Haro D , Pivneva T , Nolte C , Schaette R , Lübke J , Kettenmann H . 2009 . The principal neurons of the medial nucleus of the trapezoid body and NG2(+) glial cells receive coordinated excitatory synaptic input . J Gen Physiol . 134 : 115 – 127 . Google Scholar Crossref Search ADS PubMed WorldCat Nava N , Chen F , Wegener G , Popoli M , Nyengaard JR . 2014 . A new efficient method for synaptic vesicle quantification reveals differences between medial prefrontal cortex perforated and nonperforated synapses . J Comp Neurol . 522 : 284 – 297 . Google Scholar Crossref Search ADS PubMed WorldCat Navarrete M , Perea G , Maglio L , Pastor J , Garcia de Sola R , Araque A . 2013 . Astrocyte calcium signal and gliotransmission in human brain tissue . Cereb Cortex . 23 : 1240 – 1246 . Google Scholar Crossref Search ADS PubMed WorldCat Oliet SH , Piet R , Poulain DA , Theodosis DT . 2004 . Glial modulation of synaptic transmission: insights from the supraoptic nucleus of the hypothalamus . Glia . 47 : 258 – 267 . Google Scholar Crossref Search ADS PubMed WorldCat Pannasch U , Freche D , Dallerac G , Ghezali G , Escartin C , Ezan P , Cohen-Salmon M , Benchenane K , Abudara V , Dufour A , Lübke JHR , Deglon N , Knott G , Holcman D , Rouach N . 2014 . Connexin 30 sets synaptic strength by controlling astroglial synapse invasion . Nature Neurosci . 17 : 549 – 558 . Google Scholar Crossref Search ADS PubMed WorldCat Papouin T , Dunphy J , Tolman M , Foley JC , Haydon PG . 2017 . Astrocytic control of synaptic function . Philos Trans R Soc Lond B Biol Sci . 372 . doi:10.1098/rstb.2016.0154 . WorldCat Pearson K . 1901 . LIII. On lines and planes of closest fit to systems of points in space . Lond Edinb Dubl Phil Mag . 2 : 559 – 572 . Google Scholar Crossref Search ADS WorldCat Pedregosa F , Varoquaux G , Gramfort A , Michel V , Thirion B , Grisel O , Blondel M , Prettenhofer P , Weiss R , Vanderplas J . 2011 . Scikit-learn: machine learning in Python . J Mach Learn Res . 12 : 2825 – 2830 . WorldCat Perkins GA , Tjong J , Brown JM , Poquiz PH , Scott RT , Kolson DR , Ellisman MH , Spirou GA . 2010 . The micro-architecture of mitochondria at active zones: electron tomography reveals novel anchoring scaffolds and cristae structured for high-rate metabolism . J Neurosci . 30 : 1015 – 1026 . Google Scholar Crossref Search ADS PubMed WorldCat Popov VI , Stewart MG . 2009 . Complexity of contacts between synaptic boutons and dendritic spines in adult rat hippocampus: three-dimensional reconstructions from serial ultrathin sections in vivo . Synapse . 63 : 369 – 377 . Google Scholar Crossref Search ADS PubMed WorldCat Pozzan T , Magalhaes P , Rizzuto R . 2000 . The comeback of mitochondria to calcium signalling . Cell Calcium . 28 : 279 – 283 . Google Scholar Crossref Search ADS PubMed WorldCat Ramaswamy S , Markram H . 2015 . Anatomy and physiology of the thick-tufted layer 5 pyramidal neuron . Front Cell Neurosci . 9 : 233 . Google Scholar PubMed WorldCat Reynolds ES . 1963 . The use of lead citrate at high pH as an electron-opaque stain in electron microscopy . J Cell Biol . 17 : 208 – 212 . Google Scholar Crossref Search ADS PubMed WorldCat Rizzoli SO , Betz WJ . 2004 . The structural organization of the readily releasable pool of synaptic vesicles . Science . 303 : 2037 – 2039 . Google Scholar Crossref Search ADS PubMed WorldCat Rizzoli SO , Betz WJ . 2005 . Synaptic vesicle pools . Nat Rev Neurosci . 6 : 57 – 69 . Google Scholar Crossref Search ADS PubMed WorldCat Rizzuto R , Bernardi P , Pozzan T . 2000 . Mitochondria as all-round players of the calcium game . J Physiol . 529 ( Pt 1 ): 37 – 47 . Google Scholar Crossref Search ADS PubMed WorldCat Rodriguez-Moreno J , Rollenhagen A , Arlandis J , Santuy A , Merchan-Perez A , DeFelipe J , Lübke JHR , Clasca F . 2017 . Quantitative 3D ultrastructure of thalamocortical synapses from the “Lemniscal” ventral posteromedial nucleus in mouse barrel cortex . Cereb Cortex . 1 – 17 . doi: 10.1093/cercor/bhx187 WorldCat Rokach L , Maimon O . 2005 . Clustering methods /Data mining and knowledge discovery handbook . Boston : Springer, Boston, MA . p. 321 – 352 . Google Preview WorldCat Roland PE , Zilles K . 1996 . Functions and structures of the motor cortices in humans . Curr Opin Neurobiol . 6 : 773 – 781 . Google Scholar Crossref Search ADS PubMed WorldCat Rollenhagen A , Klook K , Sätzler K , Qi G , Anstötz M , Feldmeyer D , Lübke JH . 2015 . Structural determinants underlying the high efficacy of synaptic transmission and plasticity at synaptic boutons in layer 4 of the adult rat barrel cortex . Brain Struct Funct . 220 : 3185 – 3209 . Google Scholar Crossref Search ADS PubMed WorldCat Rollenhagen A , Lübke JH . 2006 . The morphology of excitatory central synapses: from structure to function . Cell Tissue Res . 326 : 221 – 237 . Google Scholar Crossref Search ADS PubMed WorldCat Rollenhagen A , Sätzler K , Rodriguez EP , Jonas P , Frotscher M , Lübke JH . 2007 . Structural determinants of transmission at large hippocampal mossy fiber synapses . J Neurosci . 27 : 10434 – 10444 . Google Scholar Crossref Search ADS PubMed WorldCat Rosenmund C , Stevens CF . 1996 . Definition of the readily releasable pool of vesicles at hippocampal synapses . Neuron . 16 : 1197 – 1207 . Google Scholar Crossref Search ADS PubMed WorldCat Sakata S , Harris KD . 2009 . Laminar structure of spontaneous and sensory-evoked population activity in auditory cortex . Neuron . 64 : 404 – 418 . Google Scholar Crossref Search ADS PubMed WorldCat Sanchez-Vives MV , McCormick DA . 2000 . Cellular and network mechanisms of rhythmic recurrent activity in neocortex . Nat Neurosci . 3 : 1027 – 1034 . Google Scholar Crossref Search ADS PubMed WorldCat Saviane C , Silver RA . 2006 . Fast vesicle reloading and a large pool sustain high bandwidth transmission at a central synapse . Nature . 439 : 983 – 987 . Google Scholar Crossref Search ADS PubMed WorldCat Schikorski T . 2014 . Readily releasable vesicles recycle at the active zone of hippocampal synapses . Proc Natl Acad Sci USA . 111 : 5415 – 5420 . Google Scholar Crossref Search ADS PubMed WorldCat Schikorski T , Stevens CF . 2001 . Morphological correlates of functionally defined synaptic vesicle populations . Nature Neurosci . 4 : 391 – 395 . Google Scholar Crossref Search ADS PubMed WorldCat Schneggenburger R , Sakaba T , Neher E . 2002 . Vesicle pools and short-term synaptic depression: lessons from a large synapse . Trends Neurosci . 25 : 206 – 212 . Google Scholar Crossref Search ADS PubMed WorldCat Schoch S , Gundelfinger ED . 2006 . Molecular organization of the presynaptic active zone . Cell Tissue Res . 326 : 379 – 391 . Google Scholar Crossref Search ADS PubMed WorldCat Silver H , Feldman P , Bilker W , Gur RC . 2003 . Working memory deficit as a core neuropsychological dysfunction in schizophrenia . Am J Psychiatry . 160 : 1809 – 1816 . Google Scholar Crossref Search ADS PubMed WorldCat Smith HL , Bourne JN , Cao G , Chirillo MA , Ostroff LE , Watson DJ , Harris KM . 2016 . Mitochondrial support of persistent presynaptic vesicle mobilization with age-dependent synaptic growth after LTP . eLife . 5 . doi: 10.7554/eLife.15275. WorldCat Song S , Sjöstrom PJ , Reigl M , Nelson S , Chklovskii DB . 2005 . Highly nonrandom features of synaptic connectivity in local cortical circuits . PLoS Biol . 3 : e68 . Google Scholar Crossref Search ADS PubMed WorldCat Spacek J, Harris KM. 1997. Three-dimensional organization of smooth endoplasmic reticulum in hippocampal CA1 dendrites and dendritic spines of the immature and mature rat. J Neurosci. 17:190–203. Spirou GA , Rowland KC , Berrebi AS . 1998 . Ultrastructure of neurons and large synaptic terminals in the lateral nucleus of the trapezoid body of the cat . J Comp Neurol . 398 : 257 – 272 . Google Scholar Crossref Search ADS PubMed WorldCat Südhof TC . 2012 . The presynaptic active zone . Neuron . 75 : 11 – 25 . Google Scholar Crossref Search ADS PubMed WorldCat Sätzler K , Söhl LF , Bollmann JH , Borst JG , Frotscher M , Sakmann B , Lübke JHR . 2002 . Three-dimensional reconstruction of a calyx of Held and its postsynaptic principal neuron in the medial nucleus of the trapezoid body . J Neurosci . 22 : 10567 – 10579 . Google Scholar Crossref Search ADS PubMed WorldCat Testa-Silva G , Verhoog MB , Goriounova NA , Loebel A , Hjorth J , Baayen JC , de Kock CP , Mansvelder HD . 2010 . Human synapses show a wide temporal window for spike-timing-dependent plasticity . Front Synaptic Neurosci . 2 : 12 . Google Scholar PubMed WorldCat Testa-Silva G , Verhoog MB , Linaro D , de Kock CP , Baayen JC , Meredith RM , De Zeeuw CI , Giugliano M , Mansvelder HD . 2014 . High bandwidth synaptic communication and frequency tracking in human neocortex . PLoS Biol . 12 : e1002007 . Google Scholar Crossref Search ADS PubMed WorldCat Umeda T , Ebihara T , Okabe S . 2005 . Simultaneous observation of stably associated presynaptic varicosities and postsynaptic spines: morphological alterations of CA3-CA1 synapses in hippocampal slice cultures . Mol Cell Neurosci . 28 : 264 – 274 . Google Scholar Crossref Search ADS PubMed WorldCat Verkhratsky A , Kettenmann H . 1996 . Calcium signalling in glial cells . Trends Neurosci . 19 : 346 – 352 . Google Scholar Crossref Search ADS PubMed WorldCat Verstreken P , Ly CV , Venken KJ , Koh TW , Zhou Y , Bellen HJ . 2005 . Synaptic mitochondria are critical for mobilization of resting pool vesicles at Drosophila neuromuscular junctions . Neuron . 47 : 365 – 378 . Google Scholar Crossref Search ADS PubMed WorldCat von Economo C , Koskinas GN . 1925 . Die Cytoarchitektonik der Hirnrinde des erwachsenen Menschen . Wien und Berlin : J. Springer . Google Preview WorldCat von Gersdorff H , Borst JG . 2002 . Short-term plasticity at the calyx of Held . Nat Rev Neurosci . 3 : 53 – 64 . Google Scholar Crossref Search ADS PubMed WorldCat Watanabe S , Rost BR , Camacho-Perez M , Davis MW , Sohl-Kielczynski B , Rosenmund C , Jorgensen EM . 2013 . Ultrafast endocytosis at mouse hippocampal synapses . Nature . 504 : 242 – 247 . Google Scholar Crossref Search ADS PubMed WorldCat Watanabe S , Trimbuch T , Camacho-Perez M , Rost BR , Brokowski B , Sohl-Kielczynski B , Felies A , Davis MW , Rosenmund C , Jorgensen EM . 2014 . Clathrin regenerates synaptic vesicles from endosomes . Nature . 515 : 228 – 233 . Google Scholar Crossref Search ADS PubMed WorldCat Wilhelm BG , Mandad S , Truckenbrodt S , Krohnert K , Schafer C , Rammner B , Koo SJ , Classen GA , Krauss M , Haucke V , Urlaub H , Rizzoli SO . 2014 . Composition of isolated synaptic boutons reveals the amounts of vesicle trafficking proteins . Science . 344 : 1023 – 1028 . Google Scholar Crossref Search ADS PubMed WorldCat Wilke SA , Antonios JK , Bushong EA , Badkoobehi A , Malek E , Hwang M , Terada M , Ellisman MH , Ghosh A . 2013 . Deconstructing complexity: serial block-face electron microscopic analysis of the hippocampal mossy fiber synapse . J Neurosci . 33 : 507 – 522 . Google Scholar Crossref Search ADS PubMed WorldCat Wimmer VC , Horstmann H , Groh A , Kuner T . 2006 . Donut-like topology of synaptic vesicles with a central cluster of mitochondria wrapped into membrane protrusions: a novel structure-function module of the adult calyx of Held . J Neurosci . 26 : 109 – 116 . Google Scholar Crossref Search ADS PubMed WorldCat Xu-Friedman MA , Harris KM , Regehr WG . 2001 . Three-dimensional comparison of ultrastructural characteristics at depressing and facilitating synapses onto cerebellar Purkinje cells . J Neurosci . 21 : 6666 – 6672 . Google Scholar Crossref Search ADS PubMed WorldCat Xu-Friedman MA , Regehr WG . 2003 . Ultrastructural contributions to desensitization at cerebellar mossy fiber to granule cell synapses . J Neurosci . 23 : 2182 – 2192 . Google Scholar Crossref Search ADS PubMed WorldCat Xu-Friedman MA , Regehr WG . 2004 . Structural contributions to short-term synaptic plasticity . Physiol Rev . 84 : 69 – 85 . Google Scholar Crossref Search ADS PubMed WorldCat Zhang ZW , Deschênes M . 1997 . Intracortical axonal projections of lamina VI cells of the primary somatosensory cortex in the rat: a single-cell labeling study . J Neurosci . 17 : 6365 – 6379 . Google Scholar Crossref Search ADS PubMed WorldCat Zhou YD , Fuster JM . 1996 . Mnemonic neuronal activity in somatosensory cortex . Proc Natl Acad Sci USA . 93 : 10533 – 10537 . Google Scholar Crossref Search ADS PubMed WorldCat Zilles K , Palomero-Gallagher N , Bludau S , Mohlberg H , Amunts K . 2015 . Cytoarchitecture and maps of the human cerebral cortex. In: Zilles K , Amunts K) , editors . Brain Mapping: An Encyclopedic Reference, ed. Toga, AW. Section Anatomy & Physiology . San Diego : Elsevier Academic Press . p. 115 – 135 . Chapter 207. Google Preview WorldCat © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Quantitative Three-Dimensional Reconstructions of Excitatory Synaptic Boutons in Layer 5 of the Adult Human Temporal Lobe Neocortex: A Fine-Scale Electron Microscopic Analysis
JF - Cerebral Cortex
DO - 10.1093/cercor/bhy146
DA - 2019-07-05
UR - https://www.deepdyve.com/lp/oxford-university-press/quantitative-three-dimensional-reconstructions-of-excitatory-synaptic-t8e5PVBboT
SP - 2797
VL - 29
IS - 7
DP - DeepDyve
ER -