For more information:ali.ma@wecistanche.com

Pls click here to Part 1

Pls click here to Part 3

Cistanche cistanche powder health benefits --neuroprotective effect

4.3. Regulatory Mechanisms of Dendrite Arborisatin and Spine Formation

Actin and microtubule bundling and reorganization are the backbone of dendritic arborization, supporting plasma membrane expansion. Actin is located at the edges of dendrites and drives exploratory activity. Microtubules form the center of the dendritic shaft and consolidate this newly formed shaft. The dynamic process of extension, branching, retraction, as well as orientation of branches, are coordinated by numerous signaling mechanisms. These signaling mechanisms have been described in detail by other authors (see reviews from [1–3]) and will be summarised here and in Table 1.

Actin monomer polymerization allows the formation of actin filaments, stabilized by microtubules, that support axonal and dendritic branch expansion from initial actin patches (reviewed by [1]). Actin and microtubule interactions themselves are stabilized by cross-linking proteins such as microtubule-actin cross-linking factor (MACF1; [94]). While microtubule invasion of filipodia supports branch formation, the disassembly of microtubules, regulated by ubiquitin-protein ligase 3a (Ube3a, also called E6AP ubiquitin-protein ligase), equally contributes to the retraction of dendritic branches [95]. The dynamic extension and reorganization of dendritic branches is a highly activity-dependent process; therefore, this can be inhibited by processes that affect ATP production. As one example, the disruption of Drp-dependent fission will inhibit the production of mitochondria small enough to move into dendritic branches and has been found to result in the substantial failure of primary branching within mouse Purkinje cells within the cerebellum (in vivo and in vitro) [96]. Alterations in microtubule transport systems can also affect branching by altering the capacity for normal actin and microtubule organization (e.g., [97]). Relatedly, there is a substantial dependence on protein translation; therefore, alterations in mRNA transport and local protein translation also impacts neuronal arborization (reviewed by [98]). Of note, the gene fragile X mental retardation 1 (Fmr1, the gene primarily disrupted in fragile X syndrome) is an important regulator of local mRNA translation in dendrites (reviewed by [99]).

Table 1. Extrinsic regulators of dendritic arborization and spine development.

Abbreviations: Slit1—Slit guidance ligand 1, Epah7—ephrin receptor A7, Sema3A—semaphorin 3A, NGF—nerve growth factor, BDNF— brain-derived neurotrophic factor, NT—neurotrophin, IGF1—insulin-like growth factor 1, HGF1—hepatocyte growth factor 1, Rac—Rac family 1 small GTPase 1, Cdc42—cell division cycle 42, TrK—tyrosine kinase receptor, CREST—calcium-responsive transactivator, CaMKIV— calmodulin-dependent protein kinase IV, CREB—cAMP-response element-binding protein.

Extrinsic regulators of arborisation included contact-dependent signaling (cell-to-cell or with the extracellular matrix), secreted chemoattractant or chemorepellent molecules, and growth factors. Protocadherins are a large family of molecules, though, to be involved in contact-dependent regulation of dendritic arborization (and axonal extension). Clustered protocadherins have specifically been found to regulate arborization via signaling through Rho GTPases (reviewed by [100]). A conditional knockout of y-protocadherins in the mouse brain results in disrupted dendritic complexity, though the effect of this is primarily relatively late in cortical development, from P18 to P28 [67]. This relatively delayed age-dependent effect is particularly interesting as protocadherins have also been shown to regulate dendrite self-avoidance within arbors, a phenomenon that is clearly delineated by P12 [69]. Other proteins regulating cell-to-cell contacts that affect dendritic arborization include dystroglycan, contactin 4 (CNTN4), and neurexin (NRXN)–neuroligin (NRLG) interactions. Dystroglycan, an extracellular matrix protein, has been shown to stimulate the growth of dendritic arbours in mouse hippocampal neurons in vitro in a manner dependent on cell division control protein 42 (Cdc42) GTPase [70]. CNTN4, a protein associated with schizophrenia and most commonly found to alter synaptic function, has also been shown to regulate arborization of hippocampal neurons in the mouse [101]. However, in this case, the effect was relatively small, possibly reflecting the primary role of CNTN4 in synapse organization and, therefore, the late stage at which it acts in development. Neurexin and neuroligin are adhesion proteins, which are also primarily associated with synaptic formation due to their respective presence on axon terminals and dendritic spines. It has been shown in the mouse brain that interactions between these molecules are important for the initial transient stabilisation of synapses that require activity to stabilise permanently [102]. Blocking these interactions results in reduced growth of the dendritic arbour as well as less synapse formation [102].

Secreted chemoattractants that regulate dendritic arborisation include semaphorin 3A (Sema3A), Slit guidance ligand 1 (Slit1) and ephrin A7 receptor (EphA7). Sema3A, which is a chemorepellent for axons, acts as a chemoattractant for dendrites, supporting growth and orientation of the primary apical dendrite [74] and is necessary for secondary and tertiary branching [75]. Slit1 is another secreted factor, in this case working through the Robo receptor that stimulates dendritic growth and branching [71]. Interestingly, EphA7 appears to be another regulatory molecule that has different effects throughout development, as EphA7 signalling restricts dendritic growth and early spine formation during early development (prior to P10 in the mouse) but promotes dendritic spine formation at later developmental stages [54].

The family of neurotrophic factors, including nerve growth factor (NGF), brain-derived neurotropic factor (BDNF), and neurotrophin-3 and -4 (NT-3, NT-4), that have multiple roles in regulating brain development and function, has also been identified as contributing to neuronal arborisation and synapse formation. The functions of BDNF in this context have been most widely explored and found to promote dendritic arborization, primarily via activation of the tropomyosin receptor kinase (Trk) B receptor (TrkB, reviewed in detail by [56]). NT-3, acting through the TrkC receptor, has also been found to promote arbour growth [81]. Mature BDNF and NT-3 both have developmental-specific expression and appear to contribute primarily to the later stages of dendritic expansion. In an elegant study, Joo et al. (2014) showed that decreasing NT-3 signaling, derived from pre-synaptic neurons in an activity-dependent manner, reduced elongation of dendrites from Purkinje neurons to the pial surface of the mouse brain between P7 and P14 (resulting in disruption that continues long-term) [81]. However, other researchers have shown the capacity for inhibition of NT-3 from E21 to P7 in the mouse to result in cortical neurons with reduced apical dendrites at P7 [56], suggesting the age-specific effects may be different depending on the brain region or NT-3 source. The actions of BDNF on dendritic development are harder to summarise and can be both constitutive and activity-dependent [79]. It is likely that a molecule as pleiotropic as BDNF has substantial compartmentalization of its signaling and signalling pathways. In support of this idea, it has recently been shown, using compartmentalised cultures of primary rodent neurons, that BDNF from post-synaptic targets is able to bind to TrkB receptors on axons and simulate dendritic arborisation via a distinct and complex intracellular signalling pathway [78]. Hepatocyte growth factor (HGF) also acts through its tyrosine kinase receptor (Met in this case), is important in stimulating many developmental events, and has been found to promote dendritic growth and branching [78,84]. Hepatocyte growth factor (HGF) also acts through its tyrosine kinase receptor (Met in this case), is important stimulating many developmental events, and has been found to promote dendritic growth and branching [84]. This is of particular interest, as HGF, as with many other growth factors, has been found to be reduced in ASD patients [86]. Similarly, neuregulins, members of the epidermal growth factor (EGF) family, have been linked to altered synaptic function in neurological disease [86]. As one example, neuregulin 1 binding to the EGF receptor ErbB4 has been shown to alter dendrite elaboration and synapse formation in glutamatergic and some GABAergic primary hippocampal neurons [103].

While the downstream signalling pathways have not been established for all extrinsic regulators of arborization, there are clearly several common and interacting pathways that ultimately regulate actin and microtubule organization. BDNF, for instance, is well established to alter dendritic arborisation via PI3K/Akt, cAMP response-binding protein (CREB), and other early response genes such as Arc and Rho-GTPases (reviewed in [79]). Ras- and Rho-GTPases, together with many protein kinases, e.g., glycogen synthase kinase- 3 (GSK-3) and PI3K and CREB/CREB-binding protein (CBP) can be considered master regulators of dendritic arborisation (reviewed in [2]). These molecules have numerous interactions and can directly or indirectly affect cytoskeletal rearrangement. For instance, active GSK-3 phosphorylates MACF1 altering its interactions with microtubules and f- actin in the leading edges of extended neurites [94]. Knockout experiments in the mouse suggest that MACF1 is important throughout the developmental process, with knockout resulting in increased numbers of primary dendrites, but a reduced dendritic length and abnormal dendritic orientation when modified at different stages of development in vitro, as well as in vivo [94]. An alternative method of modifying these pathways, and, therefore, dendritic arborisation, is via transcriptional regulators and chromatin remodelling proteins. ARD1B is an example of a chromatin regulation molecule that affects the production of cfos and Arc (via the phosphorylation of CREB), and many downstream pathways which contribute to dendritic branching [55]. The effects of altering ARD1B binding in the mouse brain (reduced number and length of apical and basal dendrites, substantial disruption in pial contact, reduced number and increased immaturity of dendritic spines) are complex, reflecting the combined actions of these pathways [55].

While many of the mechanisms described above are activity-independent, a large part of the dynamic phase of dendritic reorganisation is activity-dependent. Signalling following neurotransmitter release and binding is essential for stabilising dendritic growth and spine and synapse formation, with many of the signalling molecules and downstream pathways described above important for the cytoskeletal reorganisation entailed (reviewed in [2]). Sin et al. (2002) showed in the optic tectum of Xenopus tadpoles that light stimulation increased dendritic arborisation by a mechanism that involved glutamate receptor signalling, resulting in reduced RhoA activity and increased Rac and Cdc42 signalling [89]. Interestingly, a subsequent study on the mouse, using dark rearing to explore activity-dependent effects on dendritic arborisation, showed similar changes but of a greatly reduced magnitude [89,104]. This suggests there could be species or brain region differences in the contribution of activity-dependent control or arborisation. Disruption in the process of neurotransmitter signalling is, therefore, clearly a contributor to neurological disorder. Transfection of an ASD-specific mutation of GRIN2 (encoding the GluN2B NMDA receptor subunit) into a subpopulation of cultured neurons in the presence of wildtype GluN2B shows that even small changes in the presence of key genes/proteins can affect dendritic arborisation [104]. In this example, there was no significant change in primary neurons, but a substantial decrease in branching, and a decrease in the length of branches at all levels of the dendritic tree [104].

To date, research on glial involvement in the successful wiring of the brain has focused on effects of astrocytes and microglia at the synapse level, specifically regulating dendritic spine number and structure and synapse formation, plasticity, and function. There is substantial evidence that astrocytes affect synapse formation through a mixture of contact-dependent mechanisms and secreted factors, including thrombospondin, cholesterol and ApoE, hevin, transforming growth factor (TFG)-β, and chondroitin sulphate proteoglycan (reviewed by [105,106]). Astrocyte–neuronal contact, regulated by y-protocadherin, appears to be important for facilitating early phases of synapse formation [68]. Secreted BDNF appears to be a major driver of microglial-influenced dendritic spine formation, and there is also substantial evidence of microglial phagocytosis of pre-and post-synaptic elements as part of the normal pruning of excess synaptic connections, particularly those with low activity (reviewed by [107]). Phagocytosis is regulated by classical immune signaling systems, such as chemokines (specifically Cx3cr1) and complement (including C3 and C1q; [107–110]). Importantly, though beyond the scope of this review, there is also substantial data supporting a role of microglia and astrocytes in the regulation of synaptic plasticity in the adult brain (reviewed by [106,107]). Limited data is available as to whether these glial cells contribute to earlier stages of neuronal development such as dendritic arborization, though Yang et al. (2012) [82] have shown astrocyte-dependent dendritic arborization in the mouse brain. Interestingly, this study suggested that astrocyte knockout of Fmr1 was sufficient to induce the reduced dendritic arborization, as a result of an over-production of NT-3 [82]. Recent data also suggest that astrocytes may regulate neuronal arborization through the lipoprotein receptor class A repeat domain of low-density lipoprotein receptor-related protein 4 (LRP4, [111]).

5. Potential Protective Regulatory Mechanisms and Pharmacotherapies

5.1. Genetic Risk Factors That Alter Neuronal Arborisation and Associate with Neurodevelopmental Disorders

The high frequency of dendrite and spine abnormalities in neurodevelopmental disorders are due to a mixture of genetic susceptibility and an altered environment during development (possibly reflecting a specific injury in some cases). Many of the genetic risk factors associated with neurodevelopmental disorders (and neurodegeneration) interact with the molecular regulators of dendritic development outlined above and are summarised in Table 2. The clearest association between genetics and neurodevelopmental disorders occurs with chromosomal deletions/duplication (e.g., Methyl CpG binding protein 2, MeCP2; 22q13 etc.); however, as so many genes are affected, the neurobiology can be difficult to unpick (reviewed in [112]). MeCP2 is a transcriptional regulator that has significant effects of brain development. The effects of this gene are dose- and sex-dependent: Rett syndrome occurs in females with a reduction in MeCP2 [112–114], and duplication of the gene results in an increased risk of ASD and intellectual disability, primarily in males (reviewed in [113]). In terms of arborization, deficiency in MeCP2 results in reduced dendritic branches, spine density, and abnormal spine morphology [113]. Interestingly, the pro-inflammatory cytokine interleukin-1β has been shown to interact with MeCP2 and affect its functioning (reviewed by [115]), a possible mechanism by which genetic and environmental factors could interact to increase the risk of neurodevelopmental disorder.