Invaginating Structures in Synapses – Perspective

Abstract

Invaginating structures are common in the synapses of most animals. However, the details of these invaginating structures remain understudied in part because they are not well resolved in light microscopy and were often misidentified in early electron microscope (EM) studies. Utilizing experimental techniques along with the latest advances in microscopy, such as focused ion beam-scanning EM (FIB-SEM), evidence is gradually building to suggest that the synaptic invaginating structures contribute to synapse development, maintenance, and plasticity. These invaginating structures are most elaborate in synapses mediating rapid integration of signals, such as muscle contraction, mechanoreception, and vision. Here we argue that the synaptic invaginations should be considered in future studies seeking to understand their role in sensory integration and coordination, learning, and memory. We review the various types of invaginating structures in the synapses and discuss their potential functions. We also present several new examples of invaginating structures from a variety of animals including Drosophila and mice, mainly using FIB-SEM, with which we trace the form and arrangement of these structures.

Multiple Invagination Patterns and Synaptic Efficacy in Primate and Mouse Rod Synaptic Terminals

“…To elucidate the apparent discrepancy in the number of invaginating neurites between primates and other mammals is valuable when we apply experimental findings on mice to the regenerative research of rod synapses. 16 – 20 In this study, we compared the synaptic architecture of RSs between macaque, mouse, and human retinas focusing on their size-related configuration differences.…”

Pathways and Processes Underpinning Axonal Biology and Pathobiology

“…In addition, we document striking alterations in the structure of dendritic spines contacted by synapses of orphaned axons including dramatic hypertrophy of spine heads, enlargement and segmentation post-synaptic membrane specializations (PSDs), and development of spinules. These increases in the complexity of spine morphology are very similar to what is seen following intense synaptic activity and induction of LTP [for a recent review, see (Petralia et al, 2021)]. Robust and paradoxical spine growth during the time period when spines in WT mice are withering and retracting suggests yet to be characterized signaling processes between orphaned axons and their postsynaptic targets.…”

Spindle-Shaped Neurons in the Human Posteromedial (Precuneus) Cortex

“…Spines with different forms (including those with convoluted structure) can differ in their impact on the fine-tuned synaptic processing by having different postsynaptic density composition, number and type of postsynaptic receptors, subcellular components and organization, electrical and biochemical compartmentalization, clustering pattern, degree of cooperativity between adjacent spines and the parent dendrite, and impact on the neuronal voltage and output frequency ( Rochefort and Konnerth, 2012 ; Yadav et al, 2012 ; Yuste, 2013 ; Stewart et al, 2014 ; Dall’Oglio et al, 2015 ; Tønnesen and Nägerl, 2016 ; Berry and Nedivi, 2017 ; Lu and Zuo, 2017 ; Nakahata and Yasuda, 2018 ). Spinules, also found in PC spindle-shaped neurons, are active functional elements for synaptic development and maintenance that add to the neuronal plasticity repertoire and rapid integration of signals ( Petralia et al, 2018 , 2021 ). For example, NMDA activation can increase spinule number, length, and contact with distal presynaptic elements ( Zaccard et al, 2020 ).…”

“…According to these morphological features, spines were classified as (1) thin (2) stubby, (3) wide, (4) mushroom-like, (5) ramified, and (6) with a transitional aspect between these classes or as "atypical" (or "multiform") spines with more complex and varied shapes (Dall'Oglio et al, 2015 and references therein). Tiny protrusions extending from the head of a spine were classified as spinules (Brusco et al, 2014;Zaccard et al, 2020;Petralia et al, 2021). All computational procedures were run using Windows Microsoft R (version 10), Intel R Core TM i7-8750H CPU @2.20 GHz, 16.0 GB RAM memory, NVIDIA R GeForce GTX 1050 Ti with 4 GB for image processing.…”