Calcium Signaling in Schwann cells

Abstract

In addition to providing structural, metabolic and trophic support to neurons, glial cells of the central, peripheral and enteric nervous systems (CNS, PNS, ENS) respond to and regulate neural activity. One of the most well characterized features of this response is an increase of intracellular calcium. Astrocytes at synapses of the CNS, oligodendrocytes along axons of the CNS, enteric glia associated with the cell bodies and axonal varicosities of the ENS, and Schwann cells at the neuromuscular junction (NMJ) and along peripheral nerves of the PNS, all exhibit this response. Recent technical advances have facilitated the imaging of neural activity-dependent calcium responses in large populations of glial cells and thus provided a new tool to evaluate the physiological significance of these responses. This mini-review summarizes the mechanisms and functional role of activity-induced calcium signaling within Schwann cells, including terminal/perisynaptic Schwann cells (TPSCs) at the NMJ and axonal Schwann cells (ASCs) within peripheral nerves.

Introduction

Schwann cells

Glial cells are important regulators of neuronal structure and function in the CNS, PNS and ENS. Whereas astrocytes and oligodendrocytes of the CNS originate from the neural tube, Schwann cells of the PNS and enteric glia of the ENS are derived from neural crest cells that originate from the folds of the neural plate. Each of these glial cell subtypes, despite their different location, are transcriptionally similar (Rao et al., 2015). Formation of mature Schwann cells from ventrally migrating neural crest cells occurs through several intermediate cell populations during embryonic development, including the Schwann cell precursors found in mouse peripheral nerves between embryonic day (E) 12–13, and immature Schwann cells, found between E13-E15 (Jessen and Mirsky, 2005). Each of these populations, as well as the mature Schwann cells they differentiate into, is characterized by distinct molecular profile, trophic dependence on paracrine or autocrine signals, and spatial relationship to the local tissue environment (Jessen and Mirsky, 2005).

Schwann cell precursors and immature Schwann cells become associated with the axons of developing motor and sensory neurons in the ventral and dorsal spinal roots as well as mixed peripheral nerves. Through a variety of positive and negative signaling pathways, immature Schwann cells become one of the two populations of mature Schwann cells present in these nerves, myelinating or unmyelinating/Remak Schwann cells (Jessen and Mirsky, 2008). Myelinating Schwann cells form a 1:1 relationship with the large-caliber axons they surround, whereas unmyelinating Schwann cells ensheath several axons (0.5–1.5 μm diameter) to form Remak bundles. Myelinating Schwann cells facilitate nerve conduction as well as regulate large axon viability, whereas Remak Schwann cells appear to regulate the viability of smaller axons, many of which mediate nociception and autonomic function (Harty et al., 2017). Genetic disruptions to proteins regulating myelin synthesis underlie Charcot Marie Tooth Disease Type II and are characterized by impaired nerve conduction velocity as well as muscle weakness (Suter and Scherer, 2003). A recent study showed that a genetic deficit restricted to Remak Schwann cells also results in neuropathy (McFerrin et al., 2017). Both types of mature Schwann cells exhibit a transformation into a repair Schwann cell in response to nerve injury, allowing peripheral nerve axons the opportunity to successfully re-innervate their targets (Jessen and Mirsky, 2019).

In addition to these two types of axonal Schwann cells (ASCs), an additional subpopulation of non-myelinating Schwann cell known as teloglial cells, terminal Schwann cells or perisynaptic Schwann cells (TPSCs) is found at the peripheral terminals of sensory neurons (e.g., at Pacinian corpuscles) or motor neurons (e.g., the NMJ; Griffin and Thompson, 2006). Whether these are merely non-myelinating Schwann cells that find themselves near nerve terminals, or whether these cells acquire a specific molecular profile through positive or negative signaling pathways, is unknown. However, functional studies have shown that these cells regulate synaptic function (Robitaille, 1998), synaptic maintenance (Reddy et al., 2003; Barik et al., 2016), and the regeneration of distal motor axons in response to peripheral injury (Son et al., 1996), suggesting that these cells may possess a unique molecular identity.

Calcium Signaling

One of the most well characterized signaling events within cells is the change in concentration of intracellular Ca²⁺ [Ca²⁺]_i. Ca²⁺ can bind to a variety of specific proteins and thus influence their conformation and charge. However, the cytoplasmic concentration of Ca²⁺ is maintained at relatively low levels (~100 nM) by Ca²⁺-ATPases and Na⁺/Ca²⁺-exchangers in the plasma membrane as well as Ca²⁺-ATPases in the endoplasmic reticulum (ER), in comparison to its extracellular concentration (1mM). Accordingly, the ability of a cell to modulate this transmembrane Ca²⁺ gradient confers it with the capacity to effect diverse cellular changes in response to extracellular signals (Clapham, 2007). The main intracellular target of Ca²⁺ is calmodulin (calcium-modulated protein; CaM), a highly evolutionarily conserved cytosolic protein containing four Ca²⁺ binding EF-hand domains that interact with several hundred target proteins dependent on or independent of Ca²⁺ binding (Berchtold and Villalobo, 2014). The effects of these interactions include changes in the catalytic activity or subcellular localization of a wide variety of specific effector proteins.

In addition to extracellular Ca²⁺ sources, cells contain several other compartmentalized Ca²⁺ storage sites capable of influencing [Ca²⁺]_i, such as ER and mitochondria. Stimulus-induced influx of Ca²⁺ into the cytosol from either of these extracellular or intracellular sources comprises the basic bipartite cellular Ca²⁺ signaling system. The principal method of Ca²⁺ influx from extracellular sources is via voltage-gated or ligand-gated channels, which are activated by changes in membrane potential or extracellular ligands. Voltage-gated Ca²⁺ channels are composed of catalytic, calcium-conducting α as well as regulatory β subunits and include five different subtypes (L,N,P/Q,R,T). Ligand-gated cation channels that conduct Ca²⁺ include receptors to neurotransmitters such as nicotinic acetylcholine (ACh) receptors, N-methyl d-aspartate receptors to glutamate and purinergic X (P2X) receptors to ATP. In addition to each of these two subtypes, transient receptor potential channels are non-selective cation channels with Ca²⁺ permeability that respond to a wide variety of extracellular and intracellular signals, including Ca²⁺ itself. Due to the large number and diverse regulation of these channels, they have been described as integrators of Ca²⁺ signaling (Hasan and Zhang, 2018). The principal method of Ca²⁺ release from intracellular sources in non-striated muscle cells is through the activation of phospholipase C (PLC) in the plasma membrane through extracellular stimulation of Gαq-coupled metabotropic receptors. PLC catalyzes the hydrolysis of the phospholipid phosphatidylinositol 2,5-bisphosphate into the second messengers inositol 1,4,5-triphosphate (IP₃) and diacylglycerol. IP₃ diffuses through the cytosol to open the IP₃ receptor subtypes 1–3 (IP₃R1–3), which are Ca²⁺ channels that reside in the membrane of the ER. In contrast, the main method of intracellular release in striated muscle is through the opening of another Ca²⁺ channel localized to the specialized muscle ER known as sarcoplasmic reticulum, the ryanodine receptor (RyR). The three isoforms of this receptor (Ryr1–3) are activated by depolarization or influx of extracellular Ca²⁺ in skeletal or cardiac muscle, respectively, but can also be triggered by other stimuli in non-muscle cells (Fill and Copello, 2002). Finally, cells possess a mechanism of intracellular Ca²⁺ level regulation known as store-operated Ca²⁺ entry (SOCE), mediated by the Ca²⁺ sensor and channel proteins Stim and Orai (Putney et al., 2017). In response to depletion of intracellular stores, caused for example by prolonged stimulation-induced release of Ca²⁺ from intracellular stores in conjunction with excessive removal of cytosolic Ca²⁺ from the cell by plasma membrane Ca²⁺ ATPases, the Ca²⁺-sensing protein Stim translocates from the ER to the plasma membrane to activate the Orai store-operated Ca²⁺ channels to allow Ca²⁺ influx.

Chemical dyes that exhibit increases of fluorescence upon binding to free Ca²⁺ were initially developed by Roger Tsien and consisted of a Ca²⁺-chelating moiety together with a fluorescence reporter or fluorophore (Lock et al., 2015). In response to increased [Ca²⁺]_i, quenching of the fluorophore is blocked and fluorescence signal within the cell increases. Membrane-permeant ester forms of these dyes allow for their uptake by cell populations after dye loading; alternatively, dyes may be intracellularly injected for evaluation of Ca²⁺ signaling within individual cells. Dyes with two wavelengths of excitation and emission such as Fura-2 allow quantification of Ca²⁺ levels but suffer from limited temporal resolution, whereas single-wavelength dyes such as Fluo-4 are less quantitative but more amenable for capturing brief changes of [Ca²⁺]_i, also known as Ca²⁺ transient responses or simply Ca²⁺ transients. More recently, the development of genetically-encoded Ca²⁺ indicators (GECIs) has provided a tool to repeatably image Ca²⁺ transients in populations of individual cell subtypes within whole tissues or animals. These tools include the GCaMPs, a family of proteins containing a circularized green fluorescent protein (GFP) fused to CaM at its C-terminus and to M13, a peptide sequence from myosin light chain kinase, at its N-terminus (Nakai et al., 2001). Upon Ca²⁺ binding to the CaM domain of the GCaMPs, CaM binds M13, de-protonating GFP and enhancing its fluorescence.

In addition to the development of GECIs, new imaging technologies have illuminated both the pathways modulating [Ca²⁺]_I as described above, as well as the spatiotemporal nature of subcellular Ca²⁺ events such as sparks or intercellular events such as waves, in ex vivo tissue or in vivo animal preparations. For example, the development of specific confocal imaging platforms permitted the identification of subcellular sparks of Ca²⁺ release from individual clusters of Ca²⁺ channels in the ER of cardiac muscle cells (Cheng and Lederer, 2008). Similarly, the development of multiphoton imaging approaches revealed microdomains of Ca²⁺ signaling within the processes of astrocytes that exhibit temporal and molecular dynamics distinct from those underlying global Ca²⁺ transients in the soma or cell bodies of these cells (Bazargani and Attwell, 2016). Together, these technical developments provide tools to examine the phenomenology, mechanisms and significance of Ca²⁺ signaling in Schwann cells.

Schwann cell Calcium Signaling

TPSC Ca²⁺ Signaling: Mechanisms

An active role for glial cells in neuronal signaling was initiated by the pioneering studies of Kuffler, who described changes to the membrane potential of optic nerve glial cells of the salamander in response to potassium (K⁺) released by active neurons (Orkand et al., 1966). Consistent with this idea, subsequent work showed that astrocytes of the mammalian CNS express a variety of voltage-gated channels as well as receptors to neurotransmitters (MacVicar, 1984; Gallo et al., 1989). Utilizing the recently developed Ca²⁺ chemical dyes described above, several groups showed that the excitatory neurotransmitter glutamate induced an increase of [Ca²⁺]_I in cultured astrocytes that propagated between cells as waves (Cornell-Bell et al., 1989; Ahmed et al., 1990; Jensen and Chiu, 1990). This signal was mediated by different glutamate receptor subtypes as well as both extracellular Ca²⁺ influx and Ca²⁺ release from intracellular stores. These studies helped paved the way for the notion of the synapse as a tripartite entity between pre- and post-synaptic neurons/effectors as well as perisynaptic glia (Pfrieger and Barres, 1996; Araque et al., 1999).

At the NMJ, the processes of TPSCs extend up to, but are largely excluded from, the synaptic cleft between presynaptic motor nerve terminal and postsynaptic muscle endplate, and are thus located in close proximity to neurotransmitters released by motor neurons (Peper et al., 1974), similar to the perisynaptic location of astrocyte processes in the CNS. Seminal work demonstrated activity-induced Ca²⁺ signaling in TPSCs in response to high-frequency stimulation of the peripheral nerve (Reist and Smith, 1992; Jahromi et al., 1992). In these studies, the adult frog cutaneous pectoris muscle was loaded with cell-permeant Fluo-3 and changes of Ca²⁺ were imaged in the presence of the nicotinic AChR blockers curare or α-bungarotoxin (α-BTX) to reduce muscle movement induced by nerve stimulation. Blockade of neurotransmitter release eliminated the nerve stimulation-induced Ca²⁺ transients in TPSCs, and bath application of muscarine and ATP, mimicking the release of the neurotransmitters ACh and ATP by vertebrate motor axons, induced similar responses, even in the absence of extracellular Ca²⁺, suggesting that they were mediated by the release of Ca²⁺ from intracellular stores (Jahromi et al., 1992). Therefore, similar to astrocytes of the CNS, perisynaptic glia of the PNS also exhibit increases of [Ca²⁺]_I in response to neuronal activity-induced release of neurotransmitter.

Subsequent studies carried out in the lab of Robitaille extended these findings to TPSCs of postnatal and adult mice, as well as elucidated the mechanisms underlying these responses. Key findings include the observation that muscarinic and purinergic receptors mediate activity-induced Ca²⁺ responses in adult mouse, although via P1 (adenosine A1) rather than P2 receptors (Rochon et al., 2001), that these responses are mediated exclusively by P2Y₁R purinergic receptors at young postnatal TPSCs (Darabid et al., 2013 and 2018; Heredia et al., 2018a), that the intracellular release of Ca²⁺ from stores is mediated by IP₃R (Castonguay and Robitaille, 2001), that neuromodulators such as Substance P and nitric oxide regulate activity-induced Ca²⁺ transients in TPSCs (Bourque and Robitaille, 1998; Descarries at al., 2001), and that the magnitude of these responses correlates to the level of neurotransmission, in addition to reflecting a degree of intrinsic control, in both normal and diseased conditions (Rousse and Robitaille, 2010; Todd et al., 2010; Arbour et al., 2015).

TPSC Ca²⁺ Signaling: Effects

This ability of individual TPSCs to monitor or decode the magnitude of neurotransmission at individual synapses led Robitaille and his colleagues to test the idea that activity-induced Ca²⁺ transients in TPSCs regulate synaptic function. A key finding in this arena was the demonstration that pharmacological activation within individual TPSCs of G-protein signaling with GTPγS, including Gαq-PLC-IP₃R pathways, led to synaptic depression. Conversely, pharmacological blockade of G-protein signaling in TPSCs with GTPβS reduced the synaptic depression induced by high-frequency stimulation. Together, these findings supported the idea that activity-induced Ca²⁺ transients (ostensibly the most significant component of G-protein signaling affected by the manipulations) in TPSCs regulate synaptic efficacy (Robitaille, 1998). Subsequent studies, using more selective chelating strategies to interfere with [Ca²⁺]_I elevations, extended this finding to include the regulation of synaptic function in a variety of contexts (Castonguay and Robitaille, 2001; Todd et al., 2010; Darabid et al., 2013), including polyneuronal synapse elimination during early postnatal development (Darabid et al., 2018). In this study, it was shown that TPSCs at dually innervated NMJs responded differentially to nerve stimulation of the weaker and stronger nerve inputs. Interestingly, stimulation of the stronger but not weaker input produced a long-lasting synaptic potentiation, and blocking TPSC [Ca²⁺]_I elevations via electroporation and photoactivation of the caged Ca²⁺ chelator Diazo2 or bath application of the P2Y₁R antagonist MRS2179 prevented this form of synaptic plasticity. Additional experiments showed that these effects were dependent on the activation of presynaptic P1 adenosine A2 receptors, suggesting that the enhanced release of Ca²⁺ from intracellular stores in TPSCs induced by activation of the strong nerve input led to the enhanced production, release or activation of adenosine. This study thus identifies ATP/adenosine as an important gliotransmitter released by TPSCs (Araque et al., 2014). Finally, in neonatal mice injected with P2Y₁R antagonist and thus lacking activity-induced Ca²⁺ transients in TPSCs and concomitant synaptic plasticity, synapse elimination was delayed in the soleus muscle, suggesting that that TPSCs are able to differentiate between competing nerve inputs of varying synaptic efficacy and selectively potentiate the them (Darabid et al., 2018). These studies are at partial odds with those of Heredia et al., (2018a), which failed to observe a delay of synapse elimination in the diaphragm muscle of P2ry1 mutant mice lacking P2Y₁R. In each study, TPSC activity-induced Ca²⁺ responses at the neonatal NMJ were shown to completely depend on P2Y₁R. Therefore, synapse elimination in the diaphragm, in contrast to that of the soleus, appears not to depend on these TPSC responses to either discriminate multiple nerve inputs and/or to potentiate the stronger of them. Rather, other molecular mechanisms mediating this process, such as those initiated by postsynaptic action potential generation, may be responsible (Favero et al., 2009). One such possible mediator is thrombin, an activity-dependent retrograde protease capable of destabilizing motor axon branches (Zoubine et al., 1996; Gould et al., 2019). It is also possible that synapse elimination at diaphragm NMJs may normally depend in part on P2Y₁R-mediated TPSC Ca²⁺ responses as it does at soleus NMJs, but congenital elimination of this receptor may have upregulated some of these alternative pathways.

The previous studies capitalized on the remarkable technical ability to simultaneously image Ca²⁺ dye-injected TPSCs, stimulate multiple presynaptic inputs, and record from individual postsynaptic muscle cells, thus permitting a powerful analysis of the effects of modulating TPSC Ca²⁺ signaling on the morphology and function of individual synapses. In contrast, recent studies utilized mice targeting the expression of GECIs such as GCaMP3 to Schwann cells in order to examine the mechanisms and function of activity-induced Ca²⁺ signaling in these cells in an entire muscle (Heredia et al., 2018a–c). Wnt1-Cre transgenic mice, which drive expression in the neural crest, were crossed to conditional GCaMP3-expressing mice. Interestingly, at postnatal ages, nerve stimulation produces robust Ca²⁺ signals in TPSCs but ASCs, despite the fact that each of these cell types express GCaMP3 (Figure 1). Using this approach, it was determined that these responses depend exclusively on P2Y₁R-mediated release of Ca²⁺ from intracellular stores (Heredia et al., 2018a). The ability to generate such data from populations of Schwann cells across the entire diaphragm permitted the evaluation of the functional significance of this purinergic pathway at the level of the whole muscle in addition to the individual synapse. Although no changes in the level of transmitter release were observed in mice lacking P2Y₁R at individual synapses in response to single or high frequency stimulation of the phrenic nerve, enhanced muscle fatigue was observed. This occurred at least in part due to the enhanced susceptibility to the fatigue-enhancing effects of potassium (K⁺); muscles from P2ry1 mutant mice were more sensitive to high K⁺ challenge and TPSCs were less able to respond to bath-applied K⁺. These studies suggest that activity-dependent Ca²⁺ transients within TPSCs may regulate K⁺ uptake at the NMJ by these cells, similar to Bergmann glia in the cerebellum (Wang et al., 2012). Consistent with this idea, TPSCs depolarize in response to bath application of K⁺ (Heredia et al., 2018a). However, previous studies failed to show an effect on TPSC membrane potential in adult frog cutaneous pectoris in response to nerve stimulation (Robitaille, 1998).

Figure 1.

Activity-induced Ca²⁺ responses in TPSCs of the early postnatal diaphragm. Postnatal day 7 (P7) diaphragm muscle of Wnt1-Cre; conditional GCaMP3 mice was imaged before (Pre-stim) and after (40s, 45s Nerve Stim) phrenic nerve stimulation in the presence of the myosin inhibitor BHC to block movement. Top right and bottom left panels are low and high power spatial maps of fluorescence intensity of TPSCs generated in response to nerve stimulation. Bottom right panel depicts regions of interest, representing individual TPSCs, selected for analysis. Adapted from Heredia et al., 2018a.

These studies thus point to a variety of signaling mechanisms and consequences of nerve activity-induced TPSC Ca²⁺ signaling. However, significant issues remain to be addressed. First, activity-induced TPSC responses were never observed past P20 in Wnt1-Cre, conditional GCaMP3 mice. Although the reasons for this are unclear, we recently were able to detect robust activity-induced Ca²⁺ responses in TPSCs of the diaphragm at P80 in Sox10-Cre, conditional GCaMP6f mice. Interestingly, these responses were also dependent on P2Y₁R signaling (Figure 2). Second, although assumed it has not yet been shown that TPSCs exhibit endogenous activity-induced Ca²⁺ transients. This is due in large part to the difficulty of imaging TPSCs of muscle receiving intact, active projections from motor neurons in the spinal cord. Live imaging studies of fluorescent TPSCs in sternomastoid muscle of anesthetized mice have been performed (Kang et al., 2014), so adoption of similar strategies together with a protocol to reduce muscle movement, independently of manipulating transmitter release, seem possible. Third, it is not clear if TPSCs possess microdomains of Ca²⁺ elevations and if so what these spatial domains are and whether they depend on similar or different signaling mechanisms. In order to begin to address this, we imaged activity-dependent Ca²⁺ transients in TPSCs of mice expressing membrane-targeted GCaMP6f. Although these responses were only imaged using standard fluorescence microscopy at low power, they appeared throughout the TPSC and were exclusively dependent on P2Y₁R signaling (Figure 3). Relatedly, it is not clear if stimuli other than somatic nerve activity regulate TPSC Ca²⁺ signaling. Unlike astrocytes in the CNS, TPSCs are cells with relatively restricted spatial connections at the synapse, without processes that extend to blood vessels or other structures. However, it remains possible that signals between these cells and the vasculature are shared, as enhanced skeletal muscle activity triggered by somatic motor nerve stimulation induces functional hyperemia (Thomas and Segal, 2004; Hong and Kim, 2017), and a spatial relationship exists between the arterioles and motor innervation of the diaphragm (Correa and Segal, 2012). Another unresolved is whether TPSCs can propagate activity-induced Ca²⁺ signals to adjacent TPSCs or to ASCs through gap junctions or other mechanisms as waves. These studies will require a combination of approaches, such as mice expressing GECIs in all Schwann cells together with microinjection or electroporation of caged agonists or antagonists. Another important unresolved issue is whether TPSCs and more specifically TPSC Ca²⁺ signaling contributes to neuromuscular dysfunction in motor neuron diseases such as amyotrophic lateral sclerosis or spinal muscular atrophy, which display evidence of altered motor neuron excitability and function prior to structural deficits. Along these lines, it is of note that Arbour et al. (2015) found that the Ca²⁺ response to neurotransmitter release within TPSCs is itself altered pre-symptomatically in a mouse model of ALS. Specifically, the sensitivity of this response to the activation of muscarinic AChRs was increased. Finally, despite the important role played by activity and TPSCs in regulating distal re-innervation after peripheral nerve injury, it is not clear if activity-induced TPSC Ca²⁺ signaling participates.

Figure 2.

Activity-induced Ca²⁺ _responses in TPSCs of the adult diaphragm. Diaphragm muscle of P80 Sox10-Cre, conditional GCaMP6f mice was imaged in response to high-frequency nerve stimulation (40 Hz, 45s) of the phrenic nerve in the presence of the myosin inhibitor BHC to block movement. Left two images (A,B) and middle two images (C,D) are spatial maps of fluorescence intensity and regions of interest generated in response to stimulation in the absence or presence of the P2Y₁R antagonist MRS2500; far-right image (E) is α-BTX-labeled nicotinic AChR clusters simultaneously imaged with a Gemini Splitter according to Heredia et al., 2018c.

Figure 3.

Activity-induced Ca²⁺ responses in the membrane of TPSCs are mediated by P2Y₁R. Diaphragm muscle of P7 Sox10-Cre; conditional LCK-GCaMP6f (membrane-targeted GCaMP6f) mice was imaged in response to high-frequency stimulation (40 Hz, 45s) of the phrenic nerve in the presence of the myosin inhibitor BHC to block movement. A and B are spatial maps of fluorescence intensity generated in response to stimulation in the absence or presence of the P2Y₁R antagonist MRS2500, respectively. Bottom graph is a comparison of background-subtracted fluorescence responses. Arrow indicates onset of nerve stimulation.

ASCs: Mechanism and Effects

The initial studies of activity-induced Ca²⁺ transients in TPSCs failed to detect them in myelinating Schwann cells of preterminal axons and distal nerve branches. However, Lev-Ram and Ellisman (1995), using the Ca²⁺ dye Fluo-3, reported an increase of [Ca²⁺]_I near the Nodes of Ranvier of myelinating Schwann cells of the frog sciatic nerve that was EGTA-sensitive and thus dependent on extracellular Ca²⁺ influx. The change in fluorescence was considerably less than that in TPSCs and required high-power imaging of the nerve in the absence of dye-loaded muscle, which may explain the failure of these earlier studies to observe it. Although the nature of the conductance mediating Ca²⁺ influx was not identified, these authors provided support for the hypothesis that it induced Ca²⁺ release from internal stores via RyR.

In contrast to this dependence of myelinating ASC Ca²⁺ transients on extracellular sources, several examinations of Ca²⁺ signaling with myelinating and non-myelinating ASCs demonstrated an ATP-dependent mechanism. For example, bath application of ATP onto intact nerve induced elevations of [Ca²⁺]_I mediated by P2Y₁R and P2Y₂R and release from ER stores in non-myelinating and myelinating ASCs, respectively (Mayer at al., 1998). Consistent with these findings, nerve activity triggered Ca²⁺ transients in an ATP-dependent manner in mouse sciatic nerve-derived pre-myelinating ASCs cultured with dorsal root ganglia (DRG) neurons (Stevens and Fields, 2000). This pathway reduced the proliferation of these cells, suggesting that activity-induced, ATP-mediated Ca²⁺ signaling regulates ASC development. In agreement with this data generated with chemical Ca²⁺ dyes, electroporation of GECIs into ASCs of the early postnatal mouse sciatic nerve demonstrated activity-induced Ca²⁺ transients in the cytosol and mitochondria of myelinating ASCs that were dependent on P2Y₂R and the mitochondrial Ca²⁺ uniporter, respectively (Ino et al., 2015). Interestingly, reducing expression of either of these proteins resulted in hypomyelination of sciatic nerve axons, consistent with a role for this pathway in balancing Schwann cell proliferation vs. differentiation. Further evidence for this idea comes from the recent finding that ATP-induced, P2Y₂R-mediated signaling suppresses SC proliferation, although these effects are mediated by β-arrestin-mediated rather than PLC-mediated signaling (Coover et al., 2018).

In addition to these studies showing the dependence of activity-induced Ca²⁺ transients in ASCs on external influx or internal release, a recent report provided evidence for a TRP channel-mediated store-operated Ca²⁺ conductance in myelinating Schwann cells (Vanoye et al., 2019). Interestingly, this conductance was augmented in Schwann cells expressing peripheral myelin protein 22 mutants underlying CMT1A, suggesting a potential relationship between ASC Ca²⁺ signaling and this inherited neuropathy. Together, these studies highlight several pathways by which activity-induced Ca²⁺ transients in Schwann cells regulate key aspects of Schwann cell development and function.

Conclusion

Recent developments in the tools used to bind and image intracellular Ca²⁺ have greatly enhanced the empirical evaluation of nerve activity-induced Ca²⁺ transients in Schwann cells at the NMJ and along axons (Figure 4). Key studies have demonstrated a variety of mechanisms underlying these responses in TPSCs as well as myelinating and non-myelinating ASCs, and in some cases have examined the functional significance of these responses in genetic mutant mice deficient in the underlying mechanisms. Despite the relatively small number of stimuli released by peripheral nerves in the vicinity of Schwann cells (ACh, ATP, K⁺), compared to CNS neurons onto astrocytes or enteric neurons onto enteric glia, these responses exhibit distinct mechanisms depending on Schwann cell subtype, age, or species. Future studies combining advanced imaging methods together with animal models of peripheral neuromuscular injury or disease will doubtlessly yield additional mechanisms and effects of this important neuroglial interaction.

Figure 4.

Summary of the mechanisms and effects of activity-induced Ca²⁺ signaling in TPSCs at the NMJ and ASCs along peripheral nerves. A, At the NMJ between motor nerve terminal (green) and motor endplate of skeletal muscle fiber (orange), the TPSC (yellow) contains purinergic P2Y₁ and muscarinic receptors that transduce ATP and ACh released by active motor neurons. These receptors result in the production of the second messenger IP₃, which triggers the release of Ca²⁺ from the endoplasmic reticulum (ER; blue semicircle) to regulate synaptic plasticity, synapse elimination and muscle fatigue. B, Along peripheral sensory and motor axons, ATP released from active axons through pannexins or volume-activated channels (purple) stimulate purinergic P2Y₂ receptors, which triggers IP₃-mediated release of Ca²⁺ from ER into the cytosol and into the mitochondria via the mitochondrial uniporter (pink). This purinergic pathway downregulates ASC proliferation and upregulates ASC myelination of peripheral axons. A second nerve-derived signal of unknown origin (??) stimulates the influx of extracellular Ca²⁺ into ASCs via unknown Ca²⁺ or non-selective cation channels (orange bars).

Highlights.

This article highlights the various approaches taken (e.g., chemical dye, genetically-encoded indicators) to image calcium signaling in Schwann cells. It is actually the only review that tries to capture everything about calcium signaling in Schwann cells, both axonal and terminal/perisynaptic.