astrocytic calcium ion surges and tDCS

Posted comment on ´Calcium imaging reveals glial involvement in transcranial direct current stimulation-induced plasticity in mouse brain` by H. Monai, M. Ohkura, M. Tanaka, Y. Oe, A. Konno, H. Hirai, K. Mikoshiba, S. Itohara, J. Nakai, Y. Iwai and H. Hirase and published on 22nd March 2016 in Nature Communications 7:11100 doi: 10.1038/ncomms11100 (2016).


In their article, Monai and colleagues discuss a possible neurochemical mechanism involved in transcranial direct current stimulation (tDCS), which has been described for example as helpful in alleviating depression and enhancing learning. Work by others using in vitro brain slices has shown that NMDA receptors are likely to play a role in the mechanism, which in general is unclear. One possibility put forward by Monai`s group and others is that astrocytes are involved in the NMDA receptor plasticity and the mechanism includes activation of the astrocytes resulting in rising intracellular calcium levels which may lead to the secretion of signaling molecules into the synapse. This release eventually leads to glutamate receptor plasticity at the post-synaptic membrane. Hence, Monai and colleagues investigated whether tDCS produces its therapeutic effects by causing astrocytic activation.

In their study, Monai and team used transgenic mice (mouse line G7NG817) which expressed G-CaMP7 (a green fluorescent protein Ca2+ indicator protein) based in the astrocytes and a subpopulation of excitatory neurons. They found high level of expression of the fluorescent protein in the astrocytes of the cortex, hippocampus (particularly the CA3 region), thalamus and striatum. Expression of G-CaMP7 was also observed in the neurons, but not in the cortical GABA cells. A level of tDCS of 0.1 mA direct current for 10 min induced large amplitude calcium ion surges in the astrocytes of the cortex. These surges exhibited higher amplitudes than spontaneous calcium ion events, but led to no changes in local field potential. There was also no change in local travelling wave propagation. Similar results were obtained for the anodal, contra-anodal and distal regions of the cortex. Long calcium ion surges were found to be more frequent during tDCS. The amplitude changes observed were the same for awake and anaesthetized mice, but in the latter onset of the surge was found to be more variable. In awake mice, the surges began several seconds after the tDCS onset whereas the calcium ion surges were found to be of a lower frequency in anaesthetized mice.

Monai and colleagues found that the t-DCS induced calcium surges in awake mice could be blocked by the administration of an alpha 1 adrenergic receptor inhibitor, prazosin, or by destroying the noradrenergic innervation with DSP-4. The surges were also blocked in vivo by local application of prazosin. Increases in astrocytic calcium ions were also not observed in IP3R2 (inositol triphosphate receptor type 2) knock-out mice and therefore, it was concluded that astrocytic GPCR activation is the prevalent mechanism of tDCS-induced Ca2+ surges. This view was further supported by behavioural investigations. Transcranial DCS was sufficient to alleviate a mouse model of depression by chronic restraint stress, but could not produce the same therapeutic effect after prazosin administration, DSP-4 treatment or IP3R2 deficiency.

To prove that the calcium surges were linked to astrocytes and not neurons, Monai and colleagues used two-photon imaging in layer 2/3 of the primary visual cortex. These experiments showed that the calcium surges were linked to SR101-positive astrocytes. Transcranial DCS evoked astrocytic calcium ion responses which had significantly higher amplitudes than spontaneous events, whereas the neuronal calcium ion events during tDCS had similar amplitudes to spontaneous events. The astrocytic calcium surges also occurred nearly seven times more frequently during the course of tDCS (10 min) than during the baseline, while neuronal activity did not show any obvious changes. The authors also used transgenic mice to confirm that the astrocytes were responsible for the calcium surge. They used transgenic mice with expressed G-CAMP7 in neurons or astrocytes using cell-type-specific recombinant adeno-associated viruses (AAV2.1-hSyn1-G-CaMP7 and AAV9-hGFAP-G-CaMP7, respectively) in C57BL/6 mice. Astrocytic soma, but not neuronal soma, were found to give rise to the long-lasting calcium ion surges associated with tDCS.

Monai and colleagues also investigated the role of NMDA receptors in calcium surges. They looked at visual evoked potential (VEP) of primary visual cortex of anaesthetized C57BL/6 mice after a flash stimulation of 60 secs before and after tDCS. It was found that the VEP slope increased by 50% after tDCS and remained at this increased level for at least 2hours after application. The effect was blocked by AP-5, an NMDA receptor antagonist and topical application of the alpha 1 adrenergic receptor antagonist, prazosin, but was not affected by the application of the muscarinic receptor, atropine. This indicated that NMDA receptors and alpha 1 adrenergic receptors play roles in the calcium surges induced with tDCS. There was also no VEP slope enhancement in IP3R2 knock out mice. Therefore, it was concluded that astrocytic calcium ion rises were involved in the tDCS enhancement of the VEP effect.

Therefore, Monai and colleagues concluded that tDCS induced astrocytic activity brings about plasticity changes in the cortex through calcium ion and Ip3 signalling. They showed that the tDCS-induced enhancement of a sensory evoked response is NMDAR dependent and as astrocyte calcium ion levels are positively related to the extracellular level of the NMDAR co-agonist d-serine, tDCS-induced astrocytic Ca2+ elevations possibly lead to NMDAR-dependent synaptic plasticity. The team also concluded that tDCS induced plasticity could be blocked by prazosin or DSP-4 treatment, thus indicating an involvement of alpha 1 adrenergic receptors which induce the G signaling cascade for IP3 production. The authors quote in their conclusion supporting work by Panktratov and Lalo who showed that application of noradrenaline raised extracellular d-serine and ATP levels and lowers the threshold for LTP induced plasticity in mouse cortical slices. They also conclude that the activation of A1AR is the prevalent mechanism for astrocytic calcium ion elevation in awake mice and suggested that the tDCS induced noradrenergic drive includes activation of the locus coeruleus and/or direct induction of transmitter release from noradrenegeric axon terminals in the cortex.


What makes this article interesting is that glial cells are again found to be more than just support cells for the all-important neuron and neuronal synapse. We know that different types of glial cells are important for different functions, eg. microglia are important in the degeneration of neuronal cells and oligodendrocytes are important in myelin production. This article focuses on another type of glial cell, the astrocyte, and expands its function in the action potential recovery phase and later on in the development of neuronal plasticity. The authors in their study showed that not only do astrocytes ´mop up` excess released neurotransmitters, binding of these neurotransmitters can cause intracellular astrocytic calcium surges that have an influence ultimately on neuronal NMDA dependent plasticity. This action can occur even spontaneously and hence, the excitable nature of the astrocytic glial cell provides yet another aspect to neuronal function that could go dysfunction and evoke deleterious changes in the neuronal and synaptic area. This additional role of the astrocyte has provoked lots of discussion simply because of the discrepancies in reported results, which have primarily arisen from experimenters using different brain areas, in vivo and in vitro samples and various experimental conditions.

Monai and colleagues have shown in their experiments that astrocytes produce internal calcium ion surges on binding of neurotransmitters to G protein linked receptors on the glial cell membrane. Using this finding, their aim was to show that these same calcium surges and resulting events could be produced by applying transcranial direct current stimulation (tDCS) and hence, one mechanism (or even, the mechanism) that is involved in the tDCS effect would be elucidated. Their finding supports observations about tDCS that it is effective in alleviating neuropsychiatric and neurological conditions such as depression in humans, and enhances learning and memory formation. Previous work on cortical slices shows that tDCS increases the excitability of the motor cortex in a NMDA receptor dependent manner, but the mechanisms involved in vivo are largely unknown with the exception that astrocytic calcium ion/IP3 signalling appears to play a significant role in synaptic plasticity in the cortex and hippocampus. Monai and colleagues` investigation confirmed this and showed that the tDCS induced plasticity was NMDA receptor dependent. They also found that the mechanism involved alpha 1 noradrenergic receptors which they believed transduced the G protein signaling cascade for IP3 production and was linked to the resulting calcium surges. This noradrenergic drive involved activation of the locus coeruleus area and/or direct induction of transmitter release from noradrenergic axon terminals in the cortex.

However, since there are discrepancies between the findings relating to astrocytic action it is not possible to say definitively what is going on in the case of tDCS. So, what do we know about astrocytic function? We know that astrocytes are one type of glial cell with the others being oligodendrites for myelination, microglia (shape shifters) as scavengers removing dead and damaged tissue from the nervous system, nerve/glial antigen 2 (NG2)-positive glia, which include oligodendrocyte and astrocyte progenitor cells as well as NG2+ cells that persist in the mature brain and astrocytes. It is likely that astrocytic cells share many of the characteristics and mechanisms of neurons. In fact, Liu showed that a single transcriptions factor, ACII, can convert astrocytes into functional neurons. Astrocytes are part of the action potential/firing stage of an active cell networking system and they can influence whether a neurite grows or retracts and regulate the content of extracellular space eg. they surround the synapse, remove excess neurotransmitter and control potassium ion concentration after the action potential has occurred, or in times of neuronal stress.

Astrocytes are capable of such actions by their physiology. The most common type are protoplasmic astrocytes and these cells have a very complex morphology and contact most, if not all, other cell types in the brain. The cells form from their soma (diameter 7-9 μm) elaborate and dense, fine non-overlapping processes that interact closely with the synapses present and supporting blood vessels (greater than 99% of the cerebrovascular surface is sheathed by astrocyte processes). It has been said that processes from a single astrocyte can envelop approximately 140,000 synapses which means that one astrocytic cell can occupy a ´working` volume of approximately 66,000 cubic μm. The cells are linked with each other by gap junctions and patch clamping experiments with a gap-junction permeable dye show that a single astrocyte rapidly leads to the filling of hundreds, even thousands of other astrocytes. Therefore, astrocytes likely function as a ´syncytium` contacting essentially all other cellular elements in the brain, including neurons, oligodendrocytes, NG2+ cells, microglia, and blood vessels.

In addition to this abundabt communication, there is also diversity within individual astrocytes with respect to interactions with the local environment. For example, it is possible that within a single astrocyte, a subset of processes (microdomains) can interact autonomously with neuronal synapses within its immediate environment while other regions of the same astrocyte interact with different groups of synapses or with other synaptic and neuronal elements such as blood vessels. These microdomains may not communicate with each other which implies that one astrocytic cell can carry out multiple functions simultaneously. However, this is unclear at this time and requires further investigation.

Known astrocytic functions appear to be linked to two mechanisms. The first is that described by Monai and colleagues that of glial cell surface receptor binding and calcium surges. Astrocytes are assumed to be like other glial cell types in that they have many signaling proteins similar to those found in neurons, eg. ion channels and receptors such as those for glutamate, GABA and noradrenaline. These specific neurotransmitter receptors are on the cell membranes and can trigger events within the glial cells and Monai and colleagues reported this in the case of noradrenergic receptors. This finding confirmed work from the mid 1990s which showed that activation of G protein-coupled receptors on the astrocytic cell membrane surface by synaptically released neurotransmitters produces rises in intracellular calcium concentration. An IP3 signalling cascade is involved and this demonstrates not only that astrocytes display a form of excitability like the neuron, but also that astrocytes may be active participants in brain information processing. The ultimate consequence of such a binding is the increase in NMDA receptor plasticity on the neuronal post-synaptic membrane which will affect overall neuronal area functioning.

Although the authors and others report a link to NMDA receptor plasticity, some researchers (eg. Goldman) report instead an increase in AMPA receptor trafficking and increased plasticity of the neuronal area. Goldman found that grafting human cells onto mice cells led to a 4 times increase in synaptic activity. Han and colleagues, as reported in this blog`s post of January 2015, supports this observation since they found that grafting of human glial progenitor cells in the mouse forebrain led to increased synaptic plasticity and learning in the adult linked to glial cell increases.  An increase in human astrocytes was observed at 4-5 months in the hippocampus and deep neocortex layers, but by 12-20 months it was also observed in other areas such as the amygdala, thalamus, neostriatum and cortex. An investigation into synaptic activity in neuronal cells from the hippocampal dentate granule layer, an area used because of its large number of engrafted cells and the region´s known role in spatial memory, found a significant increase in the engrafted human glia cell`s  basal level of excitatory synaptic transmission. This long term potentiation (LTP) enhancement was not linked to increased NMDA receptor activity (or increased glutamate release), altered adenosine concentrations or, changed D-serine release but instead to increased TNF alpha which induces the addition of AMPA receptors to neuronal membranes and AMPA GluR1. The insertion is regulated through protein kinase C (PKC)-mediated phosphorylation of appropriate sites. Such increased neuronal activity was mirrored by enhanced learning in the chimeric mice with increased spatial memory and quicker contextual fear and tone conditioning. This observation was supported by Hennessey and colleagues who showed that astrocytes in degenerating brains caused by acute sterile inflammatory insult are primed to produce exaggerated responses via strong nucleur localization of NK-kB subunit p65 and increased synthesis of the chemokines, CXCL1 and CCL2. Administration of IL-1beta and TNF-alpha produced a more robust response in degenerating rat brain than the control.

Another function of astrocytes is potassium spatial buffering which is where extracellular potassium ions occuring during stimulation are taken up. The potassium ions enter through potassium channels causing the astrocyte to depolarize. The specialized inwardly rectifying potassium channels involved in this are also known to be linked to GPC receptors and are ATP sensitive. Potassium ion entry increases the internal concentration which is dissipated over a large area by the extensive network of astrocyte processes. There is no evidence that these GPC receptors are linked to the calcium surges that induce higher sensitivity to neuronal firing, but if they are this could mean that the internal calcium ion surge may not be caused just by neurotransmitter binding on the astrocytic surface, but also by the inward current of potassium ions released by presynaptic cells on neuronal activation.

Therefore, astrocytes appear to be stimulated by the activation and release of neurotransmitters from the activated presynaptic neuronal cell, or possibly by the internalization of potassium ions excluded during firing. These actions affect G protein receptors activities and cause internal calcium surges which presumably activate protein kinases, phosphorylate relevant proteins and cause specific gene transcription changes just like in neuronal cells. It is possible that the result of this is the release of gliotransmitters (signal molecules that could be noradrenaline, glutamate or GABA) from the glial cells which will then bind to the relevant receptors on the post-synaptic membrane. The overall result of this is that there is possibly a ´second wave` stimulation of the post-synaptic membrane, the first being the direct binding of neurotransmitters released pre-synaptically directly on firing. This second wave is linked to AMPA receptor addition to the post-synaptic membrane which is associated with LTP. The implication of this ´second wave` response is the slight delay observed in post-synaptic effects. Therefore, the post-synaptic neuronal response is augmented and temporally extended by the astrocytic response to both neurotransmitter and potassium ion presence in the synaptic cleft. Using this hypothesis, we can suggest that if tDCS can cause astrocytic calcium ion surges, then it could work by ultimately causing effects that are normally associated with neurotransmitter firing triggered by other means. The excess of electrons administered by the direct current stimulation can cause changes in presynaptic membrane electron fields which can result in the release of neurotransmitters from the presynaptic neuronal cells and/or firing of the cell. The neurotransmitters released or the excluded potassium ions can then bind to the astrocytic G protein linked receptors and cause the effects described above. In this way, the ultimate result is AMPA receptor addition to the neurons and induced LTP. This changed plasticity will present as the changes in learning and depression suggested as being associated with tDCS.

Therefore, the role of astrocytes in the neuronal firing scenario and neuronal plasticity appears to be important and can possibly be manipulated by applying direct current. This may provide a mechanism by which neuronal areas understimulated or defective in normal stimulation can be induced to fire, but it may be detrimental in areas where hyperexcitability is being reported eg. in the hippocampus and entorhinal areas as recorded in dementia. Therefore, more research is needed to investigate this astrocytic effect.

Since we`re talking about the topic…………………………………..

………would using radioactively marked AMPA receptors in the hippocampal area of transgenic mice show that tDCS is linked to AMPA receptor insertion into the cell membrane?

……….can we assume that removal of astrocytic cellular calcium ions with EDTA will have an effect on neurotransmitter release?


This entry was posted in astrocytes, calcium ions, glutamate receptors, tDCS, Uncategorized and tagged , , , . Bookmark the permalink.