Central nervous system (CNS) astrocytes are glial cells performing crucial tasks encompassing energy metabolism, neurotransmission, ion and water stable levels, and immune defense and control local blood flow/oxygen levels. Arising from neural stem cells, astrocytes differentiate into subtypes that vary according to animal species. Human cerebral cortex astrocytes are sturdier and cytologically and functionally more complex, control wider domains, and spread calcium signals more quickly than their rodents’ counterparts. They actively partake in CNS homeostasis maintenance and functioning by teaming up with their client neurons, other glial cell types, and cerebrovascular cells. Alterations of astrocytes’ activities deeply impact on age-related chronic ailments like Alzheimer’s disease (AD), the commonest senile dementia; AD involves the growing accumulation of amyloid-β peptides (Aβs) and hyperphosphorylated Tau proteins the astrocytes, and neurons supply following the interaction of their calcium-sensing receptors (CaSRs) with exogenous Aβs. The activated Aβ∙CaSR signaling triggers a self-propagating mechanism that spreads the neuropathology among adjacent and far away astrocytes and their neuronal clients causing neurons’ death. CaSR antagonists or calcilytics suppress these noxious effects in vitro. Hence, calcilytics are potential therapeutics that could halt the spread of AD neuropathology and safeguard the patients’ neuronal viability, cognition, memory, and ultimately life.
- Alzheimer’s disease
- tau protein
- calcium-sensing receptor
Between the 16th and 18th week of intrauterine life, a pool of stem cells of the neural plate generates every neural cell type, excepting microglia, in humans. Once differentiated, the astrocytes undergo a complex maturing process through which they acquire their specific morpho-functional characteristics. When these processes achieve completion, human astrocytes account for up to 50%, if not more, of the entire CNS cell population. These cells are larger in size and endowed with more numerous branches than their much less abundant (<20% of all CNS cells) rodents’ counterparts . Being so plentiful, astrocytes have a relevant role in brain environment homeostasis maintenance [2, 3]. They metabolically sustain neurons, recycle neurotransmitters, affect synapse activity, control local blood flow, and partake in blood-brain barrier functional integrity (see for details [2, 3, 4] and below). Aging and CNS diseases, neurodegenerative ones included, can induce an activated or inflammatory or reactive condition in the astrocytes [5, 6].
2. Human astrocytes’ varieties
As their designation indicates, astrocytes have a typical star-shaped morphology as they emit different numbers of cytoplasmic branches according to their subtype. Astrocytes of several subtypes dwell in the human CNS. Some of them display locational predilections, e.g., fibrous astrocytes for the white matter and protoplasmic astrocytes for the gray matter. More recently, it has been realized that the classical protoplasmic and fibrous astrocytes can be differentiated into several subtypes, some of which proper only of the human cerebral cortex. Such subtypes share a specific marker, the glial fibrillary acidic protein (GFAP), which is an intermediate filaments’ constituent expressed by all the astrocytes cultured in vitro. However, in vivo only the fibrous astrocytes express GFAP in the white matter (see for Ref. ). Recent studies have singled out a novel marker expressed by both protoplasmic and fibrous astrocytes, the aldehyde dehydrogenase-1 family member L-1 (Aldh1L1) [1, 8, 9].
2.1. Radial astrocytes
Radial astrocytes are the first ones to appear in the course of embryogenesis. At that point in time, they aid neurons’ migration by acting as scaffolds . Later, they differentiate becoming
2.2. Fibrous astrocytes
These white-matter-located astrocytes present very long and thin processes which mostly do not emit branches. The processes’ terminal end-feet envelope the axonal nodes of Ranvier and also gets in touch with the walls of cerebral vessels. Most notably, fibrous astrocytes partake in the repair of injured brain tissue, especially at the spinal cord level .
2.3. Protoplasmic astrocytes
They are the most abundant astrocytic type. Their somata give out numerous (up to 200) long branches, which end up with leafy feet or end-feet in part touching the blood vessels’ walls and in part enwrapping several thousands of synapses [1, 8]. Near the pia mater’s inner surface, the astrocytes’ end-feet cluster together forms the CNS
It is worth recalling here that both Golgi silver staining and GFAP immunolabeling of brain tissue sections make the astrocytes appear as star-like cells. However, the astrocytes are the possessors of a certain number of cytoplasmic branches these methods do not stain. Therefore, such methods do not reveal the astrocytes’ true morphology as visible under the light and/or fluorescence microscope. Another concept of old is that during development the astrocytes’ branches form an interdigitated scaffold permitting the organization of the neurons. Recently, it has become clear that independent and distinct astrocytic domains develop with no connection with similar neighboring domains within the hippocampus . As abovementioned, the morpho-functional features of human protoplasmic and fibrous astrocytes differ from rodents’ ones. For instance, the diameters of gray matter-located human protoplasmic astrocytes are 2.6-fold longer, and their GFAP-positive processes are 10-fold more abundant. A single protoplasmic astrocyte can control from 270,000 to 2.0 million synapses placed inside its spatial domain. Most important, the branches of a single astrocyte touch, envelop, and regulate not only a huge number of synapses but also the capillary vessels controlling the blood flow going to those same synapses. This organized structure has been interpreted as the indication of a control of synaptic activity by the astrocytes independently of neuronal activity. Although unable to transmit neural impulses, human astrocytes propagate calcium ion [Ca2+] waves at speeds of up to 36 μm/s, i.e., 4–10-fold faster than rodents’ astrocytes do [15, 16, 17].
2.4. Additional astrocytes’ subtypes
Besides the above-described canonical kinds, several other astrocyte subtypes have been recognized. Emsley and Macklis  have used a combined approach consisting of S100β immunostaining, GFAP expression, and human GFAP promoter-prodded enhanced green fluorescent protein (eGFP) expression in transgenic mice, to identify within several subtypes of CNS astrocytes. The latter incorporate radial glia, protoplasmic astrocytes, fibrous astrocytes, ependymal glia, tanycytes, Bergmann glia, and velate glia. The cytoarchitectonics and functional requirements of their local placements mainly determine the morphological features, growth rates, and relative densities of these subtypes . NG2 cells are an additional CNS glial cell type likely possessed of stem cell features and hence capable of giving raise to astrocytes, neurons, and oligodendrocytes (OLGs) during both intra- and extrauterine life. NG2 glial cells functionally interact with neurons at the level of synapses. Studies are under way to clarify the heterogeneity of NG2 glia .
2.5. Human cortex-specific astrocytic subtypes
At variance with other mammalian species, humans have developed two novel cerebral cortical astroglia subtypes: the
3. Astrocytes’ physiology
In the past and still now, some scientists have been holding astrocytes as neuron-supporting and at the same time debris-scavenging cells protectively regulating the homeostasis of a microenvironment from which neurons derive the necessary nutrients [12, 19]. Astrocytes also control the workings of “tripartite synapses” by enveloping them with their branches, thus barring the diffusion of released neurotransmitters and preventing the firing activity of one neuron from altering that of adjacent neurons . In addition, astrocytes’ synaptic regulation does not influence only the tripartite synapses their branches envelop but also far away synapses via astrocytes’ signals, a process named
Since astrocytes cannot be electrically excited, their plasma membranes do not propagate action potentials as instead neurons do. The membrane potential of astrocytes at rest has very low values, ranging from −85 to −90 mV. This is due to their intense expression of TREK-1 and TWIK-1 potassium ion [K+] channels . As recent lines of evidence show, astrocytes residing in separate brain areas express dissimilar types and levels of ion channels and hence are equipped with distinctive electrophysiological characteristics. The huge group of ion channels implicated is also differently expressed during astrocytes’ developmental stages .
In addition, astrocytes express various kinds of metabotropic receptors, which are coupled to a number of intracellular second messenger systems. For example, astroglia are known to adjust neuronal excitability and synaptic transmission through the metabotropic glutamatergic receptor subtype 5 (mGluR5). The results of experiments using brain slices showed that in response to an assortment of neurotransmitters, comprising acetylcholine, adenosine, ATP, endocannabinoids, GABA, glutamate, norepinephrine, and prostaglandins, metabotropic receptors could raise the intracellular Ca2+ levels ([Ca2+]i) via phospholipase C (PLC)- and inositol (1,4,5)-triphosphate (IP3)-dependent activities .
4. Astrocytes and AD neuropathology
An aberrant reactivity of astrocytes is a telltale sign of chronic neurodegenerative ailments like AD and Parkinson’s disease [1, 3, 5]. While AD advances an
AD hits nearly 2% of the people of the Western world particularly after 60 years of age . AD’s clinical course can be dissected into (
In the healthy brain, neurons produce and release at their synapses tiny amounts of nontoxic Aβ42 monomers, the intra- and extracellular amounts of which remain at low (i.e., pM), physiological values owing to a set of removing mechanisms operated by several proteases, phagocytosis by microglia and astrocytes, and disposal into the blood circulation . In aged brains, the ability to clear the Aβs from the CNS increasingly plummets likely because of local microcirculation problems. Consequently, as the
Conversely, as the
Moreover, Aβ42-os and Aβ42 fibrillar aggregates bind various plasma membrane receptors, comprising the calcium-sensing receptors (CaSRs) and the receptors for advanced glycation end products (RAGEs) which can activate the astrocytes (see for further details [3, 11, 43, 44]). Such multiple receptor interactions with Aβs stir up astrocytes’ JAK2 and MEK1/MEK2/ERK-1/ERK-2 signaling pathways stimulate the direct binding of STAT1 and HIF-1α/HIF-1β complexes to the
5. The CaSR
A highly conserved gene, the
Notably, being positively charged, both soluble or fibrillar Aβs specifically form complexes with the plasma membrane CaSRs. Subsequently, the Aβs∙CaSR complexes coalesce into patches which are rapidly endocytosed and can be detected within EEA1-positive early endosomes in the cytoplasm (Figure 1) [54, 55]. However, it has not been ascertained whether Aβs’ binding site[s] is [are] of the orthosteric or allosteric kind or both [22, 44].
Various species of G-proteins mediate CaSR’s intracellular signaling by (
CaSR’s expression occurs in every portion of the rat and human brain. By using the in situ hybridization method, Yano et al.  demonstrated that the CaSR is intensely expressed in several areas of the adult rat CNS. In relation to AD, we recall here that CaSR’s expression abounds in the hippocampus especially at the level of the somata and axon terminals of the pyramidal neurons, suggesting the functional modulation of such cells by CaSR’s signaling [47, 58]. Notably, the N-methyl-D-aspartate receptor (NMDAR) brain location is superimposable on the CaSR’s. Both NMDARs and CaSRs play crucial roles in the induction of long-term potentiation (LTP) . Typically, CaSR’s expression occurs not only in neurons but also in human primary astrocytes, astrocytoma cell lines, oligodendroglia, and microglial cells . Interestingly, total CaSR protein levels increase significantly though transiently in Aβ-exposed NAHAs . Furthermore, the intensity of CaSR’s immunoreactivity significantly increases with age in the hippocampus of 3xTg AD-model mice , particularly where Aβs and p-Taues also accumulate, a clear indication of the involvement of this receptor in AD pathophysiology in vivo (see also below).
The intracellular Ca2+ concentration ([Ca2+]i) can vary widely under both normal and pathological conditions. The Ca2+ influx into cultured astrocytes is linear and normally increases only up to 1.8 mM, suggesting that CaSR signaling controls it [61, 62]. In the past, aberrations of cell surface and intracellular Ca2+-controlling mechanisms were posited to happen in various neurodegenerative ailments, AD included [61, 62, 63, 64, 65]. Reportedly, exogenous Aβ42 and its well-established proxy, Aβ25–35, trigger [Ca2+]i surges and oscillations which persist for hours in the neurons and astrocytes too and concur with the loss of the inner mitochondrial membrane potential. This in turn promotes the release of reactive oxygen species (ROS) and oxidative stress in both neurons and astrocytes. Coculturing such reactive astrocytes with neurons caused the neurons’ death within 24 h unless the Aβ-elicited [Ca2+]i surges were forestalled .
6. Human cortical astrocytes, CaSRs, and AD promotion
It is time for us to zoom in on our preclinical model of cortical untransformed phenotypically stable, i.e.,
As we recalled above, CaSR’s expression takes place with dissimilar intensities, in every CNS cell type, astrocytes included [44, 57]. Recent studies have brought to light some of the physiological roles the CaSR plays in the human CNS, like modulation of neurons’ dendrites and axons growth and of OLGs development [57, 72]. Using the NAHAs as our experimental system, we first demonstrated that exogenous Aβ25–35—instigated CaSR signaling elicits the concurrent expression of nitric oxide synthase-2 (NOS-2) and of GTP cyclohydrolase-1 (GCH-1). GCH-1 makes the BH4 [tetrahydrobiopterin] cofactor that dimerizes and activates the NOS-2 moieties, thus allowing the synthesis of nitric oxide (NO) to occur [44, 73, 74]. Exogenous fibrillar Aβs also induce via direct CaSR signaling activation the cytoplasmic stabilization and nuclear translocation of the hypoxia-inducible HIF-1α•HIF-1β transcription complex in NAHAs. This elicits the vascular endothelial growth factor-A (
However, the most exciting discoveries were subsequently made possible by the advent of very sensitive ELISA kits assaying Aβs. In untreated NAHAs, the metabolic processing of amyloid precursor holoprotein (APP) takes place along the nonamyloidogenic pathway [NAP] being mediated by the activity of the α-secretases (mainly ADAM10) and extracellularly sheds all the soluble sAPPα it produces. Notably, sAPPα is a neurotrophic and neuroprotective compound positively affecting neurons’ functions and viability. Moreover, sAPPα synthesis precludes any Aβ40/42 production from APP as it is cut from the middle amino acid sequence of Aβ40/42. Therefore, NAP largely prevails over APP’s amyloidogenic processing [AP] in the untreated astrocytes, which secrete only very low basal Aβ40/42 amounts . Conversely, adding fibrillar Aβ25–35 by itself and hence stirring off Aβ25–35∙CaSR signaling remarkably reduces sAPPα’s extracellular shedding while driving an overproduction and oversecretion of neurotoxic Aβ42/Aβ42-os owing to concurrent raises in the sequential activities of BACE-1 and γ-secretase. The further addition of a microglial cytokine mixture only accelerates but not increases the total amount of Aβ42/Aβ42-os secretion by the NAHAs despite a concurring APP overexpression [44, 71]. Thus, these events could start of self-sustaining vicious cycle of Aβ42/Aβ42-os spreading within the brain [37, 44]. The same Aβ∙CaSR-induced signaling mechanism stimulates the secretion of neurotoxic Aβ42/Aβ42-os from human cortical postnatal HCN-1A neurons . Thereafter, the neurons start dying slowly like they do in vivo . Most important, we also gained preliminary evidence indicating that Tau and hyperphosphorylated (p)-Tau are both expressed by untreated NAHAs in culture and that their exposure to the usual Aβ42 proxy, Aβ25–35, significantly increases via Aβ∙CaSR-induced signaling the activity of GSK-3β , the main Tau protein kinase [84, 85]. The upshot is an increased production of p-Tau/p-Tau-os which both accumulate inside the cells and are extracellularly released inside exosomes . Novel lines of evidence suggest that extracellular vesicles, which comprise exosomes, play important physiological and pathological roles in the CNS . The above mechanism could promote the concurrent diffusion of both p-Tau/p-Tau-os and Aβ42-os, the two main AD drivers, within the brain, though the tauopathy’s noxious effects will take longer to manifest [37, 83].
Given the relevance of the roles that the several upshots of the Aβ∙CaSR-elicited signaling could have on the promotion of AD, we were enticed to test whether an allosteric highly specific CaSR antagonist [short-termed as
We wish to stress that these results could be gained by using untransformed human cortical adult astrocytes and postnatal neurons,
7. Conclusions and future perspectives
Mounting lines of evidence lend credence to the view that the human astrocytes—the characteristics of which remarkably differ from those of their rodent counterparts—play manifold roles in the molecular mechanisms associated with AD’s pathophysiology. A growing accumulation of Aβs, p-Taues, NO, and VEGF-A hinges upon the signaling of Aβs∙CaSR complexes. This initiates a self-spreading cascade of events which culminate in neuronal synaptic disconnection, dysfunction, and death coupled with the oligodendrocyte dysfunction, axonal myelin sheaths damage and death, the activation of the microglia, the alteration of BBB permeability, and the expression of all the other neuropathological hallmarks of AD. The upshot is a concurrent degeneration of the gray and white matters. Thus, the CNS will keep shrinking; AD clinical symptoms will emerge and increase in intensity until the patients having lost their memories and cognitive abilities die. It is obvious that given their mounting numbers, the care of LOAD patients does heavily impact on their relatives and, for the huge costs of their assistance, on their National Health Services. In this disheartening scenario, the repurposing of highly specific CaSR antagonists or calcilytics as anti-AD therapeutics has the potential for shining a ray of hope.
|aMCI||amnestic minor cognitive impairment|
|APP||Aβ precursor protein|
|BOLD||blood oxygen level-dependent|
|CNS||central nervous system|
|fMRI||functional magnetic resonance imaging|
|GSK-3β||glycogen synthase kinase-3β|
|NAHAs||normal (untransformed) adult human astrocytes|
|p-Tau||hyperphosphorylated Tau protein|
|VEGF||vascular endothelial growth factor|