A tale of many families: calcium channels in plant immunity

Abstract

Plants launch a concerted immune response to dampen potential infections upon sensing microbial pathogen and insect invasions. The transient and rapid elevation of the cytosolic calcium concentration [Ca²⁺]_cyt is among the essential early cellular responses in plant immunity. The free Ca²⁺ concentration in the apoplast is far higher than that in the resting cytoplasm. Thus, the precise regulation of calcium channel activities upon infection is the key for an immediate and dynamic Ca²⁺ influx to trigger downstream signaling. Specific Ca²⁺ signatures in different branches of the plant immune system vary in timing, amplitude, duration, kinetics, and sources of Ca²⁺. Recent breakthroughs in the studies of diverse groups of classical calcium channels highlight the instrumental role of Ca²⁺ homeostasis in plant immunity and cell survival. Additionally, the identification of some immune receptors as noncanonical Ca²⁺-permeable channels opens a new view of how immune receptors initiate cell death and signaling. This review aims to provide an overview of different Ca²⁺-conducting channels in plant immunity and highlight their molecular and genetic mode-of-actions in facilitating immune signaling. We also discuss the regulatory mechanisms that control the stability and activity of these channels.

Calcium channels play vital roles in plant immunity.

Introduction

Plants are exposed to myriads of invading organisms with different lifestyles, including viruses, bacteria, fungi, oomycetes, nematodes, and insects. Plants have evolved physical barriers, including wax layers, a cell wall, cutin, and callose, which can be strengthened during infection (Thordal-Christensen, 2003). Bacterial and filamentous pathogens enter plant leaf apoplasts through stomata, hydathodes, or wound sites. Thus, stomata and hydathodes are critical in restricting pathogen entry into plants (Melotto et al., 2017; Cerutti et al., 2019). To counteract pathogen colonization, plants deploy a two-tiered immune system (Jones and Dangl, 2006; van der Burgh and Joosten, 2019; Zhou and Zhang, 2020). Additionally, a local infection can trigger systemic acquired resistance (SAR) to protect distal tissues against subsequent attacks by a broad spectrum of pathogens (Fu and Dong, 2013; Kachroo and Kachroo, 2020; Vlot et al., 2021).

Among the common and early cellular responses in plant immunity is the transient elevation of cytosolic calcium concentration [Ca²⁺]_cyt. Calcium serves as both a nutrient and a signal in cell growth, immunity, symbiosis, and systemic signaling in all eukaryotes (Luan and Wang, 2021). The free Ca²⁺ concentration in the apoplast is ∼10,000-fold higher than that in the cytoplasm at the resting state (Demidchik et al., 2018). Thus, the opening of calcium channels leads to a dynamic [Ca²⁺]_cyt increase, where it acts as a second messenger for almost every aspect of plant growth, development, and stress responses (Tian et al., 2020b). Plants generate stimulus-specific Ca²⁺ fluxes and decoding systems that transduce specific downstream signaling events (Yuan et al., 2017; Lenzoni et al., 2018; Tian et al., 2020b). Ca²⁺ influx is mainly conducted by calcium channels. In the context of recent breakthroughs of classical calcium channels in plant immunity and identification of immune receptors as noncanonical calcium channels, in this review, we aim to provide an overview of the roles of Ca²⁺-conducting channels in plant immunity and highlight their molecular and genetic mode-of-actions in immune signal transduction. We also discuss the regulatory mechanisms that control the stability and activity of these channels.

Plant immunity

The first layer of the plant immune system is initiated by the recognition of microbe/herbivore-associated molecular patterns (MAMPs/HAMPs) or plant-derived endogenous danger‐associated molecular patterns (DAMPs) and phytocytokines by pattern recognition receptors (PRRs), culminating in pattern-triggered immunity (PTI) (Yu et al., 2017; Albert et al., 2020; Zhou and Zhang, 2020; DeFalco and Zipfel, 2021; Tanaka and Heil, 2021). Plant PRRs are receptor-like kinases (RLKs) and receptor-like proteins with an ectodomain, a transmembrane domain, and an intracellular kinase domain (de Azevedo Manhães et al., 2020; Dievart et al., 2020). To mount a successful infection, pathogens deliver effector proteins into the apoplast or host cells to interfere with the immune system and/or host physiology (Dou and Zhou, 2012; Rocafort et al., 2020). During the evolutionary arms race with pathogens, plants have evolved a second layer of defense, effector-triggered immunity (ETI), mediated by nucleotide-binding domain leucine-rich repeat proteins (NBS-LRRs or NLRs) that recognize pathogen effectors directly or effector-induced perturbation (Cui et al., 2015; Lolle et al., 2020). ETI usually leads to localized cell death, known as the hypersensitive response (HR).

Despite the differences in receptor activation and early signaling mechanisms, the boundaries between PTI and ETI become increasingly blurred (Bjornson and Zipfel, 2021a). PTI and ETI trigger a series of overlapping immune responses, including Ca²⁺ influx, reactive oxygen species (ROS) burst, activation of mitogen-activated protein kinases, transcriptional reprogramming, and phytohormone production (Tsuda and Katagiri, 2010; Peng et al., 2018). However, the strength, kinetics, and duration of these events in PTI and ETI show distinct features. Additionally, some core PTI components are required for ETI and vice versa, and PTI and ETI mutually potentiate each other (Pruitt et al., 2020; Tian et al., 2020a; Ngou et al., 2021; Yuan et al., 2021). Plasma membrane (PM)-localized NADPH oxidases, which produce extracellular ROS, contain an EF-hand motif and are regulated by Ca²⁺ (Kadota et al., 2015; Marcec et al., 2019), implicating an intertwined relationship between Ca²⁺ and ROS production in the interplay of PTI and ETI (Figure 1).

Figure 1.

Calcium signaling in plant immunity. Sensing of MAMPs/HAMPs or plant-derived endogenous DAMPs by PM-localized PRRs elicits PTI. PRRs recruit co-receptors and activate receptor-like cytoplasmic kinases (RLCKs), which subsequently phosphorylate PM-resident NADPH oxidases and calcium channels to trigger ROS burst and Ca²⁺ influx, respectively. Pathogen effector proteins delivered through the type three secretion system are recognized directly or indirectly by intracellular nucleotide-binding domain leucine-rich repeat proteins (NBS-LRRs or NLRs), cumulating into ETI. ETI induces a sustained second Ca²⁺ influx peak not detected in PTI. ETI also triggers a NADPH oxidase-dependent ROS burst, which requires core components of PTI, and upregulates PTI components, implying mutual potentiation of PTI and ETI through ROS and likely Ca²⁺ signals. Rapid Ca²⁺ waves are generated in distal tissues upon wounding and herbivory attacks. Cytosolic Ca²⁺ could be derived from the apoplast and internal stores, such as vacuole and chloroplast.

Calcium signature and signaling

Specific Ca²⁺ signatures in different branches of immunity vary in multiple aspects (Demidchik et al., 2018). PTI activation leads to a fast and transient Ca²⁺ influx from the apoplast and possible intracellular Ca²⁺ stores within minutes (Wu et al., 2014). Different types and dosages of MAMPs lead to Ca²⁺ signals with specific peak time, amplitude, and duration (Yuan et al., 2017). On the other hand, ETI activation induces a much prolonged and sustained Ca²⁺ influx. Bacterial Pseudomonas syringae carrying the effector AvrRpm1 elicited a transient [Ca²⁺]_cyt increase that peaked at 10 min after infection followed by a sustained rise that peaked around two hours after infection (Grant et al., 2000). Furthermore, treatment with lanthanum chloride (LaCl₃), an extracellular calcium antagonist, and ruthenium red (RR), an inhibitor of Ca²⁺ release from intracellular compartments, suppressed ETI-associated HR and defense gene expression, suggesting that both extracellular and intracellular Ca²⁺ release contributes to ETI (Grant et al., 2000; Gao et al., 2013). Interestingly, the influx of extracellular Ca²⁺, not intracellular Ca²⁺, is essential for MAMP Pep-13-triggered immune responses as Pep-13-treated parsley (Petroselinum crispum) cells maintained the normal defense responses in the presence of RR (Blume et al., 2000). Insect infestation triggers a [Ca²⁺]_cyt elevation within seconds at the wound site and subsequently propagates from the local tissues to the distal leaves through the vasculature (Hilleary and Gilroy, 2018).

The spatial and temporal [Ca²⁺]_cyt changes are decoded by an array of Ca²⁺ sensors, including conserved calmodulins (CaM), plant-specific calmodulin-like proteins, calcium-dependent protein kinases (CDPKs or CPKs), calcineurin B-like proteins (CBLs) coupled with CBL-interacting protein kinases and NADPH oxidases (Figure 2). Concerted actions convey Ca²⁺ signals into specific cellular responses (Yuan et al., 2017; Bredow and Monaghan, 2019; Tang et al., 2020; Tian et al., 2020b).

Figure 2.

Diverse calcium channels involved in plant immunity. Different types of ion channels, such as CNGCs, GLRs, and OSCA channels, TPC1, and ANNs have been shown to conduct cytosolic Ca²⁺ influx in plant immunity. Recently, disease resistance NLR proteins have been implicated in forming a large complex coined as NLR resistosomes, which function as noncanonical ion channels conducting cytosolic calcium influx in plant ETI. TPC1 is a vacuole-localized channel, and ANNs are cytosolic proteins, but the binding of Ca²⁺ facilitates their association with membrane lipids, whereas the others are PM localized. Distinct cytosolic Ca²⁺ signals conducted by the coordinated action of calcium channels in a cell type- and tissue-specific manner are decoded by Ca²⁺ sensors, including highly conserved CaMs, plant-specific CML proteins, CDPKs or CPKs, CBLs coupled with their interacting kinases (CIPKs) and NADPH oxidases to relay signaling into specific cellular and physiological immune responses.

Diverse calcium channels in plant immunity

Calcium channels have been implicated in conducting Ca²⁺ in plant immunity. These include classical ion channels, such as cyclic nucleotide-gated ion channels (CNGCs), glutamate receptor-like proteins (GLRs), reduced hyperosmolality-induced [Ca²⁺]_cyt increase channels (OSCAs), two-pore channels (TPCs) and annexins (ANNs), as well as the recently identified noncanonical ion channels formed by NLR (resistosome) complexes (Figure 2 and  Table 1).

Table 1.

Ion channels with roles in Ca²⁺ conductivity and plant immunity

Class	Name	Species	Function in plant immunity	Reference
CNGC	CNGC2 CNGC4	Arabidopsis	Positive regulator in ETI-associated HR and PTI-induced Ca2+ influx; BIK1 phosphorylates and activates CNGC4, which is otherwise suppressed by CaM.	Clough et al., 2000; Balagué et al., 2003; Jurkowski et al., 2004; Ma et al., 2012; Tian et al., 2019
	CNGC11 CNGC12	Arabidopsis	Chimeric CNGC11-CNGC12 induced Ca2+-dependent cell death; Positive regulator in ETI disease resistance.	Yoshioka et al., 2006; Moeder et al., 2011
	CNGC19 CNGC20	Arabidopsis	BAK1/BKK1 phosphorylates CNGC19/20 to maintain Ca2+ homeostasis for cell survival; CNGC19 positively regulates JA signaling and glucosinolate accumulation upon herbivore attack.	Meena et al., 2019; Yu et al., 2019; Zhao et al., 2021a
	OsCNGC9	Rice	OsRLCK185 phosphorylates OsCNGC9 in chitin-mediated fungal resistance.	Wang et al., 2019c
	LjCNGCIVA	Lotus	Involved in root development and infection by nitrogen-fixing rhizobia.	Chiasson et al., 2017
	MtCNGC15a-c	Medicago	Regulates nuclear Ca2+ oscillation in symbiotic interaction with rhizobia.	Charpentier et al., 2016
GLR	GLR	Radish	Heterologous expression in Arabidopsis enhances JA-responsive gene expression and resistance to B. cinerea.	Kang et al., 2006
	GLR2.7 GLR2.8 GLR2.9	Arabidopsis	Positively regulates PTI-induced Ca2+ influx and disease resistance.	Bjornson et al., 2021b
	GLR3.3 GLR3.6	Arabidopsis	Positively regulates DAMP oligogalacturonide- and aphid-induced Ca2+ elevation and propagation of the systemic wound signal and JA-responsive gene expression. Positively regulates disease resistance against Pseudomonas syringae pv. tomato DC3000 and H. arabidopsidis, but is not required for resistance against B. cinerea.	Kwaaitaal et al., 2011; Li et al., 2013, 2020; Manzoor et al., 2013; Mousavi et al., 2013; Vincent et al., 2017; Toyota et al., 2018
	SlGLR3.5	Tomato	Required for root-to-shoot systemic transmission of electrical signals and JA increase in leaves in response to root-knot nematode infection.	Wang et al., 2019a
	GbGLR4.8	Cotton	Required for fungal Fusarium elicitor-induced Ca2+ influx and disease resistance.	Liu et al., 2021
OSCA	OSCA1.3 OSCA1.7	Arabidopsis	BIK1 phosphorylates OSCA1.3/1.7 to mediate flg22-induced Ca2+ influx in stomata.	Thor et al., 2020
TPC	TPC1	Arabidopsis	Positive regulator in wounding and herbivory-triggered local and systemic Ca2+ signals, and resistance to aphids.	Kiep et al., 2015; Vincent et al., 2017
	OsTPC1	Rice	Positive regulator in fungal elicitor-induced defense gene expression, MAPK activation, and cell death.	Kurusu et al., 2005
Annexin	BjANN1	Brassica juncea	Ectopic expression in tobacco enhanced defense gene expression and resistance to the oomycete Phytophthora parasitica.	Jami et al., 2008
	ANN1 ANN2 ANN4	Arabidopsis	Positively regulates chitin-induced defense response; Required for DAMP extracellular (e)ATP in roots but not leaves; Activate insect-triggered systemic Ca2+ elevation and defense; Positive regulator in resistance to nematode Meloidogyne incognita; Nonselective loading of sRNA into extracellular vesicles and positive regulator to Botrytis cinerea.	Espinoza et al., 2017; Zhao et al., 2019; He et al., 2021; Malabarba et al., 2021; Mohammad-Sidik et al., 2021
	ANN8	Arabidopsis	Positive regulator in RPW8.1-mediated HR and disease resistance.	Zhao et al., 2021b
NLR resistosome	ZAR1	Arabidopsis	Required for cytoplasmic Ca2+ influx and HR.	Bi et al., 2021
	NRG1.1	Arabidopsis	Required for cytoplasmic Ca2+ influx and HR.	Jacob et al., 2021
	ADR1	Arabidopsis	Required for cytoplasmic Ca2+ influx and HR.	Jacob et al., 2021

CNGCs

Named based on sequence similarity to animal CNGCs, plant CNGCs mediate Ca²⁺ signaling in plant developmental regulation and stress responses (DeFalco et al., 2016b; Dietrich et al., 2020; Jarratt-Barnham et al., 2021).

Domain organization

Plant CNGCs belong to the superfamily of voltage-gated channels and are nonselective cation channels with some members also conducting other cations, such as K⁺ (Demidchik et al., 2018; Dietrich et al., 2020; Jarratt-Barnham et al., 2021). The Arabidopsis thaliana genome encodes 20 CNGCs that form four clades (I–IV), with clade IV being subdivided into IVa and IVb (Mäser et al., 2001). Similar to animal CNGCs, which form hetero-tetramers, there is increasing evidence for plant CNGC channels also being formed by heterogeneous subunits (Chin et al., 2013; Pan et al., 2019; Tian et al., 2019; Yu et al., 2019).

Plant CNGCs consist of six transmembrane domains and cytosolic amino (N)- and carboxyl (C)-terminal domains (Jegla et al., 2018; Dietrich et al., 2020). The C-terminal domain contains a C-linker, followed by regulatory domains, including the cyclic nucleotide-binding domain and the partially overlapping CaM-binding domain (CaMBD) (Kaplan et al., 2007). Additionally, a conserved Ile-Gln (IQ)-type CaMBD exists adjacent to the canonical CaMBD, and mounting evidence suggests the positive regulation by CaM via the IQ domain (DeFalco et al., 2016a; Fischer et al., 2017; Tian et al., 2019).

Roles of CNGCs in plant immunity

The best-studied CNGCs concerning immunity are CNGC2 and CNGC4. The Arabidopsis cngc2 and cngc4 mutants were named defense no death 1 (dnd1) and dnd2, respectively (Clough et al., 2000; Jurkowski et al., 2004). The cngc4 mutant was also named HR-like lesion mimic (hlm1) (Balagué et al., 2003). Both cngc2 and cngc4 mutants display autoimmune phenotypes, including conditional spontaneous cell death, increased salicylic acid (SA) accumulation, and enhanced resistance against biotrophic and necrotrophic pathogens. This corresponds with the constitutive activation of SA and jasmonic acid (JA) pathways in these mutants (Clough et al., 2000; Jurkowski et al., 2004; Genger et al., 2008).

In addition, CNGC2 and CNGC4 have been connected to PTI responses (Ma et al., 2012; Tian et al., 2019). They likely form a heteromeric channel that is kept inactive by binding of CaM (Chin et al., 2013; Tian et al., 2019). Upon MAMP perception, the PRR-associated receptor-like cytoplasmic kinase (RLCK) BOTRYTIS-INDUCED KINASE 1 (BIK1) phosphorylates CNGC4 leading to the de-repression of the channel (Tian et al., 2019), which is consistent with earlier observations that BIK1 and PBL1 are required for MAMP-triggered Ca²⁺ influx (Li et al., 2014; Ranf et al., 2014). CaM binds CNGC4 at its IQ domain in a Ca²⁺-independent manner, similar to CNGC2 and CNGC12 (DeFalco et al., 2016a; Fischer et al., 2017; Tian et al., 2019).

As CNGC2/4 are Ca²⁺-conducting channels, [Ca²⁺]_cyt was expected to be reduced in these mutants. However, they over-accumulate apoplastic Ca²⁺ (Wang et al., 2017) and are under Ca²⁺ stress (Chan et al., 2008). Additionally, the dnd phenotype disappears when mutants are grown in a low level of Ca²⁺, including the attenuated response to MAMP treatment (Wang et al., 2017; Tian et al., 2019). Thus, there are likely other calcium channels, potential feedback regulations, and/or yet-to-be-discovered roles of CNGC2 in Ca²⁺ flux, sensing, and homeostasis (Finka et al., 2012; Chakraborty et al., 2021).

CNGC11 and CNGC12 are implicated in immunity via the autoimmune mutant constitutive expresser of PR genes 22 (cpr22), which bears a fusion of the N-terminus of CNGC11 and the C-terminus of CNGC12 (Yoshioka et al., 2006). The CNGC11-12 chimera is likely a mis-regulated channel that mediates constitutive Ca²⁺ influx, resulting in autoimmunity (Urquhart et al., 2007; Moeder et al., 2019). This may be partly due to the loss of CaMBD of CNGC12, which negatively regulates channel activity (DeFalco et al., 2016a). A CNGC12 truncation without CaMBD caused autoimmunity similar to the CNGC11-12 chimera (DeFalco et al., 2016a). Notably, CNGC11 is not an active Ca²⁺-conducting channel in Xenopus oocytes (Zhang et al., 2019), questioning whether it requires another subunit or other stimulus to be activated. The cngc11 and cngc12 mutants displayed partially compromised ETI, but not PTI responses (Moeder et al., 2011).

The rice (Oryza sativa) cell death and susceptible to blast 1 (cds1) mutation is caused by a loss-of-function of OsCNGC9, a homolog of Arabidopsis CNGC14 (Wang et al., 2019c). Unlike other autoimmune mutants, cds1 displays impaired resistance to the fungus Magnaporthe grisea, as well as reduced Ca²⁺ influx, oxidative burst, and PTI-related gene expression at the seedling stage. However, the mutant is more resistant after flowering, suggesting a growth stage-dependent regulation in this process (Wang et al., 2019c). OsCNGC9 is phosphorylated by OsRLCK185, which interacts with the chitin receptor complex (Wang et al., 2019c).

While RLCKs phosphorylate and activate certain CNGCs, phosphorylation by PRR co-receptor BRI1-ASSOCIATED KINASE 1 (BAK1) and its homolog SOMATIC EMBRYOGENESIS RECEPTOR KINASE 4 (SERK4) leads to destabilization of CNGC20 and 19 (Yu et al., 2019). CNGC19 and 20 form homo- and heteromeric Ca²⁺-conducting channels (Yu et al., 2019; Zhao et al., 2021a) that positively regulate bak1/serk4 mutant-mediated autoimmunity and cell death (Yu et al., 2019). Notably, CNGC19 and 20 have an unequal contribution in this process. The cngc20, but not cngc19 mutants, suppress the bak1/serk4-triggered autoimmunity. However, CNGC19 bears an additive effect with CNGC20 in regulating bak1/serk4 autoimmunity since the cngc19/20 mutant has a stronger autoimmune suppression function than cngc20 (Yu et al., 2019). Additionally, a recessive gain-of-function mutant of CNGC20 (cngc20-4) with a mutation in the S5 transmembrane region facing into the channel pore leads to constitutively elevated [Ca²⁺]_cyt (Zhao et al., 2021a). Notably, the cngc19 or 20 mutants did not affect plant PTI (Yu et al., 2019; Zhao et al., 2021a). These reports highlight the crucial role of maintaining Ca²⁺ homeostasis for cell survival.

CNGC19 also promotes defense against the insect Spodoptera litura. The cngc19 mutants display attenuated Ca²⁺ influx and vascular spread of local wound signals upon insect attack with reduced JA biosynthesis and glucosinolate production (Meena et al., 2019). In addition, CNGC19 contributes to the plant interaction with the endophytic fungus Piriformospora indica (Jogawat et al., 2020). This suggests that CNGC19 plays a role in the Ca²⁺ signaling triggered by different microorganisms and insects; yet, it awaits to be ascertained how CNGC19 is mechanistically linked to specific immune signaling pathways.

Several CNGCs, including BRUSH in Lotus japonicus and MtCNGC15a-c in Medicago truncatula, are involved in the symbiotic interaction with rhizobia (Charpentier et al., 2016; Chiasson et al., 2017). MtCNGC15a–c mediate nuclear Ca²⁺ oscillations during the interaction with rhizobia. They target the nuclear envelope and interact with DMI1 (DOES NOT MAKE INFECTIONS1), which was thought to conduct K⁺ (Charpentier et al., 2016); however, a recent structural study suggested that DMI is rather a Ca²⁺-regulated Ca²⁺ channel (Kim et al., 2019). So far, nuclear Ca²⁺ oscillations have primarily been associated with symbiotic interactions, while there is limited evidence for pathogen responses (Thor and Peiter, 2014; Keinath et al., 2015).

GLRs

Plant GLRs, homologs of mammalian ionotropic glutamate receptors (iGluRs), are ligand-gated ion channels regulating diverse aspects of plant physiology (Grenzi et al., 2021a). The Arabidopsis genome encodes 20 GLRs, belonging to three clades (Chiu et al., 2002). Similar to iGluRs, plant GLRs possess an extracellular amino-terminal domain, a bilobed clamshell-like ligand-binding domain, three complete and one partial transmembrane helices, and a cytoplasmic tail (Grenzi et al., 2021a). However, unlike iGluRs, which are gated by glutamate (Glu), plant GLRs are gated by multiple amino acids (Qi et al., 2006), implicating a broad ligand specificity for plant GLRs.

Members of GLRs function as Ca²⁺-permeable channels in plant immunity and wound responses (Forde and Roberts, 2014). The first report connecting a GLR to plant immunity came from heterologous expression of a radish (Raphanus sativus) GLR in Arabidopsis, which resulted in enhanced expression of JA-regulated genes and resistance against the fungus Botrytis cinerea (Kang et al., 2006). In Arabidopsis, GLR3.3 positively regulates disease resistance against Pseudomonas syringae pv. tomato DC3000 and Hyaloperonospora arabidopsidis but is not required for resistance against B. cinerea (Li et al., 2013; Manzoor et al., 2013). GLR3.3 is also involved in Ca²⁺ influx in response to the DAMP oligogalacturonide (Kwaaitaal et al., 2011). Genetic analyses showed that Ca²⁺ influx induced by green peach aphid feeding depends on BAK1 and GLR3.3/GLR3.6 (Vincent et al., 2017), implying a possible link between PRR complexes and GLRs. It is possible that similar to CNGCs, GLRs are directly activated by PRRs and associated proteins.

The function of GLRs in transducing long-distance signaling in response to wounding and herbivores has been well established (Mousavi et al., 2013; Toyota et al., 2018). Wounding or insect chewing induces the release of Glu or other amino acids into the apoplast, which either locally or systemically activate GLRs through direct binding. GLR3.3/GLR3.6 sense apoplastic Glu and trigger Ca²⁺ influx. Consequently, this signal propagates to distal leaves where it triggers systemic defense responses (Toyota et al., 2018). Propagation of the Ca²⁺ signal is abolished in the glr3.3 glr3.6 double mutant. In contrast, flg22 or oligogalacturonides do not initiate systemic Ca²⁺ increases, supporting the importance of GLRs for long-distance wound signaling (Toyota et al., 2018).

It remains unknown whether Glu itself is transmitted systemically upon wounding. A recent study indicates that GLR3.3 coordinates with the JA transporters JAT3/JAT4 to regulate systemic translocation and signaling of JA (Li et al., 2020). This may explain the reduced expression of wound-induced JA-response genes in the distal leaves of glr3.3 and glr3.6 mutants (Mousavi et al., 2013). Additionally, tomato (Solanum lycopersicum) SlGLR3.5 (homologous to Arabidopsis GLR3.3) is required for the root-to-shoot systemic transmission of electrical signals and JA increase in leaves in response to root-knot nematode infection (Wang et al., 2019a). Thus, Glu may function as a wound-induced DAMP that activates GLRs in mediating long-distance Ca²⁺ and JA signaling. This possibility is corroborated by the observations that overlapping genes are induced by Glu and MAMPs/DAMPs, and Glu induces defense responses locally and systemically (Goto et al., 2020).

Other GLR clade members also are implicated in plant immunity. GLR2.7/GLR2.8/GLR2.9 plays a role in MAMP-induced Ca²⁺ influx and bacterial disease resistance (Bjornson et al., 2021b). A genome-wide association study (GWAS) in cotton (Gossypium hirsutum) identified GhGLR4.8 as an atypical resistance gene to the fungus Fusarium oxysporum f. sp. vasinfectum (Fov) (Liu et al., 2021). GhGLR4.8 belongs to the GLR4 clade, a new subfamily widespread in angiosperms but absent in Brassicaceae (Chen et al., 2016). GhGLR4.8 is indispensable for Ca²⁺ influx induced by secreted proteins (SEPs) of Fov, suggesting that GhGLR4.8 functions as a Ca²⁺-permeable channel in cotton (Liu et al., 2021). Interestingly, co-infiltration of Fov SEPs and GhGLR4.8 induced HR, a hallmark of ETI (Liu et al., 2021). Future work needs to determine whether MAMPs or effectors in Fov SEPs induce GhGLR4.8-mediated Ca²⁺ influx and whether/how GhGLR4.8 might also be involved in long-distance Ca²⁺ signaling in response to Fov. It also remains to be determined whether different GLR clades have similar or distinct activation mechanisms.

OSCA channels

OSCA channels were independently identified using heterologous expression of Arabidopsis integral membrane proteins in Chinese Hamster Ovary cells for osmosensitive calcium conductance or calcium-imaging-based genetic screens for Arabidopsis mutants in response to osmotic stress (Hou et al., 2014; Yuan et al., 2014). The osca mutants exhibit a reduced hyperosmolality-induced [Ca²⁺]_cyt increase and these channels belong to the transmembrane (TMEM) 63 proteins, evolutionarily conserved mechanically activated ion channels in eukaryotes (Hou et al., 2014; Yuan et al., 2014; Murthy et al., 2018). Plants have expanded this family with 15 genes in Arabidopsis and 11 in rice (Li et al., 2015). Structural studies indicate that OSCAs consist of 11 transmembrane (TM) helices, which bear similarities to mammalian TMEM16 proteins, and a cytosolic domain and form homo-dimers (Liu et al., 2018; Zhang et al., 2018; Maity et al., 2019).

OSCAs have mostly been linked to osmotic stress-sensing (Hou et al., 2014; Yuan et al., 2014). OSCA1.2 functions as a hyperosmolality-activated mechanosensitive calcium channel (Yuan et al., 2014; Murthy et al., 2018). However, the biological role of most family members has not been determined. Recently, it was reported that OSCA1.3/1.7 regulate stomatal closure in plant PTI. Upon flg22 treatment, BIK1 phosphorylates OSCA1.3/1.7, thereby activating calcium channel activity and promoting stomatal closure and disease resistance (Thor et al., 2020). Interestingly, a reduced Ca²⁺ influx was observed only in guard cells but not in whole leaves of osca1.3/1.7 mutants. Furthermore, while stomata closure induced by the DAMP Pep1 was also impaired, abscisic acid-induced stomatal closure still occurs normally in the mutants, suggesting a specific role of OSCA1.3/1.7 in stomatal immunity, but not in general guard cell regulation (Thor et al., 2020). It will be interesting to determine how OSCA1.3/1.7 specifically regulate stomatal immunity. Since OSCAs are involved in osmotic sensing, the role of OSCAs in stomatal immunity may be related to water homeostasis. Future studies addressing whether infections cause turgor changes mediated by OSCAs or other mechanosensitive channels will elucidate the role of osmotic sensing in plant immunity.

TPCs

In addition to the apoplast, internal compartments contribute to the rise in [Ca²⁺]_cyt in plant immunity. However, the interplay of the PM and endomembrane calcium release channels remains opaque. The TPCs, present in plants and animals, are slow vacuolar calcium channels co-regulated by voltage and calcium (Peiter et al., 2005; Patel et al., 2016). TPCs form dimers, where each subunit consists of two nonidentical Shaker-like pore-forming sections with six transmembrane domains and a pore region in each section, making it a quasi-tetramer (Guo et al., 2016; Kintzer and Stroud, 2016). Two EF-hand motifs in the cytosolic linker part suggest Ca²⁺-dependent regulation, which further potentiates voltage activation (Guo et al., 2016; Kintzer and Stroud, 2016).

Plant TPCs have been implicated in plant immunity. Arabidopsis has a single TPC gene (Peiter et al., 2005). A gain-of-function allele resulting in the overactivation of two-pore channel 1 (TPC1) in Arabidopsis led to increased JA accumulation and resistance to B. cinerea (Bonaventure et al., 2007). However, neither overexpression nor tpc1 null mutant plants displayed altered MAMP flg22/elf18-induced Ca²⁺ influx and defense responses (Ranf et al., 2008), questioning the role of TPC1 in response to bacterial elicitors. Interestingly, wounding and herbivores trigger rapid TPC1-dependent local and systemic Ca²⁺ signals (Kiep et al., 2015; Vincent et al., 2017). Additionally, the overactivation of TPC1 inhibited aphid proliferation (Vincent et al., 2017), suggesting that TPC1 may play a role in plant systemic Ca²⁺ signaling and insect resistance. Rice OsTPC1 was reported to play a positive role in elicitor-induced defense gene expression, MAP kinase activation, and HR (Kurusu et al., 2005). It remains unknown how TPC1 mediates Ca²⁺ release from the vacuole in coordination with PM channels.

ANNs

ANNs are conserved phospholipid- and Ca²⁺-binding proteins that function as atypical Ca²⁺-permeable channels/transporters (Saad et al., 2020). However, whether ANNs form channels or transport Ca²⁺ by a different mechanism needs to be clarified, as ANNs do not possess a transmembrane domain. ANNs are cytosolic proteins, but the binding of Ca²⁺ facilitates their association with membrane lipids (Davies, 2014; Demidchik et al., 2018). In animals, ANNs are involved in exocytosis and endocytosis by binding to membrane lipids in a Ca²⁺ dependent manner (Gerke et al., 2005). Plant ANNs are characterized by the C-terminal repeated ANN domains with an endonexin fold for Ca²⁺-binding. The Arabidopsis genome encodes eight ANNs (Clark et al., 2012). Different ANNs exhibit diverse functions, including calcium channel/transporter, phosphodiesterase, F-actin binding, peroxidase, and RNA-binding (Konopka-Postupolska and Clark, 2017; He et al., 2021).

Members of plant ANNs are transcriptionally induced by SA and pathogen infection (Vandeputte et al., 2007), indicating a likely involvement in plant–pathogen interactions. Ectopic expression of ANN1 from Brassica juncea in tobacco (Nicotiana tabacum) confers tolerance to Phytophthora parasitica (Jami et al., 2008). Furthermore, ANN1 interacts with the chitin receptor CHITIN ELICITOR RECEPTOR KINASE 1 (CERK1) in mediating chitin-induced Ca²⁺ influx (Espinoza et al., 2017). Intriguingly, ann1 mutants showed increased chitin-induced ROS production and MAP kinase induction (Espinoza et al., 2017). Additionally, ANN1 is involved in Ca²⁺ influx induced by DAMP extracellular (e)ATP in roots but not leaves (Mohammad-Sidik et al., 2021). ANN1 plays a role in systemic but not local calcium elevation and the defense response to herbivores (Malabarba et al., 2021). Notably, ANN1 and ANN2 bear RNA-binding properties and contribute to small RNA loading into extracellular vesicles, which likely contributes to their role in defense against B. cinerea (He et al., 2021). It remains unknown whether and how the above-mentioned immunity-related phenotypes of ANN1 are linked to its putative Ca²⁺-binding activities. In addition, ANN8 negatively regulates RESISTANCE TO POWDERY MILDEW 8.1 (RPW8.1)-mediated resistance to powdery mildew (Zhao et al., 2021b).

In light of the wide distribution of ANNs in eukaryotes, nematodes deploy ANN-like effectors to suppress plant defense, likely by manipulating the function of endogenous ANNs. The soybean (Glycine max) nematode Heterodera glycines secretes an ANN-like effector, which complements the function of ANN1, indicating a functional mimicry (Patel et al., 2010). This is a strategy often employed by pathogens to colonize their hosts and suppress host immunity. Furthermore, a Meloidogyne incognita effector hijacks ANN1 and ANN4 to suppress plant immunity, likely by impairing H₂O₂-induced Ca²⁺ influx (Zhao et al., 2019). Consistently, ANN1 and ANN4 play a positive role in the resistance to this nematode (Zhao et al., 2019). Despite the recent emerging roles of ANNs in plant immunity, there is very little functional data on how they conduct Ca²⁺ and how they are regulated.

NLR resistosome as calcium channels

NLRs are tripartite-domain proteins with a variable N-terminus, a central nucleotide-binding and oligomerization domain, and a C-terminal leucine-rich repeat domain. Based on features of the N-terminal domain, plant NLRs are classified into three groups: coiled-coil domain-containing NLRs (CNLs), Toll/interleukin-1 receptor (TIR) domain-containing NLRs (TNLs), and RPW8 domain-containing NLRs (Saur et al., 2020; Bi and Zhou, 2021).

NLR-triggered ETI is often accompanied by HR, a form of programmed cell death (PCD) at the site of infection. Ca²⁺ influx is a hallmark of PCD in animals and plants and plays a central role in NLR signaling (Huysmans et al., 2017). Activation of the NLR RPM1 (resistance to Pseudomonas syringae pv. maculicola 1) by the bacterial effectors AvrB or AvrRpm1 triggers a second prolonged Ca²⁺ influx (Grant et al., 2000). LaCl₃, a calcium channel blocker, suppresses the NLR ZAR1 (HOPZ-activated resistance 1; Wang et al., 2019b) and RPM1-induced HR (El Kasmi et al., 2017). It has been assumed that NLR activation by effectors activates calcium channels to facilitate cytoplasmic Ca²⁺ influx. However, the identity of calcium channels in ETI remained elusive until recently.

The structural analysis of the NLR resistosomes, including ZAR1 (Wang et al., 2019b, 2019d), RPP1 (recognition of Peronospora parasitica 1) (Ma et al., 2020), and Roq1 (recognition of XopQ 1) (Martin et al., 2020), have advanced our understanding of NLR-triggered cell death and Ca²⁺ conductance (Bi et al., 2021; Jacob et al., 2021). The Cryo-EM structural study of the ZAR1 resistosome demonstrated inactive monomeric and active pentameric states of a full-length plant CNL. Uridylylation of RLCK PBL2 by the Xanthomonas campestris effector AvrAC leads to binding of the modified PBL2^UMP to RKS1, an RLCK without kinase activity (pseudokinase), which forms an inactive complex with ZAR1 in the resting state. Binding of PBL2^UMP causes a conformational change that triggers the release of ADP form ZAR1. In the new conformation, the ZAR1:RKS1:PBL2^UMP complex binds ATP, which induces a second conformational change and forms a multimeric resistosome complex (Wang et al., 2019b, 2019d). The N-terminal α1 helices of five ZAR1 protomers form a funnel-shaped structure, which is crucial for effector-induced ZAR1 PM targeting, cell death, and immunity (Wang et al., 2019b, 2019d). This structure shares similarities with previously characterized pore-forming proteins, like mammalian MLKL (MIXED-LINAGE KINASE-LIKE), which has also been suggested to form a cation channel (Xia et al., 2016). Therefore, it was proposed that the ZAR1 resistosome forms pores at the PM and functions as a cation channel.

Indeed, the electrophysiological analysis of the ZAR1 resistosome showed that it possesses cation channel activity in Xenopus oocytes (Bi et al., 2021). The ZAR1 resistosome pore is permeable to variable cations, such as Na⁺, K⁺, Cs⁺, Mg²⁺, and Ca²⁺, but with distinct current amplitudes (Bi et al., 2021), indicating that ZAR1 acts as a Ca²⁺-permeable cation channel. Consistent with in vitro electrophysiological data, the co-expression of ZAR1 resistosome components in zar1 mutant protoplasts induced Ca²⁺ influx and ROS production (Bi et al., 2021). Single-molecule imaging revealed that the ZAR1 oligomer forms pentameric complexes located at the PM (Bi et al., 2021). Organelles such as chloroplasts and vacuoles were damaged before the loss of PM integrity and abrupt cell rupture (Bi et al., 2021). However, the pore formed by ZAR1 may not be the direct executer for cell death, as the ZAR1 resistosome forms much earlier than the PM loses integrity (Bi et al., 2021).

On the other hand, it remains to be determined whether other plant CNLs also form similar resistosome structures that mediate Ca²⁺ influx. The MADA motif, conserved in ∼20% of plant CNLs, matches the N-terminal α1 helix of ZAR1 (Adachi et al., 2019); thus, it is conceivable that these NLRs could form ZAR1-like resistosomes and also function as calcium channels.

Two other small groups of NLRs in the NRG1 (N REQUIREMENT GENE 1) and ADR1 (ACTIVATED DISEASE RESISTANCE 1) families (known as helper NLRs), which carry an N-terminal RPW8-like domain, are required for sensor TNLs to induce cell death and immunity (Dong et al., 2016; Qi et al., 2018; Castel et al., 2019; Wu et al., 2019). The TIR domains of TNLs bear nicotinamide adenine dinucleotide (NAD⁺) hydrolase (NADase) activity, which is required for activation of downstream immune responses (Horsefield et al., 2019; Wan et al., 2019). Recently, two structural studies reported the assembly of Roq1 and RPP1 into homotetrameric resistosomes upon binding of Xanthomonas effector XopQ and H. arabidopsidis effector ATR1, respectively (Ma et al., 2020; Martin et al., 2020). Unlike ZAR1, the RPP1 resistosome contains a stabilized tetrameric TIR domain structure with NADase activity, which is thought to be responsible for activating helper NLRs (Castel et al., 2019; Lapin et al., 2019; Ma et al., 2020).

Intriguingly, the structure of the helper NLR NRG1.1 CC_R domain resembles that of the resting state ZAR1 and MLKL proteins (Jacob et al., 2021). Similar to ZAR1, activated NRG1.1 oligomerizes and forms puncta at the PM (Jacob et al., 2021). Furthermore, cell death caused by active NRG1.1 correlates with the number of pores formed by active NRG1.1 (Jacob et al., 2021). Ca²⁺ imaging and electrophysiological analysis of active NRG1.1 and ADR1 showed that NRG1.1 and ADR1 act as Ca²⁺-permeable channels and subsequently trigger cell death (Jacob et al., 2021). Recently, it has been shown that activation of TIR signaling promotes the self-association of the ADR1 family member ADR1-like 1 (ADR1-L1) (Wu et al., 2021). However, the signal and mechanism from activated NLRs that activates helper NLRs ADR1 and NRG1 remain largely unknown.

Furthermore, the predicted pore size formed by the activated ZAR1 and NRG1/ADR1 complexes is much smaller than that formed by animal gasdermin, which causes the bulk release of macromolecules from cells destined to die (Saur et al., 2020), suggesting other direct executors may exist in NLR-mediated cell death. The increase in [Ca²⁺]_cyt after forming these resistosome complexes may instead activate a cell death signaling cascade. Altered endoplasmic reticulum (ER) and vacuolar structures were observed 3 h before cell rupture. It has been known that in plants, the vacuole plays a role in regulating NLR-triggered cell death. For example, activation of tobacco NLR N results in vacuole disruption, and the release of the vacuolar protease VPE1 has been connected to HR formation (Hatsugai et al., 2004). Indeed, CNGC11-12 mediated cell death was VPE1-dependent (Urquhart et al., 2007). In addition, the activation of the CNLs RPM1 and RPS2 induces the fusion of the central vacuole with the PM (Hatsugai et al., 2009). Thus, it remains to be determined what the role of resistosome-mediated Ca²⁺ influx is and whether other calcium channels are involved in ETI. ETI potentiates PTI (Pruitt et al., 2020; Ngou et al., 2021). Thus, during ETI, PTI-related channels may also be activated.

Regulation of calcium channels mediating specific Ca²⁺ signals

Ca²⁺ influx needs to contain stimulus-specific information and then create unique Ca²⁺ signals to activate distinct downstream signaling. To achieve this, precise temporal and spatial regulation of each channel is of paramount importance. How can such precise regulation of each channel be achieved? The regulation of these channels occurs at multiple levels, including the transcript and protein levels, by ligands, through the interaction with other proteins, including subunit oligomerization and physical and electric status changes of the PM (Demidchik et al., 2018). Likely, multiple regulatory mechanisms are at work to generate specific signals.

Transcriptional regulation-stimulus induction and cell-type/tissue specificity

The members of the expanded plant calcium channel families display distinct expression patterns of tissue specificity and in response to different stimuli, indicating that transcriptional regulation is an important determinant of their biological roles. The clade I CNGC3, 11, 12, and 10, as well as the clade IV CNGC4, 19, and 20 are transcriptionally upregulated upon pathogen infections (http://bar.utoronto.ca/; Balagué et al., 2003). GLR2.7/2.8/2.9 are transcriptionally upregulated by different MAMPs and DAMPs and glr2.7/2.8/2.9 mutants showed a moderately compromised Ca²⁺ influx upon MAMP treatment (Bjornson et al., 2021b), supporting their role in PTI responses. This partial effect suggests that either other GLR members or calcium channels also contribute to PRR-activated Ca²⁺ influx.

Different cell-types/tissues may utilize different channels/subunits to create specific Ca²⁺ fluxes. The cell-type specificity may occur at the level of transcription and be regulated by other mechanisms. One example is OSCA1.3, which mediates PTI in guard cells, but not at the whole leaf level (i.e. mesophyll cells), even though it is expressed in both cell types (Thor et al., 2020). This example indicates that although the channels are transcriptionally regulated, another mechanism is at work for their cell-type specific function.

Activation by ligands or mechanical triggers

While some channels are activated by polarization (voltage) changes, most plant calcium channels require binding of a ligand or a mechanical trigger to initiate Ca²⁺ conductance (Demidchik et al., 2018).

Plant CNGCs are likely gated by cyclic nucleotides, based on their similarity to animal counterparts (Kaplan et al., 2007). However, the effects and requirement of cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate remain controversial and inconclusive (Dietrich et al., 2020; Jarratt-Barnham et al., 2021). Furthermore, the existence of plant cNMP-producing enzymes is not yet conclusively proven (Gehring and Turek, 2017; Blanco et al., 2020), although some evidence suggests a role of cAMP in immunity (Ma et al., 2009; Sabetta et al., 2019) and increased cAMP levels were reported upon pathogen infection (Jiang et al., 2005). Thus, the production of cNMPs upon recognizing microbes and the role of cNMPs in regulating plant CNGCs await to be further illustrated (Dietrich et al., 2020; Jarratt-Barnham et al., 2021).

Unlike animal GLRs, many amino acids, as well as the tripeptide glutathione and the ethylene precursor 1-aminocyclopropane-1-carboxylic acid can gate and activate plant GLR channels (Li et al., 2013; Mou et al., 2020; Grenzi et al., 2021a). The structural analysis of GLR3.4 and ligand specificity suggested a homo-tetrameric channel with an amino acid and glutathione-dependent gating mechanism, further substantiating distinct features of the GLR gating mechanism from animal GLRs (Green et al., 2021). GLRs have been connected to systemic signaling upon wounding or herbivory (Mousavi et al., 2013; Toyota et al., 2018). Electrophysiological analysis identified two GLRs that contribute to cytosolic Ca²⁺ increases to propagate systemic wound signaling and are gated by extracellular glutamine and pH (Shao et al., 2020). Whether the ligands act locally and trigger systemic Ca²⁺ or electric signals or travel through the vasculature remains to be determined (Johns et al., 2021).

Other channels, including OSCAs, are probably activated by sensing tensions in the PM caused by changes in turgor pressure (Liu et al., 2018; Maity et al., 2019). The cytosolic domain of OSCAs contains two long helices that lie parallel to the PM. They contact the PM lipid molecules and may sense PM tensions caused by changes in turgor pressure (Engelsdorf et al., 2018; Liu et al., 2018; Maity et al., 2019). Cell wall integrity) impairment by pathogens can also lead to the activation of mechanosensitive calcium channels such as MCA1 (MATING PHEROMONE INDUCED DEATH 1 (MID1)-COMPLEMENTING ACTIVITY 1) (Engelsdorf et al., 2018).

ANNs are activated by ROS, key molecules produced in plant PTI and ETI (Laohavisit et al., 2012). Interestingly, NADPH oxidases responsible for ROS production in plant immunity contain an EF-hand motif and are regulated by Ca²⁺ (Kadota et al., 2015; Castro et al., 2021). Furthermore, the identification of the RLK HPCA1 as a receptor for extracellular H₂O₂, which triggers an increase in [Ca²⁺]_cyt, suggests an intertwined relationship between Ca²⁺ and ROS production and signaling (Wu et al., 2020). However, HPCA1 is not required for MAMP-induced Ca²⁺ bursts, suggesting the existence of additional ROS receptors for MAMP responses. Nevertheless, multiple channels belonging to different families likely act together to create specific Ca²⁺ signals in plants in response to pathogen infection. Initial Ca²⁺ influx by a particular channel may trigger a cascade response of opening and closure of other channels.

Post-translational regulations

Post-translational modifications, such as phosphorylation, ubiquitination, and N-glycosylation, have emerged as important regulatory mechanisms in modulating immune responses (Kong et al., 2021). The direct phosphorylation of CNGC19 and CNGC20 by BAK1 regulates their stability and homeostasis, resulting in the increased accumulation of CNGC19/20 in the bak1 serk4 double mutant (Yu et al., 2019). CNGC20 undergoes proteasome-dependent degradation and is likely ubiquitinated (Yu et al., 2019). It is not known whether and how BAK1-mediated phosphorylation relates to ubiquitination. There are multiple putative glycosylation sites in CNGC19 and CNGC20. Conceivably, CNGC19/CNGC20 are glycosylated, and glycosylation regulates CNGC19/CNGC20 maturation and transportation to the PM.

In contrast, CaM-gated CNGC2 and CNGC4 are activated by BIK1-mediated phosphorylation upon MAMP elicitation (Tian et al., 2019). Intriguingly, BIK1 phosphorylates the C-terminal cytosolic domain of CNGC4, but not CNGC2, and phosphorylation sites in CNGC4 are not conserved in CNGC2 (Tian et al., 2019). The mechanism linking phosphorylation of CNGC4 by BIK1 and channel activation remains an open question. Similarly, the rice BIK1 homolog, OsRLCK185, phosphorylates OsCNGC9 to activate its channel activity (Wang et al., 2019c). Furthermore, OSCA1.3 is phosphorylated by BIK1 upon flg22 treatment at the first cytoplasmic loop, and this phosphorylation is essential for its function in stomatal immunity (Thor et al., 2020). Apparently, phosphorylation is a general theme for the activation of different calcium channels. It is possible that other types of calcium channels, such as GLRs, could be directly regulated by PRR complexes and associated RLCKs.

Oligomerization of subunits

Mammalian CNGCs form heterotetrameric channels and the combination of subunits varies depending on the cell type. Evidence hints at heterotetrametric channel formation for CNGC2/4, CNGC19/20, and CNGC7(8)/18 (Chin et al., 2013; Pan et al., 2019; Tian et al., 2019; Yu et al., 2019; Zhao et al., 2021a). However, it is not clear whether these combinations are constitutive or dynamic depending on the context.

Considering the expanded family members of CNGCs in plants, the combination of different CNGC subunits in a tetrameric channel could generate large variations that potentially provide a wide range of regulatory mechanisms. Different subunits have unique C- and N-terminal cytosolic portions that can be subject to posttranslational modifications and interaction with a range of proteins, such as CaMs, kinases, and ubiquitin E3 ligases. These interactors can positively or negatively regulate channel activity (DeFalco et al., 2016a; Tian et al., 2019). Furthermore, CNGC7 and CNGC8 repress CNGC18 activity when they form heteromeric complexes, indicating that some subunits act as repressors (Pan et al., 2019). Binding of apo-CaM2 de-represses (opens) the channel; then increased [Ca²⁺]_cyt leads to the dissociation of Ca²⁺-CaM2, restoring the resting state of the channel (closed). Thus, Ca²⁺ oscillations can be achieved by the combination of a repressive subunit and CaM (Pan et al., 2019).

The competition between different subunits may determine whether homomeric or heteromeric channels or which types of heteromeric channels are being formed. These channels may display different activities and regulation (Chiasson et al., 2017). This could be achieved either by partner proteins, like CaM, that affect subunit interaction or by changing expression levels of subunits under specific conditions or in different cell types.

Evidence also suggests the oligomerization of other calcium channel units, including the above-mentioned plant resistosomes. Hetero-tetrameric GLR channels may exist (Price et al., 2013). However, GLR3.3 and 3.6, which are both involved in systemic signaling (Toyota et al., 2018), are expressed in different cell types and therefore unlikely to form a heteromeric channel together (Nguyen et al., 2018). OSCA1.1 and OSCA1.2 form homomeric dimers (Jojoa-Cruz et al., 2018; Zhang et al., 2018). It remains to be determined whether OSCAs also form hetero-dimers (Murthy et al., 2018). Arabidopsis ANN1 and 4 and cotton GbANN5 and 6 interact with each other, implying that ANNs form heteromeric channels (Huang et al., 2013; Huh et al., 2010).

Channelsome formation by interaction with receptors and signaling components

Emerging data suggest that calcium channels that are in complex or in proximity to Ca²⁺-decoding proteins form sensing modules and signaling hubs, which have been coined as “channelsomes” (Dietrich et al., 2020; Jaillais and Ott, 2020). For example, CNGC17 and CNGC10 interact with BAK1 and AUTOINHIBITED H⁺-ATPASEs (AHAs) and are involved in phytosulfokine and brassinosteroid signaling, respectively (Ladwig et al., 2015; Grosseholz et al., 2021). GLR3.7 interacts with a 14-3-3 protein in a CDPK-mediated phosphorylation dependent manner (Wang et al., 2019e). ANN1 was found in a PM “nanodomain” with the slow anion channel 1 (SLAC1) homolog 3 SLAC3 anion channel (Demir et al., 2013). ANNs may interact with C2 domain-containing proteins, suggesting a connection to phospholipases (Davies, 2014).

Additionally, CNGCs may form complexes with other channels/transporters. For example, MtCNGC15a–c interact with the Ca²⁺ channel does not make infections 1 (DMI1) at the nuclear envelope (Charpentier et al., 2016), and Arabidopsis CNGC15 interacts with nitrate transporter 1.1 (NRT1.1), where nitrate levels determine Ca²⁺ influx in root tips (Wang et al., 2021). This increases the level of complexity of calcium channel regulation even further.

Interactors of calcium channels can also be necessary for delivering these proteins from the ER to their final destinations. Mildew Locus O 5 (MLO5) and MLO9, which are seven-transmembrane domain-containing proteins, recruit CNGC18-containing vesicles to the PM through the SNAP receptor (SNARE) complex in pollen tube guidance (Meng et al., 2020). Furthermore, CORNICHON proteins are transmembrane proteins essential for sorting and trafficking GLRs from the ER to the PM or potentially to other internal Ca²⁺ reservoirs such as the ER, vacuole, and mitochondria, contributing to Ca²⁺ homeostasis in the cytosol (Wudick et al., 2018). Thus, regulating such cargo receptor proteins may be crucial for the timely delivery of specific channel subunits.

Conclusion and future perspectives

Cytosolic free Ca²⁺ serves as an essential second messenger to relay cellular signals in plant growth, development, and stress responses. Ca²⁺ spikes with overlapping and distinct features in timing, kinetics, amplitude, and sources ubiquitously occur in plant responses to pathogen and insect attacks. Significant progress in this field has been made in the last few years; however, some crucial questions still need to be addressed (Figure 3). A rapid transient [Ca²⁺]_cyt spike is associated with PTI, whereas a second sustained Ca²⁺ influx is likely specific for ETI. Interestingly, dynamic ROS bursts occur in PTI and ETI. It is possible that Ca²⁺ signals relay signals through the EF-hand motif of NADPH oxidases for ROS production, which further connects and potentiates PTI and ETI responses (Castro et al., 2021). Additionally, ROS derived from other sources, including chloroplasts and mitochondria may also contribute to defense activation (Castro et al., 2021). [Ca²⁺]_cyt increase triggered upon immune elicitation appears to be mainly from the apoplast. However, some cytoplasmic organelles also sequester high concentrations of Ca²⁺. It remains largely elusive how different organelles coordinate Ca²⁺ signals (Figure 3).

Figure 3.

Future perspective of Ca²⁺ signals and channels in plant immunity. Perception MAMPs/HAMPs/DAMPs by PRRs and effectors by nucleotide-binding domain leucine-rich repeat proteins (NLRs) receptors lead to PTI and ETI, respectively. PTI and ETI potentiate each other. However, the precise mechanism needs yet to be elucidated. Both PTI and ETI trigger Ca²⁺ influx with distinct features. Immunity-induced Ca²⁺ influx appears to come from different Ca²⁺ stores, such as the apoplast, vacuole, chloroplast, possibly ER, and nuclear envelope. It remains elusive how the Ca²⁺ stores coordinate for spatial and temporal Ca²⁺ dynamics in plant immunity. Ca²⁺ influx is achieved by different calcium channels, including CNGCs, GLRs, OSCAs, TPCs, ANNs, NLR-formed resistosomes, and possibly other calcium channels/transporters, which await to be discovered. Additionally, how diverse calcium channels coordinately function in plant immunity needs to be addressed. During PTI, RLCKs induce a ROS burst by phosphorylating NADPH oxidases and trigger Ca²⁺ influx by phosphorylating CNGCs and OSCAs. It is also possible that other channels like GLRs are activated by PRRs and associated proteins. Ca²⁺ signals are amplified, probably through binding to EF-hand motifs of NADPH oxidases for ROS production. ROS could be perceived by ROS sensors like HPCA1, leading to further activation of the calcium channels, triggering sustained cytosolic Ca²⁺ elevation. Ca²⁺ transporters could coordinate with calcium channels to shape specific Ca²⁺ signatures. How NLR resistosomes trigger cell death during ETI is still not clear. Rapid Ca²⁺ waves are generated that move to distal tissues upon wounding and herbivory attacks and are propagated by GLRs. It remains to be shown how other channels play a role in systemic immune signaling. Some outstanding questions in the field are summarized.

Ca²⁺ influx is usually achieved by Ca²⁺-permeable ion channels, including CNGCs, GLRs, TPCs, and ANNs, all of which have been linked to immunity. Additionally, Ca²⁺-ATPases and Ca²⁺/H⁺ exchangers mediate Ca²⁺ efflux and sequestration and play a role in establishing stimulus-specific Ca²⁺ signatures in response to infection. Two tonoplast-localized Ca²⁺-ATPase pumps, auto-inhibited Ca²⁺-ATPase, isoform 4 (ACA4) and ACA11, negatively regulate flg22-induced Ca²⁺ signals (Hilleary et al., 2020). It remains unknown how the different types of calcium channels coordinately function in plant immunity (Figure 3). In addition, assembly, homeostasis, activation, and deactivation of most calcium channels remain to be elucidated. Furthermore, the resistosomes formed by the NLRs ZAR1 and ADR1/NRG1 convey Ca²⁺-conducting activities; however, it is unknown whether they function specifically as unconventional calcium channels or display a broad conductivity for other molecules. It awaits to be determined whether other NLRs also form calcium channels. Considering the largely expanded members of NLRs in plants (>200 per Arabidopsis accession and >400 panNLRome (van de Weyer et al., 2019) and their activation by pathogen-encoded effector proteins, a complex regulatory network would be needed to fine-tune the resistosome-mediated channel activities and to maintain proper Ca²⁺ levels, a critical factor for cell survival. Layered transcriptional regulations of the calcium channels contribute to cell-type and tissue-expression specificity and responsiveness to stimuli. Additionally, post-translational modifications further dictate their homeostasis, assembly, activation, and attenuations.

It should also be considered that different cell types may utilize different channels to conduct Ca²⁺. As mentioned, OSCA1.3 specifically mediates PTI in guard cells but not at the whole leaf level (Thor et al., 2020). Recently, with the emergence of genetically encoded Ca²⁺ sensors, the cell-to-cell and systemic propagation of immune signals has gained increased attention. Ca²⁺ and ROS, as well as potentially electric signals, form a self-amplifying cascade that transmits signals locally and systemically at the whole plant level (for a detailed review see Johns et al., 2021). Several Ca²⁺-conducting channels from the GLR, CNGC and mechanosensitive small conductance-like families are involved in the propagation of a systemic signal triggered by light stress (Fichman et al., 2021). A similar interplay may occur for immune signaling as well.

The precise recording of Ca²⁺ signatures is essential for deciphering Ca²⁺ signaling in plant immunity. Recently, genetically encoded biosensors, such as aequorin and GCaMPs, have vastly improved our understanding of Ca²⁺ fluxes (Grenzi et al., 2021b). However, so far, they were mostly used to record Ca²⁺ changes at the whole organ level and did not reveal specific calcium signatures. The use of improved high-resolution imaging systems will contribute to dissect the role of various calcium channels in creating specific Ca²⁺ signatures in plant immunity (Grenzi et al., 2021b; Yoshinari et al., 2021).

Accession numbers

CNGC2 (AT5G15410), CNGC4 (AT5G15410), CNGC11 (AT2G46440), CNGC12 (AT2G46450), CNGC19 (AT3G17690), CNGC20 (AT3G17700), GLR2.7 (AT2G29120), GLR2.8 (AT2G29120), GLR2.9 (AT2G29100), GLR3.3 (AT1G42540), GLR3.6 (AT3G51480), OSCA1.3 (AT1G11960), OSCA1.7 (AT4G02900), ANN1 (AT1G35720), ANN2 (AT5G65020), ANN4 (AT2G38750), ANN8 (AT5G12380), TPC1 (AT4G03560), ZAR1 (AT3G50950), NRG1.1 (AT5G66900), ADR1 (AT1G33560).

 

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plcell) are: Keiko Yoshioka (keiko.yoshioka@utoronto.ca) and Libo Shan (lshan@tamu.edu).