Part 1:Connecting The Neurobiology Of Developmental Brain Injury: Neuronal Arborisation As A Regulator Of Dysfunction And Potential Therapeutic Target
Mar 21, 2022
ali.ma@wecistanche.com
Click to Cistanche herba for anti-Alzheimer's disease
Ane Goikolea-Vives and Helen B. Stolp *
Department of Comparative Biomedical Sciences, Royal Veterinary College, London NW1 0TU, UK; agoikoleavive18@rvc.ac.uk
* Correspondence: hstolp@rvc.ac.uk
Abstract: Neurodevelopmental disorders can derive from a complex combination of genetic variation and environmental pressures on key developmental processes. Despite this complex etiology, and the equally complex array of syndromes and conditions diagnosed under the heading of neurodevelopmental disorder, there are parallels in the neuropathology of these conditions that suggest overlapping mechanisms of cellular injury and dysfunction. Neuronal arborization is a process of dendrite and axon extension that is essential for the connectivity between neurons that underlies normal brain function. Disrupted arborization and synapse formation are commonly reported in neurodevelopmental disorders. Here, we summarise the evidence for disrupted neuronal arborist-check for Citation: Goikolea-Vives, A.; Stolp, H.B. Connecting the Neurobiology of Developmental Brain Injury: Neuronal Arborisation as a Regulator of Dysfunction and Potential Therapeutic Target. Int. J. Mol. Sci. 2021, 22, 8220. https://doi.org/ 10.3390/ijms22158220
Academic Editor:
Giuseppe Lazzarino
Received: 2 July 2021
Accepted: 28 July 2021
Published: 30 July 2021
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open-access article distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).
tion in these conditions, focusing primarily on the cortex and hippocampus. In addition, we explore the developmentally specific mechanisms by which neuronal arborization is regulated. Finally, we discuss key regulators of neuronal arborization that could link to neurodevelopmental disease and the potential for pharmacological modification of arborization and the formation of synaptic connections that may provide therapeutic benefit in the future.
Keywords: dendritic arborization; dendritic spine; synapse formation; neurodevelopmental disorder; perinatal brain injury
1. Introduction
Dendritic arbors, together with dendritic spines in spiny neurons, are fundamental in regulating both the information received by a neuron and the way that this information is processed and acted upon. As a result, changes in dendritic arborization, or the formation of dendritic spines, have a dramatic effect on brain function. This is evidenced by a substantial body of research correlating alterations in dendrites and dendritic spines with the severity of cognitive and behavioral symptoms of neurodevelopmental, neuropsychiatric, and neurodegenerative disorders.
Dendrite formation is a relatively late and extended developmental event, after a prolonged period of proliferation, that follows a broadly stereotypic pattern for all neurons. It is driven by a combination of intrinsic genetically regulated processes, particularly important during early phases of neurite extension, that are then dynamically influenced by a multitude of extrinsic cues, including activity-dependent regulation [1–3]. For the identification of potential therapies to correct disrupted arborization and connectivity in neurodevelopmental disorders, it is necessary to understand the consequences of genetic and environmental events on dendritic arborization, the time-dependence of these dis- ruptions, and the capacity for structural or functional compensation as part of normal development. While progress is being made in our understanding of many of these areas, there is still a lack of overview and integration of information necessary to make the re-quired progress in therapeutic discovery. To facilitate this progress, we will review the links between neurodevelopmental disorders and disrupted dendritic development, considering the potential consequences of disruption for the functioning of neural networks. The mechanisms underlying alterations in dendritic and synaptic density in neurodevelopmental disorders will be explored, particularly focusing on those mechanisms that show promise for therapeutic intervention.
2. Disrupted Neuronal Arborisation in Neurodevelopmental Disorders
Neuronal morphology is a major determinant of neuronal connectivity and normal brain function [4,5]. The dendritic branching pattern, as well as dendritic and spine density, size, and morphology, determines the efficacy of the synaptic input transmission, integration, and processing [5,6]. Many neurodevelopmental pathologies exhibit dendritic and spine abnormalities, summarized in Figure 1 [7–10]. For instance, brain post-mortem studies from autistic patients reported reduced dendritic branching complexity in the hippocampal CA1 and CA4 regions [11], a reduction in the number of dendrites in the dorsolateral prefrontal cortex [12], and increased spine densities in cortical pyramidal neurons [13]. Dendritic abnormalities are a core feature of syndromes such as Down syndrome, Rett syndrome, fragile X syndrome, and phenylketonuria; patients displayed a decreased number and length of dendritic arbors as well as abnormal morphology and number of dendritic spines in the cerebral cortex (reviewed in [14]). Disorders such as epilepsy and traumatic brain injuries (TBI), in which excitotoxicity is involved, have also been associated with aberrant dendritic spine structure and distribution [15]. Post-mortem reports from patients with epilepsy showed decreased dendritic branching complexity, fewer branches, as well as decreased spine density and dendritic swelling in layer III cortical pyramidal neurons [16]. Dendritic varicosities and loss of dendritic spines have been observed in the hippocampus of these patients [17]. Perinatal hypoxic/ischemic brain injury can result in long-term neurologic defects or death of the new-born (reviewed in [18]). Animal studies have shown that hypoxic-ischemic events lead to the loss of dendritic spines, appearance of dendritic varicosities, reduced dendritic length, and dendritic branching in rat cortical pyramidal neurons [19,20] and sheep cortical and subcortical neurons [21–24].
Figure 1. Schematic diagram summarising disease-associated alterations in dendritic arborization and spine formation. A neurotypical neuron elaborates complex branching and long and numerous dendrites. It also develops relatively stable and mature spines. Individuals with ASD often show reduced dendritic branching complexity and increased presence of immature/thorny spines and increased spine density. Neurons from fragile X and Rett syndrome patients have been found to display shorter and less abundant dendrites and abnormally long and thin spines of increased density. In patients with epilepsy, neurons have been observed to form shorter and less branched arbors that often display varicosities and distorted spines, and spine density is also decreased. Neurons from schizophrenic individuals display decreased spine size and abnormal spine necks, smaller somas, and a reduced number of dendrites. In subjects with hypoxia/ischemia or TBI, neurons elaborated less and shorter dendrites, with varicosities and constrictions, and showed reduced spine density and the presence of abnormal spines. Decreased dendritic length and branching as well as spine loss has been observed in neurons from patients with AD and PD. AD—Alzheimer’s disease, ASD—autism spectrum disorder, PD—Parkinson’s disease, TBI—traumatic brain injury.
3. Neuronal Arborisation and Synapse Formation as Part of Cortical Circuit Formation
The earliest cortical circuits in humans are formed in the prelate by gestational week 5 [25–28]. Neurons within the prelate create primitive and temporary synaptic connections with adjacent cells acting like provisional targets until migrating neurons arrive to form more stable connections. These preplate neurons are also the first neurons to project outside the cerebral cortex. As the cortical laminae develop, neurons generate short- and long-distance connections to create local and globally interlinked neural networks. Developmental processes following migration and differentiation initially result in excessive neuronal arborization and synaptic connectivity. These require refinement, first by spontaneous activity and later by extrinsic stimuli-dependent activity, in order to form and establish mature neural circuits [29]. Spontaneous neuronal activity is necessary for the initial development of connectivity, and it reaches the cerebral cortex through the thalamocortical pathway, even before the radial migration of cortical neurons has been completed [30]. Seminal studies conducted in the visual system of prenatal cats demonstrated that blockage of spontaneous firing of action potentials before eye-opening impaired normal axon terminal branching of retinal ganglion cells [31] and of thalamocortical pathway neurons, which led to the aberrant formation of ocular dominance columns in the primary visual cortex [32]. In the developing mouse somatosensory cortex, before any sensory stimuli can be received, the absence of spontaneous activity that originates from the thalamus results in cortical hyperexcitable circuits and aberrant development of functional columnar structures [33].
As the brain matures and begins to receive sensory input, the number, type, and strength of synapses varies as a result of neuronal activity. This neuronal activity not only enables the addition of unique information into neuronal patterns but also promotes circuit refinement and is essential in the development of mature circuitry. After birth, dendritic morphogenesis is particularly susceptible to activity-dependent inputs, and it is crucial to determine the neuronal dendritic structure and the type of connections to establish [34]. Additionally, dendritic branches can remodel in response to damage caused by injury or disease. The capacity to reshape and adapt to change is termed plasticity, and it has been shown to be present throughout adult life [35,36].
During the first 18 months of life, the rate of dendritic morphogenesis and synaptogenesis increases, and developmental processes such as experience-dependent synapse remodeling and pruning approach a critical period in which incorrect timing and rate has been proposed to lead to the development of several neurodevelopmental disorders [26,37,38]. Dendritic and synaptic pruning is driven by an interplay of neurons, microglia and astrocytes [39]. Pruning occurs in two phases: directly after birth—early childhood—to ensure the correct formation of sensory circuits; and during the transition from childhood, adolescence, and adulthood to remodel circuits involved in higher cognitive functions including self-regulation [39,40]. Abnormal pruning leads to aberrant dendritic arborization and synaptic function.
Interestingly, it appears that dendrite maturation and expression of behavioral symptoms of some neurodevelopmental disorders are temporally correlated [41]. For instance, autism spectrum disorder (ASD) onset matches with the dendritic growth and arborization that occurs during early childhood [42], and the expression of attention-deficit hyperactivity disorder (ADHD) symptoms at mid and late childhood and schizophrenia during late adolescence with dendritic and synaptic pruning [40,41]. ASD pathology has been associated with disrupted excitatory/inhibitory balance and abnormal connectivity of higher-order association areas [43]. Since ASD is often accompanied by an increased brain size during the first 3 years of life, it has been postulated that it could be due to a dendritic overgrowth or deficiency in pruning and maintenance of normal cell numbers [42]. ADHD and Tourette’s syndrome appear late in childhood and are characterized by a deficient connectivity in neural circuits associated with self-regulation and inhibitory capacity. Patients with ADHD have also been shown to reach peak cortical thickness later than their neurotypical counterparts [44]. The typical onset of schizophrenia occurs during adolescence or young adulthood. In schizophrenic patients, during puberty, cortical thinning occurs at a faster pace and extends to neighboring regions compared to age-matched controls [45]. The reason for the excessive thinning has been proposed to be either due to reduced dendritic branching and decreased cell number, or more commonly due to excessive synaptic pruning or irregular synaptic remodeling [46,47]. From this information, we can conclude that a failure to maintain correct dendritic maturation leads to abnormal neuronal function and circuit establishment, which ultimately results in the development of atypical behavioral symptoms associated with neurodevelopmental disorders. In this context, it is important to now consider the specific timing of dendritic maturational events and the signaling mechanisms that underpin them.
4. Developmental Timeline and Regulation of Neuronal Arborisation and Synapse Formation
The morphology of mature neurons is characterized by the multitude of highly branched processes that extend from the cell body. These neurites initially extend in a similar manner, prior to specialization, into axons and dendrites, with spine formation part of the late dendritic specialization in spiny neurons [2,3,8]. Phases of dendritic arborization can be summarised as (i) growth (characterized by an initial slow phase and the subsequent fast elongation), followed by (ii) dynamic extension and retraction, leading into a final period of (iii) dendrite stabilization (Figure 2) [2]. In addition to these neuronal-dependent processes, there is a prolonged period of pruning that is particularly dependent on environmental cues (reviewed in [1]). Therefore, there is a stereotypic element of dendritic arborization, though there is variation in the timing of events between individual cell types, brain regions, and species.
Figure 2. Schematic diagram summarising the comparative timeframe in which key events in dendritic arborization occur. Dendritic branching commences with primary branch formation, immediately following neuronal migration to its final position. Processes of branching then occur to form secondary and tertiary branches, as well as branch elongation. During this period, is the initial formation of dendritic spines and, subsequently, of synapses. Reorganization and stabilization of the dendritic branches, spines, and synapses occur relatively late in the developmental processes. The comparative time frame of these events is shown for the mouse, sheep, non-human primate, and human brain. E—embryonic day, GW—gestational week, P—postnatal day.
4.1. Progression and Timing of Dendrite Development
Data collected from numerous species shows a general pattern of dendritic expansion during the early years of life, where the lengthening of dendritic branches and increasing branch complexity correlate with synaptic formation. This process stabilizes before a period of reorganization and synaptic pruning during early adolescence as adult patterns of arborization and synaptic connectivity are established (see Figure 2). Typical early-branching patterns, visualized with Golgi staining, show a single apical primary dendrite together with 3–9 basal dendrites [48]. These extend following the completion of migration, with the leading migratory edge thought to transition from the primary apical neurite at this point. The switch from migration to dendritic extension appears, at least in the mouse, to be facilitated by the removal of Sox11 inhibition in the early postnatal cortex [49]. The axon generally specializes from a basal neurite, while the rest contribute to the dendritic tree. In the human brain, the first dendritic branching has been reported between 16–26 weeks of gestation, increasing up to 36 weeks [50,51], and an established, though rudimentary, dendritic structure is present at term, with cortical neurons 30–55% of their maximum length [52]. A similarly established arborization has been described at term for non-human primates [53]. In cortical neurons, the basal dendrites appear to establish their complexity earlier than the apical dendrites, with no new branch orders identified in basal dendrites after term [52]. In the mouse brain, these steps largely occur in the first postnatal week, with branching broadly equivalent to the human term, around postnatal day (P)7-10 (see comparative data in [49,54–56] as examples). In the sheep, another common animal for modeling developmental brain injury, dendritic arborization within the cortex commences at around 0.7–0.85 of gestation (see data in [21,23,24]).
An analysis of layer V neurons in the human prefrontal cortex suggests that there is a rapid phase of dendritic expansion and branching that continues until 5 years of age [57]. This is then followed by a long period of local dynamic reorganization of the dendritic branches. The majority of data regarding this period of dynamic reorganization comes from rodent or cell culture studies exploring the molecular mechanisms regulating these stages of dendritic arborization (discussed below). A substantial body of work in zebrafish also exists, utilizing the capacity for genetically enhanced time-lapse imaging to unpick specific developmental events. While this review focuses primarily on findings from the mammalian brain, the zebrafish data is an important addition to the field and is reviewed in [58]. Data from the mouse brain clearly shows the formation of primary branches by P10, with the continued elaboration of secondary and tertiary branches until approximately P40 [50]. Variation in dendritic arborization between cortical regions is detectable in the neonatal human brain, with the primary motor cortex appearing to develop first (based on identification of longer dendrites and a greater number of dendritic spines) [59]. Synaptic density in the visual and auditory regions also appears to develop relatively early [60]. In both measures, the prefrontal cortex appears to lag in its maturation, with less complex dendritic branches [52] and a reduced synapse number in early life [60]. Data from the chimp brain shows a similar pattern of maturation: neurons in the prefrontal cortex continue to be less elaborate until after adolescent pruning, though they ultimately show more complexity in their branching pattern than neurons in other cortical regions [53].
Evidence for sex differences in dendrite arborization is beginning to grow, from a mixture of in vivo and in vitro studies. These studies show a clearly increased complexity within the dendritic arbors of hippocampal neurons in male mice at P28, compared to their female counterparts [61]. This result was replicated in primary neuronal-glial cultures from P0 hippocampal tissue in the same mouse strain and appeared to be at least partly estrogen-dependent [61]. These differences in the formation of dendritic arbors between males and females may help explain the well-recognized sex differences in the presentation and diagnosis of neurodevelopmental disorders. While the study of Keil et al. (2017) links these sex differences to activation of the estrogen receptor [61], a study by Beyer and Karolczah (2000) on primary mouse midbrain dopaminergic neuronal cultures suggests that estrogen may also stimulate the growth of neurons independently of the
estrogen receptor, instead of being dependent on cAMP- and PKA-derived phosphorylation of CREB [62]. Additionally, there is some evidence from mouse studies to suggest that sex differences in microglial development (innate and following inflammation) may contribute to the observed differences in neuronal arborization and synapse number [63].
4.2. Dendritic Spines and Synaptic Development
Dendritic spines are microscopic membrane protrusions comprising the receptive postsynaptic compartment of synapses in the brain [15]. Spines contain neurotransmitters, neuropeptides, receptors, signaling molecules, ion channels, and other proteins that participate in synaptic transmission. Newly formed dendrites lack synapses and spines. During spinogenesis, thin finger-like dynamic protrusions named filopodia emerge from the dendritic arbor. These filopodia can form immature synapses at contact with axons; synapses can occur along the whole length of the filopodium and at its base and can receive multiple synapses [64]. As spinogenesis proceeds, filopodia length and frequency decreases, and dendrites start to produce thin, stubby, and mature mushroom-shaped spines from retracted filopodia [65].
Dendritic spine formation on spiny neurons follows dendritic branching after a natural delay, with immature spines detectable on neurons in the human hippocampus by 36 weeks of gestation, a time when multiple dendritic branches are present [50]. Spine formation is likely to occur even earlier in the cortex, as synapses can be detected from as early as 27 weeks post-conception age, ranging from 3–10 synapse/100 um depending on cortical region [60]. The synaptic density increases to a peak (~60 synapses/100 um) at around 4 years of age and then declines during adolescence to an adult density of approximately 35 synapses/100 um [60]. In the mouse cortex, spines are clearly present in immature states prior to P10 and visibly mature by P20 [54]. Synapses are detectable from P5, increasing rapidly to a large stable number between P10–17 [66]. Patterns of synapse formation vary throughout the brain, beginning earlier in inner cortical layers (V, VI) compared to outer ones (II, III), following the inside-out development of the cortex [60].
