Targeting microglia L-type voltage-dependent calcium channels for the treatment of central nervous system disorders

Abstract

Calcium (Ca²⁺) is a ubiquitous mediator of a multitude of cellular functions in the central nervous system (CNS). Intracellular Ca²⁺ is tightly regulated by cells, including entry via plasma membrane Ca²⁺ permeable channels. Of specific interest for this review are L-type voltage-dependent Ca²⁺ channels (L-VDCCs), due to their pleiotropic role in several CNS disorders. Currently, there are numerous approved drugs that target L-VDCCs, including dihydropyridines. These drugs are safe and effective for the treatment of humans with cardiovascular disease and may also confer neuroprotection. Here, we review the potential of L-VDCCs as a target for the treatment of CNS disorders with a focus on microglia L-VDCCs. Microglia, the resident immune cells of the brain, have attracted recent attention for their emerging inflammatory role in several CNS diseases. Intracellular Ca²⁺ regulates microglia transition from a resting quiescent state to an “activated” immune-effector state and is thus a valuable target for manipulation of microglia phenotype. We will review the literature on L-VDCC expression and function in the CNS and on microglia in vitro and in vivo and explore the therapeutic landscape of L-VDCC-targeting agents at present and future challenges in the context of Alzheimer’s disease, Parkinson’s disease, Huntington’s disease, neuropsychiatric diseases, and other CNS disorders.

1 |. INTRODUCTION

Regulation of intracellular calcium (Ca²⁺) levels in cells of the central nervous system (CNS) is critical for the maintenance of homeostasis. Intracellular Ca²⁺ is a second messenger for numerous CNS processes and is a vital regulator of canonical neuron functions such as action potential firing and neurotransmitter release. Intracellular Ca²⁺ also controls essential functions in microglia, the immune cells of the CNS. Intracellular Ca²⁺ regulates microglia transformation from a quiescent, homeostatic resting state to an activated immune-effector state. Ca²⁺ orchestrates several important functions including phagocytosis, proliferation, migration generation of neurotrophic factors, cytokine production and release, and formation of reactive oxygen species (ROS) (Brawek & Garaschuk, 2013; Sharma & Ping, 2014). Given the emerging critical role of dysregulated and dysfunctional microglia in numerous neurological conditions including Alzheimer’s disease (AD), Parkinson’s disease (PD), Huntington’s disease (HD), schizophrenia, and autism spectrum disorder (ASD), there is an increasing need to identify potential treatments for CNS disorders that target microglia. Ca²⁺ channels offer a potential target for altering microglia phenotype in these and other disorders. Recently, the United States National Institutes of Health has encouraged repurposing of existing therapeutics for treatment in disorders with high unmet therapeutic needs. Several existing Ca²⁺-channel blocker (CCB) drugs are approved, safe, and well-tolerated in the clinic and are prime targets for repurposing. Of these CCBs, dihydropyridine L-type voltage-dependent Ca²⁺ channel (L-VDCC) antagonists (e.g., nimodipine, isradipine, nicardipine, nifedipine) have demonstrated their neuroprotective utility in several CNS disorders and for their anti-inflammatory action on microglia.

2 |. REGULATION OF INTRACELLULAR CA²⁺

Intracellular Ca²⁺ is a ubiquitous second messenger in numerous signal transduction pathways throughout the body. Extracellular Ca²⁺ levels are 20,000-fold higher than cytosolic Ca²⁺ levels, creating a concentration gradient that is essential for the induction of numerous Ca²⁺-dependent signaling cascades; a large amount of cellular energy is devoted to maintenance of this gradient using the plasma membrane Ca²⁺ ATPase (PMCA) transporter and Na⁺/Ca²⁺ exchanger (NCX) (Clapham, 2007). Cells utilize cytosolic Ca²⁺ to directly modulate signaling cascades involving Ca²⁺-dependent kinases and phosphatases (e.g., calcineurin), or indirectly via the Ca²⁺ binding protein calmodulin (CaM) (Bootman, 2012) and these pathways can feed back onto Ca²⁺ channels to regulate intracellular Ca²⁺ levels (Ben-Johny & Yue, 2014). When cytosolic Ca²⁺ levels rise, Ca²⁺ is rapidly sequestered by buffers and Ca²⁺-binding proteins as well as organelles including the endoplasmic reticulum (ER), mitochondria, and lysosomes (McBrayer & Nixon, 2013; Toescu & Verkhratsky, 2003).

There are two main sources of cytosolic Ca²⁺: Ca²⁺ released from intracellular organelle stores and Ca²⁺ entry from the extracellular space. Opening of Ca²⁺ channels on organelles or the plasma membrane results in a rapid, local 100- to 1000-fold rise in Ca²⁺ in the immediate microenvironment; coordinated channel opening can be used to induce global elevation of Ca²⁺ throughout the cell or across populations of cells (Berridge, Bootman, & Lipp, 1998). Ca²⁺-induced Ca²⁺ release (CICR) uses rises in intracellular Ca²⁺ to trigger the release of additional Ca²⁺ from intracellular stores (largely the ER) into the cytosol via activation of ryanodine receptors (RyRs) or of inositol trisphosphate (IP3) receptors (Berridge, 1993; Thibault, Gant, & Landfield, 2007). The depletion of ER Ca²⁺, in turn, leads to capacitive Ca²⁺ influx of extracellular Ca²⁺ in a process termed store-operated Ca²⁺ entry (SOCE) via plasma membrane Ca²⁺-release-activated Ca²⁺ (CRAC) channel made up of Orai1 and Orai2; L-VDCCs also contribute to SOCE (Park, Shcheglovitov, & Dolmetsch, 2010; Wang et al., 2010). ER proteins stromal interaction molecule (STIM) 1 and 2 translocate to the plasma membrane to activate CRAC channels to refill depleted ER Ca²⁺ stores via sarcoplasmic reticulum (SR)-ER calcium-ATPase (SERCA) pumps (Soboloff, Spassova, Dziadek, & Gill, 2006) and also interacts with L-VDCCs to negatively regulate their function during SOCE (Park et al., 2010; Wang et al., 2010). Notably, in non-excitable cells such as lymphocytes, high levels of STIM1 inhibit L-VDCCs and Orai1 activity dominates, while in excitable cells such as neurons with relatively lower amounts of STIM1, L-VDCC activity drives this process (Cahalan, 2010).

3 |. CA²⁺ SIGNALING IN MICROGLIA

Microglia are the innate immune cells of the CNS. Under normal conditions, “resting” microglia play an important homeostatic role, dynamically surveying their environment and performing tissue maintenance (Sierra, Tremblay, & Wake, 2014). Microglia can rapidly respond to injury and pathogenic threats by shifting toward and “activated” phenotype where they release cytokines and ROS. Sustained microglia activation results in neurotoxicity (Figure 1). Dysregulation of microglia-mediated neuroinflammation is an emerging component of numerous brain disorders, including AD (Cameron & Landreth, 2010), PD (Tansey & Goldberg, 2010), HD (Möller, 2010), and neuropsychiatric disorders (Tay et al., 2017).

FIGURE 1.

Dual role of microglia in CNS disorders. Microglia provide important homeostatic functions but can also mediate neurotoxicity and neurodegeneration and play a role in Alzheimer’s disease (AD), Parkinson’s disease (PD), Huntington’s disease (HD), and neuropsychiatric (NP) illness. Resting microglia contribute to neuroprotection and homeostasis via constant surveillance of the microenvironment, release of neurotrophic factors, and beneficial tuning of synapses. Resting microglia have few calcium (Ca²⁺) transients. Resting microglia can rapidly respond to an acute stimulus to become activated. Acute stimuli include short-term exposure to aggregated protein fibrils or oligomers associated with AD and PD (Aβ, tau, synuclein), toxins (e.g., MPTP), or acute stress. Genetic and environmental risk factors can alter microglia response to these stimuli. Acutely activated microglia increase their phagocytic capacity and briefly upregulate cytokines to resolve threats and clear protein and debris concurrently with increase of intracellular Ca²⁺ and transients. Resolution of acute neuroinflammation results overall in neuroprotection. L-type voltage-dependent Ca²⁺ channels (L-VDCCs) can reduce acute microglia activation in vitro and in vivo. Chronic stimuli, such as long-term exposure to protein aggregates (amyloid β [Aβ] plaques, tau tangles, Lewy bodies, mutant huntingtin [mHTT] inclusions) or stress and trauma, can chronically activate microglia leading to neurotoxic secretion of reactive oxygen species (ROS), cytokines, and chemokines as well as synaptotoxicity, engulfment of neurons, and seeding of toxic protein aggregates. Degenerating neurons provide positive feedback during chronic neuroinflammation, perpetuating a cycle of neurotoxicity and neurodegeneration. Notably, chronic neuroinflammation is difficult to resolve due to this toxic cycle, as chronically activated microglia become less responsive to stimuli. This decrease in responsiveness occurs at the molecular level, with increases in Ca²⁺ transients but reduced Ca²⁺ response to stimuli. L-VDCC antagonists can reduce the secretion of cytokines and ROS and may represent a potential therapy for resolution of chronic neuroinflammation that is seen during AD, PD, HD, and neuropsychiatric illness

Intracellular Ca²⁺ regulates microglial transformation from a homeostatic resting state to a neurotoxic-activated immune-effector state. Microglia Ca²⁺ influx via a variety of mechanisms was first reported in the early 1990s as a mechanism by which these cells could rapidly respond to brain injury (Walz, Ilschner, Ohlemeyer, Banati, & Kettenmann, 1993). Later studies by Hoffman and colleagues found that microglia activation by the inflammagen lipopolysaccharide (LPS) resulted in increased Ca²⁺ influx, which is sustained throughout activation. The addition of a Ca²⁺ chelator reduced LPS-induced ROS and pro-inflammatory cytokine release from microglia, indicating that extracellular entry of Ca²⁺ is essential for LPS-induced microglia activation (Hoffmann, Kann, Ohlemeyer, Hanisch, & Kettenmann, 2003). Notably, microglia activated by LPS display elevated levels of basal Ca²⁺ and their receptor-evoked Ca²⁺ increases are blunted; in essence, microglia “ignore” additional stimuli while activated (Hoffmann et al., 2003). The pro-inflammatory cytokines, tumor necrosis factor α (TNFα), interleukin 1β (IL-1β), and interferon γ (IFNγ), all increase microglia intracellular Ca²⁺ (Franciosi, Choi, Kim, & McLarnon, 2002; Goghari, Franciosi, Kim, Lee, & McLarnon, 2000; McLarnon et al., 2001) suggesting the presence of a Ca²⁺-dependent pro-inflammatory feedback cycle that could contribute to the propagation of microglia activation. Notably, NLRP3 inflammasome activation requires Ca²⁺ mobilization (Murakami et al., 2012). Anti-inflammatory cytokines, in contrast, reduce intracellular Ca²⁺ levels in microglia. For example, the application of IL-10 reduced intracellular Ca²⁺ in microglia exposed to NMDA excitotoxicity and hypoxia (Turovskaya, Turovsky, Zinchenko, Levin, & Godukhin, 2012). Conversely, an indirect blockade of Ca²⁺ entry into LPS-activated microglia stimulates anti-inflammatory cytokine production (Dolga et al., 2012). Recent advances in in vivo calcium imaging by multiphoton microscopy has allowed for the investigation of microglia calcium dynamics in the intact brain during aging, injury, and disease (Brawek & Garaschuk, 2013; Brawek et al., 2014; Olmedillas del Moral, Asavapanumas, Uzcátegui, & Garaschuk, 2019; Pozner et al., 2015; Tvrdik & Kalani, 2017). In vivo, resting surveillant microglia display few Ca²⁺ transients (Schwendele, Brawek, Hermes, & Garaschuk, 2012) but respond with large Ca²⁺ transients following nearby neuronal injury (Eichhoff, Brawek, & Garaschuk, 2011).

The activity of calcium-binding proteins in signaling pathways in microglia has not been as thoroughly described as other cell types (Möller, 2002) although recent progress has been made in this area as illustrated in Figure 2. One of the most well-known Ca²⁺ binding proteins in microglia is ionized calcium-binding adaptor molecule 1 (Iba1) which is ubiquitously used throughout scientific literature to visualize microglia during immunohistochemistry (Imai, Ibata, Ito, Ohsawa, & Kohsaka, 1996). Notably, Iba1 shares structural features with CaM and may bind Ca²⁺ or partner proteins via similar mechanisms (Yamada, Ohsawa, Imai, Kohsaka, & Kamitori, 2006). One important Ca²⁺- and Iba1-dependent function in microglia is the regulation of the actin cytoskeleton during chemotaxis (Siddiqui, Lively, Vincent, & Schlichter, 2012). CaM is also highly expressed in microglia and regulates signaling pathways involved in morphology, phagocytosis, proliferation, activation, and modulation of Ca²⁺ ion channel activity (Casal, Tusell, & Serratosa, 2001; Franco et al., 2018; Szabo, Dulka, & Gulya, 2016; Wong & Schlichter, 2014). Ca²⁺ in microglia can also transduce signals via the Ca²⁺-dependent phosphatase calcineurin to regulate microglia inflammatory state via regulation of transcription factors such as nuclear factor kappa B (NFκB) and nuclear factor of activated T cells (NFAT) (Kraft, 2015; Kurland et al., 2016; Liu et al., 2017). The specific and organized activation of signaling pathways likely relies on microglial intracellular concentrated Ca²⁺ microdomains (Berridge, 2006). Microglia Ca²⁺ microdomains have not been studied extensively due to technology limitations (Tvrdik & Kalani, 2017), although one study asserts that in microglia, global Ca²⁺ elevation triggers phagocytosis and migration via transcription of necessary genes, while local Ca²⁺ increases in cell processes regulate acute chemotactic migration (Lim et al., 2017). Further study of the subcellular organization of microglial Ca²⁺ binding proteins, Ca²⁺-dependent signaling pathways, and Ca²⁺ microdomains is necessary to understand how Ca²⁺ orchestrates numerous microglia functions.

FIGURE 2.

Calcium (Ca²⁺) signaling in microglia. Microglia use a diverse toolkit of calcium channels, pumps, and binding proteins to control cytosolic calcium levels and regulate Ca²⁺-dependent signaling pathways. Plasma membrane Na⁺/Ca²⁺ exchanger (NCX) and plasma membrane Ca²⁺ ATPase (PMCA) pumps regulate the Ca²⁺ gradient. Ca²⁺ enters the cytosol from the extracellular space via a variety of ligand-gated ion channels (e.g., purinergic ionotropic purine receptors (P2Xs) or glutamatergic N-methyl-D-aspartate receptors [NMDARs]), transient receptor potentials (TRPs), L-type voltage-dependent Ca²⁺ channels (L-VDCCs), or Orai Ca²⁺-release-activated Ca²⁺ (CRAC) channels. G protein-coupled receptors (GPCRs) on the plasma membrane can indirectly increase intracellular Ca²⁺ via phospholipase C (PLC)-mediated inositol trisphosphate (IP3) signaling to inositol trisphosphate receptors (IP3Rs) on the endoplasmic reticulum (ER) Ca²⁺ store. Ryanodine receptors (RyRs) can also release Ca²⁺ from ER stores and can be activated by Ca²⁺, mediating Ca²⁺-induced Ca²⁺ release. Stromal interaction molecule (STIM) 1 and 2 monitor ER Ca²⁺ levels and interacts with Orai and L-VDCCs to mediate SOCE while SERCA pumps on the ER transport Ca²⁺ into the ER lumen to replenish depleted ER Ca²⁺ stores. Regulation of L-VDCCs on microglia is not well characterized; microglia may express voltage-insensitive splice variants of L-VDCCs similar to T-lymphocytes, as indicated by the “?” adjacent to the voltage sensor in this figure. Ca²⁺ channels on lysosomes and mitochondria also sequester cytosolic Ca²⁺. Mitochondrial Ca²⁺ alters oxidative stress and production of reactive oxygen species (ROS). Cytosolic Ca²⁺ interacts with several binding proteins including ionized calcium-binding adaptor molecule 1 (Iba1), calmodulin (CaM), and calcineurin to regulate numerous functions including actin remodeling, chemotaxis, phagocytosis, and regulation of Ca²⁺ channels. Calcineurin regulates transcription via nuclear factor of activated T cells (NFAT) and nuclear factor kappa B (NFκB), controlling transcription of cytokines, chemokines, Nos2 for the production of iNOS, and other inflammatory factors. Ca²⁺ also regulates the formation of the inflammasome and subsequent production of IL-1β. Notably, inflammatory cytokines feed back onto intracellular and plasma membrane Ca²⁺ channels, leading to modulation of cytosolic Ca²⁺ levels

Similar to all cells, Ca²⁺ in microglia comes from intracellular stores or can enter via plasma membrane channels; there is a balanced communication between these sources via SOCE. In microglia, blocking SOCE via STIM1 and Orai1 reduces important immune functions such as phagocytosis and cytokine production (Heo, Lim, Nam, Lee, & Kim, 2015). ER Ca²⁺ release is largely regulated by the activation of IP3Rs and RyRs. In microglia, RyR antagonism prevents LPS-induced neurotoxicity mediated by microglia (Klegeris, Choi, McLarnon, & McGeer, 2007). As Figure 2 illustrates, the microglia plasma membrane also contains numerous other ligand-, ion-, and voltage-gated Ca²⁺ channels that serve as direct or indirect Ca²⁺ sources under different conditions including L-VDCCs, G protein-coupled receptors (GPCRs), transient receptor potential (TRP) channels, ligand-gated Ca²⁺ channels such as the ionotropic purine receptors (P2Xs) or glutamatergic N-methyl-D-aspartate receptors (NMDARs), NCX and PMCA transporters, among others (Möller, Kann, Verkhratsky, & Kettenmann, 2000; Walz et al., 1993). This review focuses on the role of L-VDCCs on microglia during health and disease.

4 |. L-VDCCS

Voltage-dependent Ca²⁺ channels (VDCCs) are expressed throughout the body, inclusive of both the brain and peripheral tissues and excitable and non-excitable cells. VDCCs are composed of four subunits: α1, α2δ, β, and γ. The α2δ and β subunits regulate channel trafficking and expression and perform other key functions reviewed elsewhere (Dolphin, 2012), and regulate signal transduction pathways (Servili, Trus, & Atlas, 2019). This review will focus on the α1 subunit, which forms the ion conduction pore, the voltage sensor, and contains drug binding sites and is thus the major determinant of the biophysical properties of VDCCs. Notably, in addition to their ion channel function, L-VDCCs also include transcription factors at their C terminals (Gomez-Ospina, Tsuruta, Barreto-Chang, Hu, & Dolmetsch, 2006; Lu et al., 2015) and can trigger gene expression independent of Ca²⁺ entry (Servili et al., 2019). Subtypes of VDCCs are distinguished by their distinct α1 subunits which are divided into three distinct subclasses, namely Cav1, Cav2, and Cav3 (Catterall, 2011). These subtypes of VDCCs with unique biophysics and pharmacology confer varied biological functions to the distinct cell types where they are differentially localized. The focus of this review, the L-VDCC, is slow inactivating high voltage-activated channels with large single-unit conductance which utilize the Cav1 subclass of the α1 subunit; within the Cav1 subclass are four specific subtypes: Cav1.1, Cav1.2, Cav1.3, and Cav1.4 (Lipscombe, Helton, & Xu, 2004). Cav1.2 and Cav1.3 are expressed on neurons and other cells throughout the brain with differential expression both subcellularly within the same neuron (Hell et al., 1993) and across different brain regions (Kabir, Martínez-Rivera, et al., 2017). Cav1.3 and Cav1.4 are critical to the function of ear hair cells and retinal photoreceptors, respectively (Pangrsic, Singer, & Koschak, 2018). Cav1.1 is largely restricted to skeletal muscle, where it is essential for excitation-contraction coupling; Cav1.2 and Cav1.3 also regulate excitation-contraction coupling in cardiac myocytes and smooth muscle cells (Lipscombe et al., 2004). Cav1 L-VDCCs are also localized to endocrine cells (Schulla et al., 2003; Yang, He, Zhang, Barrett, & Hu, 2019) as well as lymphoid and myeloid immune cells (Davenport, Li, Heizer, Schmitz, & Perraud, 2015; Komoda, Shiraki, Oyama, Nishikido, & Node, 2018; Suzuki, Inoue, & Ra, 2012) which express all Cav1 subtypes at varying levels (Choi et al., 2019; Fenninger et al., 2019; Komoda et al., 2018; Sharma et al., 2016). Timothy syndrome (TS), a rare autosomal-dominant disorder affecting multiple organ systems, demonstrates the importance of Cav1.2 throughout the body: mutations in CACNA1C, the gene encoding Cav1.2, lead to heart arrhythmias and structural cardiac defects, immune system dysfunction, neurodevelopmental difficulties such as ASD, and physical malformations such as syndactyly (Dixon, Cheng, Mercado, & Santana, 2012; Liao & Soong, 2010; Splawski et al., 2004).

In addition to differential subunit distribution, the α1 L-VDCC subunit genes are also alternatively spliced between different cells and tissues. In the brain, there are hundreds of splice variants of Cav1.2 (Clark et al., 2019) and different Cav1.3 splice variants regulate dendritic spine morphology on neurons (Stanika et al., 2016). Cav1.2 mRNA is also extensively alternatively spliced in cardiac tissue (Wang, Papp, Binkley, Johnson, & Sadée, 2006) which significantly changes the biophysical properties of the channel (Bartels et al., 2018). In the immune system, T-lymphocyte cells express splice variants of Cav1.4 unique from retinal Cav1.4 in that these variants remove exons that render the resulting channel insensitive to membrane depolarization (Badou et al., 2006; Kotturi & Jefferies, 2005); a similar voltage-insensitive Cav1.3 channel has also been discovered in hepatoma cells (Brereton, Harland, Froscio, Petronijevic, & Barritt, 1997). These data suggest that cells may each have their own unique splice variants of L-VDCC subunits that regulate their tissue-specific function via alterations in biophysical and pharmacological properties.

5 |. L-VDCCS ON NON-EXCITABLE CELLS

The presence, purpose, and function of L-VDCCs have been well-characterized on excitable cells including neurons, myocytes, and excitable endocrine cells where L-VDCCs mediate excitation-contraction coupling, excitation-transcription coupling, and excitation-secretion coupling (Catterall, 2011; Hofmann, Flockerzi, Kahl, & Wegener, 2014), but L-VDCC function on non-excitable cells, such as peripheral immune cells, has been the topic of considerable debate. L-VDCCs are clearly vital for normal T-lymphocyte function since genetic ablation of Cav1.1 or Cav1.4 causes Ca²⁺-mediated dysfunction in these cells, including reduced transcriptional activation of NFAT reduced proliferation and blunted cytokine production (Badou et al., 2006; Fenninger et al., 2019; Fenninger & Jefferies, 2019; Matza et al., 2016) and patients with TS display immune deficiencies (Splawski et al., 2004). On excitable cells, membrane depolarization leads to a clearly detectable Ca²⁺ current while this current is elusive on T lymphocytes (Cahalan & Chandy, 2009). Indeed, the voltage dependence of the channel is removed via alternative splicing of Cav1.1 and Cav1.4 L-VDCCs in T lymphocytes (Kotturi & Jefferies, 2005; Matza et al., 2016), raising the question of how L-VDCCs are regulated on these cells. T-lymphocyte Cav1.2 is inhibited by STIM1 during T-cell receptor activation (Park et al., 2010; Wang et al., 2010) suggesting that perhaps Cav1.2 is instead important in survival and maintenance of resting T lymphocytes rather than during activation (Fenninger & Jefferies, 2019); however, the signaling pathways that regulate Ca²⁺ influx via T-lymphocyte Cav1.4 and Cav1.1 L-VDCCs are currently unknown. T lymphocyte-specific Cav1.1 splice variants that lack voltage sensitivity are constitutively open at resting potential (Matza et al., 2016). While myeloid-specific splice variants have not been identified, L-VDCC antagonists have immune-dampening effects on monocytes and microglia similar to their effect on lymphocytes (Briede et al., 1999; Das, Burke, Wagoner, & Plow, 2009; Davenport et al., 2015; Espinosa-Parrilla, Martínez-Moreno, Gasull, Mahy, & Rodríguez, 2015; Komoda et al., 2018). Taken together, these data suggest that the role of L-VDCCs in non-excitable cells displays markedly different biophysical properties from those present on excitable cells, yet are vital for cellular function.

6 |. L-VDCCS IN THE CNS

6.1 |. L-VDCCs on neurons

Cell culture and animal models have led to a thorough characterization of L-VDCCs on neurons which has been described extensively by other reviews (Striessnig, Pinggera, Kaur, Bock, & Tuluc, 2014); here we will briefly highlight studies investigating L-VDCC function on neurons. Neuronal L-VDCCs contribute to long-term potentiation (LTP) following high-frequency afferent input to the hippocampal CA1 region, which is induced by 200 Hz stimulation compared to the 25 Hz stimulation that induces N-methyl-D-aspartate receptor (NMDAR)-dependent LTP (Grover & Teyler, 1990). The afterhyperpolarization (AHP) that follows trains of CA1 pyramidal neuron actional potentials controls the threshold required for LTP is driven by L-VDCCs, specifically Cav1.3 (Gamelli, McKinney, White, & Murphy, 2011). Functionally, the involvement of L-VDCCs in LTP likely explains their role in learning and memory. Post-retrieval antagonism of hippocampal CA1 L-VDCCs results in spatial memory retention deficits (Da Silva, Cardoso, Bonini, Benetti, & Izquierdo, 2013). Furthermore, conditional deletion of neuronal Cav1.2 results in suppression of hippocampal-dependent memory as well as neurogenesis in the hippocampus (Temme, Bell, Fisher, & Murphy, 2016) and deletion of Cav1.3 results in impaired contextual fear consolidation and altered neurophysiology of the amygdala (McKinney, Sze, Lee, & Murphy, 2009). Forebrain-specific neuronal deletion of Cav1.2 enhanced cell death of hippocampal neurons and reduced levels of brain-derived neurotrophic factor (Lee et al., 2016). Notably, the balance between NMDA- and L-VDCC-dependent learning and memory changes during aging, suggesting a role for L-VDCCs age-associated memory deficits (Foster, 2007).

L-VDCCs also have critical roles in several neuron functions beyond LTP and are involved in non-cognitive behaviors in rodents. In brain development, spontaneous Cav1.2 Ca²⁺ transients regulate the growth and migration of neurites (Kamijo et al., 2018), which may be related to the role of this channel in neurodevelopmental disorders. Cav1.2 and Cav1.3 alter sociability, stress response, anxiety- and depressive-like behaviors in mice (Busquet et al., 2010; Dao et al., 2010; Dedic et al., 2018; Ehlinger & Commons, 2019; Lee et al., 2012; Kabir, Che, et al., 2017; Terrillion, Francis, Puche, Lobo, & Gould, 2017) which are discussed extensively later in the section on neuropsychiatric disorders. Increases in Cav1.2 and Cav1.3 expression are associated with the development of morphine tolerance and dependence (Alboghobeish, Naghizadeh, Kheirollah, Ghorbanzadeh, & Mansouri, 2019). Neuronal Cav1.2 and Cav1.3 are both present in neurons of the SN pars compacta, a region critical in the coordination of movement and the pathophysiology of PD where Cav1.3 drives pacemaking activity (Chan et al., 2007; Nedergaard, Flatman, & Engberg, 1993). Deletion or antagonism of Cav1.2 can be neuroprotective by triggering resilience to oxidative stress in neurons (Michels et al., 2018), particularly in SN neurons (Guzman et al., 2018) which will be further discussed in a later section covering PD.

6.2 |. L-VDCCs on astrocytes and oligodendrocytes

L-VDCCs on non-neuronal CNS cell types are not well-characterized. Cultured astrocytes express protein and mRNA for L-VDCCs (He et al., 2016; Yaguchi & Nishizaki, 2010), display L-VDCC currents (D’Ascenzo et al., 2004), and respond to L-VDCC antagonists (Cheli, Santiago González, et al., 2016; Hashioka, Klegeris, & McGeer, 2012). In vivo, L-VDCCs are found on early postnatal hippocampal astrocytes (Akopian, Kressin, Derouiche, & Steinhäuser, 1996), and Cav1.2 expression is increased on astrocytes near amyloid plaques in mouse models of AD (Daschil et al., 2013) as well as in a mouse model of hyperammonemia (Wang, Du, Liang, Verkhratsky, & Peng, 2015), suggesting a role for L-VDCCs in astrogliosis. In vivo astrocytes also upregulate the gene expression of Cav1.2 in response to chronic fluoxetine selective serotonin reuptake inhibitor (SSRI) antidepressant treatment (Du, Liang, Li, Hertz, & Peng, 2014). Conditional deletion of astrocytic Cav1.2 also results in deficits of hippocampal neurogenesis (Völkening, Schönig, Kronenberg, Bartsch, & Weber, 2017). In oligodendrocytes, L-VDCC activation and expression regulates oligodendrocyte development both in vitro and in vivo (Cheli, González, et al., 2016; Paez et al., 2007).

6.3 |. L-VDCCs on microglia in vitro

Dihydropyridine-sensitive putative L-VDCC Ca²⁺ currents have been reported on cultured mouse and human microglia for at least two decades. Initial reports by Colton and colleagues (Colton, Jia, Li, & Gilbert, 1994) found that the application of the L-VDCC activator, BAY-K8644, increased voltage-dependent Ca²⁺ influx, which could be blocked by the dihydropyridine L-VDCC antagonist nifedipine. Activation of these putative L-VDCCs on microglia also induced microglia superoxide production (Colton et al., 1994), suggesting a role for these channels in the regulation of microglia activation and/or mitochondria function.

Following initial reports of dihydropyridine-sensitive Ca²⁺ channels on microglia, several laboratories set forth to examine the physiological role of these channels using pharmacological manipulation. Addition of proteins associated with neurodegenerative diseases, amyloid β (Aβ) or prion protein (PrP), to microglia cultures increased their immune activation (as measured by cytokine production) as well as intracellular Ca²⁺ and these processes were blocked by both the dihydropyridine nifedipine or the non-dihydropyridine L-VDCC antagonists verapamil or diltiazem (Silei et al., 1999). Nimodipine, another dihydropyridine L-VDCC antagonist, reduced microglia Ca²⁺ influx when exposed to the chemokine RANTES (Hegg, Hu, Peterson, & Thayer, 2000) and prevented microglia production of cytokines and nitric oxide (NO) and reduced neurotoxicity following LPS treatment (Li, Hu, Liu, Bao, & An, 2009). Others have observed similar reductions in microglia-mediated activation and neurotoxicity by dihydropyridine drugs following treatment with inflammagens, cytokines, or Aβ (Gao et al., 2013; Hashioka et al., 2012; Huang et al., 2014; Sanz et al., 2012). The majority of the literature demonstrates agreement that L-VDCC antagonists are anti-inflammatory, although one recent report found that these drugs may potentiate microglia inflammatory response in vitro (Wang, Saegusa, Huntula, & Tanabe, 2019). Microglia are the tissue-specific myeloid cells of the CNS, so there is likely overlap with the function of peripheral myeloid lineage cells, in which L-VDCCs have also been examined. Peripheral blood mononuclear cells (PMBCs) and macrophages use L-VDCCs to regulate their immune response including regulation of phagocytosis and production of pro-inflammatory cytokines (Allanore et al., 2005; Gupta et al., 2009; Matsumori, Nishio, & Nose, 2010).

The presence of L-VDCCs on microglia in vitro is still debated (Kettenmann, Hanisch, Noda, & Verkhratsky, 2011). There are conflicting reports suggesting microglia do not express L-VDCCs (Kettenmann et al., 2011; Schampel et al., 2017) while others have found gene and protein expression of L-VDCCs on microglia (Espinosa-Parrilla et al., 2015). RNA-sequencing databases suggest that microglia do express variable and low amounts of L-VDCC gene expression, particularly CACNA1D (encoding Cav1.3), and different splice variants (Hammond et al., 2019; Zhang et al., 2014). The overall uncertainty about whether microglia even express L-VDCCs may be due to activation-dependent differential expression similar to other microglia channels (Nguyen et al., 2017), or an uncharacterized splice variant of L-VDCC subunits that is microglia-specific, similar to that which is observed in T lymphocytes (Kotturi & Jefferies, 2005; Matza et al., 2016). In the brain, there are hundreds of splice variants of Cav1.2 (Clark et al., 2019) and it is not clear which, if any of these splice variants may be specific to microglia. Microglia may express their own voltage-insensitive unique splice variant of L-VDCCs similar to T lymphocytes (Kotturi & Jefferies, 2005) which would explain the elusiveness of typical L-VDCC currents on microglia. Functionally, various kinds of inflammagens increase microglia Ca²⁺ influx via dihydropyridine-sensitive mechanisms (Hegg et al., 2000; Silei et al., 1999) and both dihydropyridine and other L-VDCC antagonists produce an anti-inflammatory phenocopy (Espinosa-Parrilla et al., 2015; Gao et al., 2013; Huang et al., 2014; Li et al., 2009; Sanz et al., 2012). However, the anti-inflammatory effects of putative L-VDCC antagonists in vitro may be due to off-target effects via NADPH oxidase (Li et al., 2009) or other ion channels such as voltage-dependent Na⁺ channels, Ca²⁺-dependent K⁺ channels, or voltage-gated K⁺ channels (Zhang & Gold, 2009).

6.4 |. L-VDCCs on microglia in vivo

The in vitro literature above demonstrates a consensus that L-VDCC antagonists are anti-inflammatory in vitro. Echoing this in vitro data, the dihydropyridine nicardipine inhibits pro-inflammatory cytokine expression following a single in vivo LPS injection (Huang et al., 2014). Similarly, the dihydropyridine nimodipine is anti-inflammatory in a chronic model of intracerebroventricular LPS infusion and can reverse chronic inflammation-induced spatial memory deficits, hippocampal microglia activation, and hippocampal inflammatory gene expression (Hopp, D’Angelo, et al., 2015). In the same chronic LPS model, nimodipine also reduced inflammation in brainstem nuclei and altered inflammation-induced deficits in brainstem-dependent behaviors (Hopp, Royer, et al., 2015). However, nimodipine treatment did not alter age-associated increases in hippocampal microglia activation in aged rats (Hopp et al., 2014).

The in vivo literature on L-VDCC antagonism during CNS injury is less consistent with respect to microglia involvement. Nimodipine treatment is neuroprotective in various other injury or disease models, including excitotoxic injury to the magnocellular nucleus basalis (Harkany et al., 2000) and stroke (Babu & Ramanathan, 2011). The mechanisms of neuroprotection are unclear. For example, after ischemic-reperfusion injury L-VDCC antagonism improves behavioral outcomes while concurrently reducing microglia activation (Yanpallewar, Hota, Rai, Kumar, & Acharya, 2004). However, following facial nerve transection, L-VDCC antagonism improves motor neuron survival without reducing microglia activation (Mattsson, Aldskogius, & Svensson, 1999). Two recent studies found that the dihydropyridine L-VDCC antagonist nimodipine was neuroprotective in mouse models of multiple sclerosis: both papers found that there were fewer microglia in regions affected by the disease (Ingwersen et al., 2018; Schampel et al., 2017) but one study found no changes in microglia activation markers (Ingwersen et al., 2018) while the other found that nimodipine did reduce microglia activation markers such as inducible nitric oxide synthase (iNOS) and ROS (Schampel et al., 2017). The latter study also found that nimodipine treatment could trigger microglia cell death both in vivo and in vitro, via L-VDCC-independent mechanisms (Schampel et al., 2017) in contrast to other studies outlined above. Recently, the first study to genetically reduce expression of the L-VDCC Cav1.2 subunit specifically on microglia found that reduced expression increased microglia pro-inflammatory activation and increased neurotoxicity (Wang et al., 2019). While most of the literature seems to agree that blocking microglia L-VDCCs reduces their pro-inflammatory activation state, the only study using genetic reduction of microglia-specific Cav1.2 observed an increase in microglia pro-inflammatory markers. Clarifying the role of L-VDCCs on microglia is an important area of research relevant to several CNS disorders.

7 |. L-VDCCS IN CNS DISORDERS

7.1 |. Aging

Aging is the largest risk factor for neurodegenerative disease including AD, HD, and PD, and Ca²⁺ dysregulation is a common feature of aging and these disorders. In aging, increased CNS L-VDCC protein expression causes increased L-VDCC activity in the hippocampus (Thibault & Landfield, 1996) leading to detrimental increases in the AHP, which is associated with increased long-term depression (LTD) and spatial memory deficits due to interference with NMDAR-dependent LTP (Disterhoft, Thompson, Moyer, & Mogul, 1996; Kumar & Foster, 2004). These age-associated changes in L-VDCCs are predominantly due to increases in Cav1.3 expression (Chen, Blalock, Thibault, Kaminker, & Landfield, 2000) but may also involve trafficking and post-translational modifications of Cav1.2 that increase its activity (Davare & Hell, 2003; Núñez-Santana et al., 2014). Similarly, during aging SN pars compacta neurons become increasingly reliant on Cav1.3 channels to drive their autonomous pacemaking activity, rendering them vulnerable to stressors associated with PD (Chan et al., 2007). Increased intracellular Ca²⁺ during aging may set the stage for the development of AD and PD pathology, since Ca²⁺ dysregulation can contribute to the production and aggregation of misfolded proteins characteristic of AD and PD, including Aβ, tau, and α-synuclein (Lautenschläger et al., 2018; Stutzmann, 2007). Notably, longitudinal studies of aging demonstrate that dihydropyridine CCBs may be neuroprotective against AD and PD (Forette et al., 2002; Hoffman et al., 2009; Ritz et al., 2009; Yasar, Corrada, Brookmeyer, & Kawas, 2005).

Age-associated neuroinflammation may drive age-associated increases in L-VDCC function since TNFα can increase L-VDCC activity (Furukawa & Mattson, 1998) and blocking TNFα in aged rodents reduced age-associated increases in L-VDCC activity and LTD (Sama et al., 2012). Furthermore, age-associated inflammation may lead to changes in microglia Ca²⁺ that could be related to L-VDCC activity. Aged microglia show increased numbers of intracellular Ca²⁺ transients in vivo and dysfunctional response to stimuli (Olmedillas del Moral et al., 2019). However, treating aged rats with the L-VDCC antagonist nimodipine at a dose that can reverse LPS-induced microgliosis did not significantly reduce age-associated microglia activation and only slightly improved age-associated spatial memory deficits (Hopp et al., 2014; Riekkinen, Schmidt, Kuitunen, & Riekkinen, 1997).

7.2 |. Alzheimer’s disease

AD, the most common neurodegenerative disease, is characterized by the progressive loss of cognitive function due to neurodegeneration of the hippocampus and other connected brain regions. Neuropathologically, AD is characterized by neuron loss, deposition of extracellular Aβ plaques, accumulation of intracellular tau tangles, and gliosis (Braak & Braak, 1991; Eikelenboom et al., 2010). Familial, autosomal-dominant AD is rare and is caused by mutations in amyloid precursor protein and cleavage processes that lead to excessive Aβ deposition. However, the mutations associated with the more common sporadic AD are not as clearly linked to Aβ deposition and rather involve regulation of lipid metabolism, endocytosis, and immune function (Pimenova, Marcora, & Goate, 2017). The “calcium hypothesis of AD” links age-associated intracellular Ca²⁺ dysregulation and the amyloidogenic pathway to cognitive dysfunction and neurodegeneration in AD (Berridge, 2010). In rodent models of AD, intracellular Ca²⁺ dysregulation is observed in several cell types near amyloid plaque deposits, including neurons (Arbel-Ornath et al., 2017; Kuchibhotla et al., 2008), astrocytes (Kuchibhotla, Lattarulo, Hyman, & Bacskai, 2009), and microglia (Brawek et al., 2014). While the consequences of altered intracellular Ca²⁺ dynamics near plaques in AD are not fully understood, increased intracellular Ca²⁺ can interfere with normal synaptic plasticity and contribute to excitotoxicity in neurons (Paula-Lima et al., 2011), is involved with formation of pathological lesions (Stutzmann, 2007), may dysregulate astrocyte- mediated hemodynamics (Kelly, Hudry, Hou, & Bacskai, 2018), and perturb microglia stimuli responsiveness (Brawek et al., 2014). Furthermore, Ca²⁺ dynamics are altered in non-neuronal tissues such as fibroblasts from human AD patients and unaffected family members (Etcheberrigaray et al., 1998; Peterson & Goldman, 1986) suggesting that increased intracellular Ca²⁺ precedes amyloid and tau pathology and may play a role in disease pathogenesis.

The source of excess intracellular Ca²⁺ during AD pathology is an unresolved question. NMDA-type glutamate receptors are a potential target and one of the few current therapies for AD is memantine, an NMDA channel antagonist, which can slightly improve cognitive symptoms in AD patients (Danysz & Parsons, 2012). Other studies have implicated AMPA glutamate receptors (Alberdi et al., 2010; Resende et al., 2007; Zhao et al., 2010) or even direct Ca²⁺ pore formation by Aβ (Sepúlveda et al., 2014) as sources of excess Ca²⁺. L-VDCCs are another potential source of excessive intracellular Ca²⁺ during AD. STIM1 mediates Ca²⁺ dysregulation in fibroblasts from human AD patients by deregulating Ca²⁺ entry via L-VDCCs; Cav1.2 knockout or dihydropyridine L-VDCC antagonists reversed this phenotype (Pascual-Caro et al., 2018). Further implicating L-VDCCs in AD are early studies showing increases in dihydropyridine radioligand binding association with cell loss in AD brains relative to controls (Coon, Wallace, Mactutus, & Booze, 1999). Recent studies show an association between AD pathology and polymorphisms (Heck et al., 2015) and an increase in copy number variations (Villela, Suemoto, Pasqualucci, Grinberg, & Rosenberg, 2016) of L-VDCC subunits. Indeed, epidemiological evidence suggests that the use of dihydropyridines in humans is associated with mild improvements in dementia, reduced pathological burden, and decreased risk of developing AD (Forette et al., 2002; Hoffman et al., 2009; López-Arrieta & Birks, 2002; Yasar et al., 2005).

Evidence from cell culture and transgenic animal models also indicate a role for L-VDCCs in AD pathophysiology. In vitro, increased expression of CACNA1C (encoding Cav1.2) is associated with increases in tau pathology (Jiang et al., 2018) and apolipoprotein 4 (ApoE4), the late-onset AD risk gene, which can increase L-VDCC activity in neurons (Ohkubo et al., 2001). Aβ can directly interact with L-VDCC channel components and alter their functions (Kim & Rhim, 2011) and blockade of L-VDCCs by nimodipine attenuates Aβ neurotoxicity (Weiss, Pike, & Cotman, 2008). Ca²⁺ entry via L-VDCCs is pointed to as a potential mechanism linking amyloid and tau pathology (Pierrot et al., 2006). In mouse models of AD, Cav1.2 expression is increased on astrocytes near amyloid plaques (Daschil et al., 2013). Injection of the highly amyloidogenic 42-residue, Aβ42, induces hyperexcitability and memory deficits that are reversible with isradipine or nimodipine treatment (Gholami Pourbadie, Naderi, Janahmadi, Mehranfard, & Motamedi, 2016; Pourbadie et al., 2015, 2017). In the APP/PS1 transgenic AD mouse model, nimodipine treatment improves spatial memory via Ca²⁺-dependent mechanisms (Wang et al., 2012). Two studies in transgenic animal models of AD found that treatment with dihydropyridines could reduce Aβ plaque burden via alterations in cleavage of amyloid precursor protein (APP) into Aβ and change the clearance of Aβ from the CNS, via stimulation of autophagy (Copenhaver et al., 2011; Paris et al., 2011). In contrast, another study found that nimodipine stimulated the production of the amyloidogenic Aβ42 fragment in vitro and increased plasma levels in vivo via Ca²⁺-independent mechanisms (Facchinetti et al., 2006).

Notably, these in vivo studies did not examine the effects of dihydropyridine treatment on microglia in the context of AD pathology, despite a critical role for microglia in disease pathogenesis and progression via loss of homeostatic functions, a failure to clear protein aggregates, release of toxic factors, and contribution to protein aggregate formation (Cameron & Landreth, 2010; Pimenova, Raj, & Goate, 2018; Shi & Holtzman, 2018). Microglia activation appears early in AD (Eikelenboom et al., 2010) and is correlated with cognitive deficits (Perez-Nievas et al., 2013). Microglia from human AD brains display higher baseline Ca²⁺ levels compared to non-demented controls and have a reduced responsiveness to stimuli in vitro (McLarnon, Choi, Lue, Walker, & Kim, 2005) which is similar to results from in vivo imaging studies of AD transgenic mice showing that plaque-associated microglia have increased calcium transients and reduced stimuli responsiveness compared to resting, ramified microglia (Brawek et al., 2014). In vitro studies have demonstrated that dihydropyridines can reduce microglia pro-inflammatory response to Aβ (Sanz et al., 2012; Silei et al., 1999). Perhaps dihydropyridines could alter calcium dysregulation in plaque-associated microglia due to the involvement of L-VDCCs in SOCE (Park et al., 2010; Wang et al., 2010), which is disrupted in AD-derived microglia (McLarnon et al., 2005). Pro-inflammatory cytokines that are found in AD (Sudduth, Schmitt, Nelson, & Wilcock, 2013) would feed back onto microglia further increasing intracellular Ca²⁺ and propagating neuroinflammation (Franciosi et al., 2002; Goghari et al., 2000; McLarnon et al., 2001). Taken together, these data demonstrate the importance of examination of the microglia response to L-VDCC antagonists during AD pathology as a potential mechanism of neuroprotection.

7.3 |. Parkinson’s disease

PD, the second most common neurodegenerative disease, is characterized by the progressive loss of voluntary movement (akinesia) and, in many cases, cognitive deficits. Neuropathologically, PD is characterized by the loss of dopaminergic neurons in the SN pars compacta and accumulation of intracellular misfolded α-synuclein into Lewy bodies. What renders SN dopaminergic neurons specifically vulnerable to neurodegeneration has long remained unclear; hypotheses include oxidative stress due to dopamine use, the bioenergetic demands of highly branched axons, and dependence on Ca²⁺ for pacemaking (Surmeier et al., 2017). Polymorphisms in genes associated with mitochondria, endocytosis, and lysosome function have all been linked to PD (Billingsley, Bandres-Ciga, Saez-Atienzar, & Singleton, 2018), but their precise role in disease pathogenesis is poorly understood. Variants leading to mitochondria dysfunction are the most well-characterized and could render SN dopaminergic neurons vulnerable to stress from their high-baseline energy requirements and intracellular Ca²⁺ (Surmeier et al., 2017). The toxin 1-methyl-4-phenyl-1,2,5,6-tetrahydropyridine (MPTP) which causes a PD-like syndrome is a mitochondrial toxin (Mattson, 2012) and L-VDCC antagonists protect SN neurons from MPTP-mediated degeneration (Ilijic, Guzman, & Surmeier, 2011; Kupsch et al., 1996) via reduced mitochondrial stress (Guzman et al., 2018). During aging, SN neurons become even more reliant on L-VDCCs and Ca²⁺, rendering them vulnerable to neurotoxicity that is preventable with L-VDCC antagonists (Chan et al., 2007). Furthermore, like Aβ and tau aggregates in AD, α-synuclein contributes to Ca²⁺ dysregulation and Ca²⁺ dysregulation can lead to α-synuclein aggregation (Lautenschläger et al., 2018; Mosharov et al., 2009). Importantly, epidemiological studies support the hypothesis that excessive L-VDCC activity in the SN renders neurons vulnerable to toxicity: patients taking L-VDCC antagonists have a reduced incidence of PD (Marras et al., 2012; Pasternak et al., 2012; Pfeiffer, 2010; Ritz et al., 2009; Swart & Hurley, 2016).

In addition to selective vulnerability of SN dopaminergic neurons to degeneration, microgliosis is a hallmark pathological feature of PD (Iannaccone et al., 2013; Tansey & Goldberg, 2010) and microglia phenotype changes occur early during PD and may be an initiating factor that links protein aggregation, cognitive and motor deficits, and neurodegeneration (Joers, Tansey, Mulas, & Carta, 2017). Furthermore, the SN also shows selective vulnerability to neuroinflammatory insult (Bardou et al., 2014; Qin, Liu, Hong, & Crews, 2013). In mesencephalic neuron or neuron-glia mixed cultures, the L-VDCC antagonist, nimodipine, protected neurons from both inflammatory and MPTP toxicity in a microglia-dependent manner (Li et al., 2009). In a rodent model of chronic neuroinflammation with selective SN vulnerability, nimodipine reduced microglia activation in the SN pars compacta with no neuroprotective benefit to dopaminergic neurons (Hopp, Royer, et al., 2015). Recently, the first study using genetic approaches to reduce L-VDCC expression on microglia demonstrated an increase in neuroinflammation and potentiation of neurotoxicity following knockdown of Cav1.2 on microglia (Wang et al., 2019), in contrast to numerous other studies that found pharmacological targeting of L-VDCCs was broadly neuroprotective and anti-inflammatory. Overall, these data demonstrate an important role of L-VDCCs during PD on both neurons and microglia of the SN.

7.4 |. Huntington’s disease

Unlike AD and PD, the genetic etiology of HD is clear: a CAG repeat expansion in the huntingtin gene (Htt) leads to an autosomal-dominant neurodegenerative disease characterized by progressive, uncontrollable choreiform motor defects and cognitive and psychiatric impairments. Neuropathologically, HD is characterized by inclusion bodies composed of mutant huntingtin (mHTT) within neurons, with GABAergic medium spiny neurons (MSNs) of the striatum most vulnerable to pathology (Vonsattel & DiFiglia, 1998). Huntingtin is expressed ubiquitously throughout the body and is involved in intracellular organelle trafficking, transcriptional regulation, and antiapoptotic activity (Schulte & Littleton, 2011). Specifically, mHTT dysregulates transcription of several genes related to Ca²⁺ homeostasis, leading to defects in SOCE (Czeredys, Gruszczynska-Biegala, Schacht, Methner, & Kuznicki, 2013) which in turn leads to synapse loss in HD models (Wu et al., 2016). Furthermore, MSNs with mHTT are hypersensitive to glutamate-mediated Ca²⁺ toxicity due to sensitization of ER IP3Rs and mitochondrial Ca²⁺ overload (Tang et al., 2005). While the α subunit of L-VDCCs does not interact with mHTT as measured by co-immunoprecipitation (Swayne et al., 2005), it does interact with the α2δ L-VDCC subunit (Kaltenbach et al., 2007). We posit that L-VDCC activity may be involved in Ca²⁺ dysregulation observed in HD due to the relationship between L-VDCCs and SOCE (Park et al., 2010; Wang et al., 2010), although there is a dearth of studies reporting L-VDCC activity during HD. In a Drosophila model of HD, the deletion of the L-VDCC gene, Dmca1D, suppresses mHTT-associated neuronal phenotypes (Romero et al., 2008). In muscle from the R6/2 model of HD, L-VDCCs display persistent, low-level increases in Ca²⁺ influx, leading to Ca²⁺ overload (Braubach et al., 2014). In the YAC128 mouse model of HD, the L-VDCC antagonist nifedipine is mildly neuroprotective in vivo (Hong et al., 2015). A recent study performed a more complete analysis L-VDCCs in the BACHD mouse model of HD and found increases in Cav1.2 (but not Cav1.3) protein but not mRNA in the cortex, increases in cortical L-VDCC Ca²⁺ currents in vitro, and neuroprotection by the L-VDCC antagonists isradipine and nifedipine on cortical neurons during glutamate toxicity in vitro (Miranda et al., 2019).

Microglia are involved in HD pathology, although relatively few studies have examined their role closely (Creus-Muncunill & Ehrlich, 2019; Möller, 2010). Microglia in HD contain mHTT inclusions (Simmons et al., 2007) but the expression of mHTT in microglia is neither necessary nor sufficient to cause an HD phenotype (Petkau et al., 2019) despite inducing pro-inflammatory gene expression in microglia (Crotti et al., 2014). Since Ca²⁺ influx into microglia is known to regulate their phenotype (discussed previously), it is possible that mHTT-induced changes in microglial gene expression are due to the Ca²⁺-dysregulating effects of mHTT. Further examination of Ca²⁺ dysregulation in microglia during HD would illuminate additional cell-autonomous effects of mHTT.

7.5 |. Neuropsychiatric disorders

Neuropsychiatric disorders lead to dysfunction in mood, affect, cognition, communication, and day-to-day function and are the leading cause of disability in the United States (Murray et al., 2013). Despite their diverse symptoms, genome-wide association studies linked polymorphisms in CACNA1C, the gene that encodes Cav1.2, to both neurodevelopmental disorders such as ASD and neuropsychiatric disorders such as schizophrenia, major depression, and bipolar disorder (Green et al., 2010; Roussos et al., 2014; Smoller et al., 2013) which echoes, in part, the neuropsychiatric phenotype observed in the L-VDCC channelopathy TS (Dixon et al., 2012). The CACNA1C mutations in TS and some polymorphisms for neuropsychiatric disorders lead to toxic L-VDCC gain-of-function due to changes in channel properties or gene expression (Kamijo et al., 2018; Liao & Soong, 2010). Gain-of-function polymorphisms in the Cav1.3 gene, CACNA1D, are also associated with ASD (Pinggera et al., 2015). The CACNA1C neuropsychiatric risk single-nucleotide polymorphism (SNP) rs1006737 has been studied extensively in humans and has been associated with alterations in gray matter volume in several anatomical regions, amygdala activity, and hippocampal functional coupling activity by neuroimaging; defects in verbal fluency, working memory, startle response, and attention by functional testing; and increased neuroticism, paranoia, depression, and anxiety by personality testing (Bhat et al., 2012; Kabir, Lee, & Rajadhyaksha, 2016).

Transgenic rodent models demonstrate some behavioral defects that parallel humans with CACNA1C SNPs (Bhat et al., 2012; Kabir et al., 2016). With respect to “autism-like” behaviors, CACNA1C heterozygous mice show mild defects in social behavior but very few changes in exploratory, anxiety-like, and learning behaviors (Kabitzke et al., 2018) while CACNA1C haploinsufficient rats do display mild defects in prosocial vocalizations without other overt behavioral phenotypes and even mild improvements in spatial memory function (Braun et al., 2018; Kisko et al., 2018; Michels et al., 2019). Haploinsufficient CACNA1C mice demonstrated reduced anxiety, reduced locomotion in the amphetamine-induced hyperactivity model of psychosis, and reduced depressive-like behavioral phenotypes (Dao et al., 2010). Prefrontal cortex (PFC) specific deletion of CACNA1C post-embryonically or developmentally in the dorsal raphe nuclei also reduces anxiety-like behavior and depressive-like phenotypes (Kabir, Lee, et al., 2017; Lee et al., 2012). These behavioral changes are consistent with the idea that neuropsychiatric risk SNPs in CACNA1C are gain-of-function. However, other studies demonstrate conflicting data. Developmental deletion of glutamatergic neuron or neural-lineage CACNA1C deletion produces cognitive deficits, reduced sociability, hyperactivity, and anxiety (Dedic et al., 2018) in contrast to the behavioral phenotype produced by global CACNA1C heterozygosity or even post-embryonic knockout (Dedic et al., 2018; Kabir, Lee, et al., 2017). Furthermore, selective deletion of CACNA1C in serotonergic neurons of the dorsal raphe nuclei increases immobility during the forced swim test and increase anxiety during the open field test (Ehlinger & Commons, 2019). Like the data with CACNA1C-lacking mice, homozygous loss of CACNA1D also appears to induce an antidepressant and anxiolytic phenotype (Busquet et al., 2010). Similarly, treatment with the dihydropyridine nimodipine has antidepressant and anxiolytic effects in mice (Moreno, Rolim, Freitas, & Santos-Magalhães, 2016; Moreno et al., 2014).

Genetic predisposition to neuropsychiatric illness is known to interact with environmental risk factors such as stress (Syed & Nemeroff, 2017) and the rs73249708 and rs116625684 CACNA1C SNPs interact with adult trauma to predict depression scores (Dedic et al., 2018). L-VDCCs and stress also interact in animal models, leading to changes in stress-induced behavioral phenotypes. Heterozygous CACNA1C global deletion induces susceptibility to the detrimental behavioral effects of chronic social defeat stress, while glutamatergic neuron-specific deletion during adulthood was protective (Dedic et al., 2018) and nucleus accumbens-specific deletion during adulthood was detrimental (Terrillion et al., 2017), suggesting region- and developmental-specific effects of CACNA1C activity on stress responsivity. The type of stress also appears to be differentially affected by CACNA1C deletion, suggesting divergent pathways of social and footshock stress via CACNA1C: global CACNA1C heterozygosity is protective against footshock stress-induced behavioral deficits (Bavley, Fischer, Rizzo, & Rajadhyaksha, 2017) as is treatment with the dihydropyridine L-VDCC antagonist nisoldipine (Verma, Bali, Singh, & Jaggi, 2016).

Notably, the detrimental effects of stress are driven by neuroinflammatory signaling pathways in part mediated by microglia (Weber, Godbout, & Sheridan, 2017) and microglia activation is observed following all types of stress (Calcia et al., 2016) suggesting a crucial role for microglia phenotype in neuropsychiatric disorders. While several human neuroimaging studies have identified physiological effects of CACNA1C mutations (Bhat et al., 2012) and microglia activation (Doorduin et al., 2009; Notter et al., 2018; Suzuki et al., 2013) independently in neuropsychiatric disorders, none have examined whether there is a relationship between CACNA1C SNPs and microglia phenotype in these diseases. Furthermore, microglia phenotype has not been examined in any of the rodent models described here, and microglia-specific CACNA1C or CACNA1D deletion has not been examined in the context of mouse models of neuropsychiatric disorders, despite the characteristic immune deficiency found in TS and the emerging role of immune dysregulation in ASD, depression, bipolar disorder, and schizophrenia (Tay et al., 2017). Perhaps the inability of neural-specific deletion to replicate the effects of L-VDCC antagonist treatment or global heterozygosity suggests the potential involvement of another cell type such as microglia. Current treatments for neuropsychiatric disorders such as antipsychotics and SSRI antidepressants can reduce microglia activation via Ca²⁺-dependent mechanisms (Horikawa et al., 2010; Mizoguchi, Kato, Horikawa, & Monji, 2014), supporting a role for both regulation of microglia activation and regulation of microglia Ca²⁺ in the treatment of neuropsychiatric diseases.

8 |. L-VDCC ANTAGONIST DRUG DISCOVERY FOR CNS DISORDERS

Numerous drugs are available that target L-VDCCs due to the important role of these channels in mediating hypertension. Several classes of L-VDCC antagonists were developed in the 1960s through the 1980s, starting with diphenylmethylpiperazines, phenylalkylamines (e.g., verapamil), benzothiazepines (e.g., diltiazem), and finally the widely used dihydropyridines (Godfraind, 2017). The dihydropyridine BayK8644 acts as an L-VDCC activator. The α1 subunit contains the dihydropyridine and other Ca²⁺ channel-blocking drug-binding sites; drug binding interferes with the voltage-dependent cycling of the channel (Zamponi, Striessnig, Koschak, & Dolphin, 2015). Dihydropyridines have increased affinity for the inactivated channel while phenylalkylamines and benzothiazepines preferentially bind the channel when it is open. Despite these differences, CCBs have the common properties of safety and tolerability with limited side effects at therapeutic doses for hypertension and cardiac ischemia. Currently, the only approved CNS indication for CCBs is for the dihydropyridine nimodipine, which is approved for neuroprotection after subarachnoid hemorrhage, putatively due to prevention of vasospasm (Mijailovic et al., 2013) rather than direct neuroprotection. Recent studies have tested two different dihydropyridine CCBs in two CNS indications: PD and AD. Phase II safety and tolerability studies in PD and AD patients showed that isradipine (Simuni et al., 2010) and nilvadipine (Kennelly et al., 2011), respectively, were safe and well-tolerated. Human retrospective studies had previously shown that treatment with dihydropyridines led to significant risk reduction in PD (Ritz et al., 2009) and dementia (Forette et al., 2002). Despite promising preclinical data in animal models of PD (Guzman et al., 2018; Ilijic et al., 2011) and AD (Paris et al., 2011), Phase III trials of isradipine and nilvadipine recently failed to improve or slow progression of PD (Simuni, 2019) and AD (Lawlor et al., 2018), respectively. Similarly, it is not clear whether L-VDCC antagonist drugs are effective in the treatment of bipolar disorder due to inadequate clinical trials, despite extensive clinical use and a clear genetic link between CACNA1C polymorphisms in bipolar disorder (Cipriani et al., 2016). A randomized, double-blind, placebo-controlled trial examining the L-VDCC antagonist nicardipine for mood stabilization was recently initiated (Atkinson et al., 2019).

There are several potential explanations as to why trials with L-VDCC antagonists have thus far failed. First, these drugs may simply not be efficacious for the treatment of AD, PD, or neuropsychiatric disorders, or these drugs may be preventative rather than interventional. There could be issues with drug dosing, brain penetration, or target engagement at the appropriate brain region or cell type. For example, isradipine is more selective for Cav1.2 over Cav1.3 in SN neurons, so it might not be able to reverse (Ortner et al., 2017) the Cav1.3-dependent vulnerability to neurotoxicity (Chan et al., 2007) at the dose given. Additionally, isradipine may have off-target toxic effects that obscure its benefits, for example a recent study in mice demonstrated that Cav1.2 ablation on microglia in mouse models of PD may promote SN neurotoxicity in vivo (Wang et al., 2019). These studies highlight the need for enhanced understanding of channel subtypes and novel subtype-specific drug design and discovery.

Dihydropyridine antagonists have an increased affinity for inactivated channel states and are thus more selective for arterial smooth muscle compared to neurons due to the preferable channel inactivation kinetics of smooth muscle-specific splice variants. Furthermore, the Cav1.2 channel subtype is more sensitive to dihydropyridines than Cav1.3, which is dependent on subtype-specific channel properties: Cav1.3 channels activate at −55 mV membrane potential (peak −40 mV) while Cav1.2 channels activate at −35 mV (peak 0 mV; Xu & Lipscombe, 2001). Development of Cav1.3-specific compounds is an active area of research for CNS indications such as PD; this is also relevant to the development of novel anti-inflammatory drugs since Cav1.3 is expressed at higher levels than Cav1.2 on microglia (Zhang et al., 2014). Currently, only one Cav1.3-selective agent has been described as a novel therapeutic strategy for PD (Kang et al., 2012). Dihydropyridine tissue selectivity is also related to preferential binding to inactivated channels which are more likely to occur on cells that have a less negative resting membrane potential (RMP) and longer, slower depolarization (e.g., arterial smooth muscle, RMP = −50 to −60 mV) compared to cells with more negative membrane potentials that fire large, fast depolarizations (e.g., neurons, RMP = −70 mV) (Helton, Xu, & Lipscombe, 2005). Notably, microglia have a less negative membrane potential (−26 mV quiescent, −42 mV LPS-activated; Boucsein, Kettenmann, & Nolte, 2000; Chung, Joe, Soh, Lee, & Bang, 1998) suggesting that microglia L-VDCCs may have more favorable dihydropyridine binding compared to neurons; however, this depends on the characteristics of the currently uncharacterized microglia-specific L-VDCC splice variants (Zhang et al., 2014).

As described above, current dihydropyridine L-VDCC targeting drugs have been the focus of a resurgence of research for various CNS conditions with limited treatment options, such as AD and PD. The protective effects of dihydropyridines may be due to antagonism on their putative targets, the L-VDCCs, or due to blockade of unrelated “off-target” pathways, particularly on microglia (Chiozzi et al., 2019; Schampel et al., 2017). For example, a recent study confirmed that nimodipine and other dihydropyridines could reduce all Aβ effects on microglia (e.g., inflammasome activation, IL-1β release, mitochondrial dysfunction) independent of L-VDCCs (Chiozzi et al., 2019). These authors posit that nimodipine may reduce Aβ toxicity by increasing the rate of endosome-lysosome fusion (autophagic flux), leading to increased Aβ degradation; this mechanism of dihydropyridines has been previously hypothesized based upon unreported data on autophagic flux proteins (Anekonda & Quinn, 2011). Indeed, drug screens have identified numerous L-VDCC targeting drugs across different structural classes (verapamil, loperamide, amiodarone, nimodipine, and nitrendipine) as stimulators of autophagy, which may be due to modulation of Ca²⁺ levels or due to as-of-yet determined off-target effects (Vakifahmetoglu-Norberg, Xia, & Yuan, 2015). L-VDCC antagonists drugs may produce effects via action on NADPH oxidase (Li et al., 2009), heat-shock protein modulation (Roe et al., 2018), suppression of spleen tyrosine kinase (Paris et al., 2014), antioxidant effects (Mason, Mak, Trumbore, & Mason, 1999), or other ion channels such as voltage-dependent Na⁺ channels, Ca²⁺-dependent K⁺ channels, or voltage-gated K⁺ channels (Zhang & Gold, 2009). Furthermore, the unexpected pro-neurotoxic effect of genetic ablation of Cav1.2 on microglia (Wang et al., 2019) also suggests that L-VDCC antagonists may act via alternative mechanisms in different models.

9 |. IMPLICATIONS FOR FUTURE RESEARCH

The development of agents that specifically target L-VDCC subtypes and variants represents a key advance needed in the field. The development of Cav1.3-targeting agents is a crucial first step, followed by splice variant-specific agents. Cav1.3 is a better target for PD and may also represent a better target for novel anti-inflammatory agents targeted to microglia. Recent advances in long-read sequencing have revealed the extensive number and diversity of variants present in the brain (Clark et al., 2019), some of which are most certainly specific to cells of interest, including subtypes of neurons, microglia, and astrocytes. With these splice variant sequences, we can characterize the electrophysiological and pharmacodynamic properties of each of these variants and identify their cell type-specific distribution. In addition to investigating the biophysical and pharmacological properties of these splice variants, it will be critical to evaluate the role of these variants on transcription, since the C terminals of Cav1.2 and Cav1.3 are transcription factors (Gomez-Ospina et al., 2006; Lu et al., 2015). There are tremendous opportunities for drug development targeting L-VDCCs, but extensive additional basic research is required for advancement.

When these new compounds eventually move to the clinic, the identification of appropriate patient populations and appropriate biomarkers will be vital to the success of these hypothetical compounds. To identify patients who may respond to specific L-VDCC targeting therapies, personalized medicine approaches such as the generation of patient-specific induced pluripotent stem cells (IPSCs) would provide insight into cell type-specific L-VDCC genotype and phenotype prior to enrollment. With respect to biomarkers, there are different options depending on the cell target. For microglia, recent advances in PET imaging allow for in vivo evaluation of neuroinflammation (Shen, Bao, & Wang, 2018). For neurons, magnetoencephalography (Mandal, Banerjee, Tripathi, & Sharma, 2018) allows for in vivo measurement of L-VDCC-related neuronal discharges. While blood-based biomarkers are the gold standard, further research into the downstream, specific signaling cascades related to L-VDCC target engagement in the brain will be required.

10 |. CONCLUSION

L-VDCC antagonists are historically indicated as neuroprotective agents in a variety of CNS conditions, including dementia, AD, PD, HD, and neuropsychiatric disorders. While the action of L-VDCCs and associated pharmaceuticals on excitable cells is well-characterized, the nature and purpose of these channels on non-excitable cells, such as microglia, is underdeveloped and is likely a key characteristic of their neuroprotective effects. We present a model of microglia-mediated neuroprotection by L-VDCC antagonist drugs via the reduction of microglia immune activation in Figure 1. However, the mechanism driving this neuroprotection is not clear and may be due to action on L-VDCCs themselves or via other unknown drug targets. Our current understanding of L-VDCCs suggests that tissue-specific drugs may be an exciting avenue to promote neuroprotective and anti-inflammatory actions through varied cell types including microglia. Overall, L-VDCCs are an old target ripe for the development of novel research and treatment strategies.

Significance.

Microglia, the immune cells of the brain, play an emerging central role in several brain diseases, such as Alzheimer’s disease (AD), Parkinson’s disease, Huntington’s disease, and neuropsychiatric disorders. Intracellular calcium (Ca²⁺) is known to regulate microglia transformation from a resting state to an activated, immune-effector state. Chronically activated microglia, such as those observed during conditions such as AD, are neurotoxic. Importantly, several drugs targeting L-type Ca²⁺ channels are neuroprotective and anti-inflammatory. This review closely examines the literature on L-type Ca²⁺ channels on microglia and drugs that target them and discusses potential future drug development. A deeper understanding of signaling pathways that regulate microglia function will reveal new avenues of treatment for brain disorders where microglia play a central role. This is especially important for central nervous system disorders where there are currently no disease-modifying therapies available.