1. Introduction
Manganese (Mn) is an essential trace metal that participates in numerous physiological processes and is indispensable for plant growth and development (Peiter, Reference Peiter, Krauss and Nies2014). Hence, Mn can be a limiting factor for crop performance and yield formation (Assuncao et al., Reference Assuncao, Cakmak, Clemens, González-Guerrero, Nawrocki and Thomine2022). As a transition metal, Mn is integral to redox processes and electron transfer reactions (Lilay et al., Reference Lilay, Thiébaut, du Mee, Assuncao, Schjoerring, Husted and Persson2024). It serves as a cofactor for various metalloenzymes, functioning as both a Lewis acid and redox catalyst. Owing to its role in the oxygen-evolving complex (OEC) in photosystem II, one of the earliest symptoms of Mn deficiency is a decline in photosynthesis, resulting from the reduced stability of the OEC and decreased electron transfer (Alejandro et al., Reference Alejandro, Meier, Hoang and Peiter2023; Schmidt et al., Reference Schmidt, Jensen and Husted2016). Beyond photosynthesis, adequate Mn levels are crucial for a plethora of processes, including respiration, oxidative stress mitigation, glycosylation, matrix polysaccharide and lignin biosynthesis and phytohormone homeostasis (Andresen et al., Reference Andresen, Peiter and Küpper2018). These reactions take place in different cellular compartments, necessitating the distribution of the metal on various levels of scale, first between plant organs and subsequently between and within organelles (Bashir et al., Reference Bashir, Ahmad, Kobayashi, Seki and Nishizawa2021). Compared to other metals, such as iron (Fe) and zinc (Zn), the mechanisms of Mn sensing and transport are less well understood (Huang, Yamaji, & Ma, Reference Huang, Yamaji and Ma2024). Since the transport and functions of Mn have last been addressed in comprehensive reviews (Alejandro et al., Reference Alejandro, Höller, Meier and Peiter2020; Andresen et al., Reference Andresen, Peiter and Küpper2018; He et al., Reference He, Rössner, Hoang, Alejandro and Peiter2021; Shao et al., Reference Shao, Yamaji, Shen and Ma2017), we have witnessed a substantial progress in some areas of plant Mn biology. In particular, the Mn supply of organellar targets and the regulation of Mn transporters at the post-translational level has been the focus of recent research. It has also become apparent that plant species differ markedly in their Mn handling and their employment of orthologous transport proteins. This review focuses on new developments in Mn transport at the organismic and cellular levels, highlighting the importance of intracellular distribution and post-translational modifications.
2. Mn in soil and rhizosphere
Mn is acquired from the soil solution as Mn2+, whereas higher oxidation states, such as Mn(III) and Mn(IV), present in oxides, are unavailable to plants (Alejandro et al., Reference Alejandro, Höller, Meier and Peiter2020). Mn availability is thus primarily determined by soil chemistry, in particular pH and redox status, with acidic and reducing conditions increasing its availability. In particular, the soil’s redox potential may change rapidly and on a small scale, for example, upon drying and rewetting, demanding rapid adaptation of Mn uptake and handling. Additionally, H+ release by roots increases Mn availability in the rhizosphere by a factor of 100 per unit pH change. Exuded organic anions may chelate Mn and function as electron donors, further increasing Mn solubility. Conditions that provoke carboxylate exudation, such as phosphorus deficiency, thus lead to increased Mn availability and uptake (Lambers et al., Reference Lambers, Hayes, Laliberté, Oliveira and Turner2015). Finally, microorganisms can mediate the redox cycling of Mn, and hence affect its availability to plants. Since Mn occurs in various oxidation states in soils and plants, we employ the generalised abbreviation Mn in this review when not referring to a specific oxidation state.
3. Getting in – pathways of Mn uptake
Prior to uptake into root cells, Mn may move apoplastically in the cortical cell walls towards the root endodermis. Coating of endodermal cells by a suberin layer physically prevents Mn influx in older root sections. Interestingly, deficiency of Mn, Fe and Zn led to delayed suberisation along the root axis (Barberon et al., Reference Barberon, Vermeer, De Bellis, Wang, Naseer, Andersen, Humbel, Nawrath, Takano, Salt and Geldner2016). This response depended on an intact ethylene signalling pathway, and likely serves to increase the absorptive area. Contrary to this finding in Arabidopsis, in roots of Hordeum vulgare (barley), the strength of Mn deficiency was decisive for endodermal suberisation, believed to either increase uptake into the xylem under mild deficiency or decrease leakage under severe deficiency (Chen et al., Reference Chen, Husted, Salt, Schjoerring and Persson2019).
Figure 1. Schematic representation of a hypothetical Arabidopsis thaliana cell containing all transport proteins from various gene families with known localisation. These proteins include members of the bivalent cation transporter family (BICAT, green), metal tolerance protein family (MTP, dark blue), Zrt-Irt-like proteins (ZIP, light blue), natural resistance-associated macrophage proteins (NRAMP, orange), calcium exchangers (CAX, purple) and P2A-type ATPases (yellow). The MTP family is implicated in detoxification processes, with AtMTP11 facilitating Mn transport into the Golgi apparatus, followed by exocytosis (depicted by blue arrows). Under Fe-deficient conditions, AtMTP8 sequesters excess Mn into the vacuole. NRAMP family members mediate Mn import into the cytosol (AtNRAMP1/6) and from intracellular compartments, such as TGN/EE (AtNRAMP2), enhancing Mn accumulation in chloroplasts. AtNRAMP3/4 mediates Mn release from the vacuole. AtBICAT2/PAM71-HL/CMT1 facilitates Mn transport across the chloroplast envelope into the stroma, while AtBICAT1/PAM71 transfers Mn into the thylakoid lumen. AtBICAT3/PML3 channels Mn into the trans-Golgi cisternae, where it plays a crucial role in glycosylation and cell wall matrix polysaccharide synthesis. Other depicted transporters are described in Alejandro et al. (Reference Alejandro, Höller, Meier and Peiter2020) and He et al. (Reference He, Rössner, Hoang, Alejandro and Peiter2021).
A transporter of the natural resistance-associated macrophage protein (NRAMP) family, AtNRAMP1, is essential for the high-affinity uptake of Mn into Arabidopsis root cells (Cailliatte et al., Reference Cailliatte, Schikora, Briat, Mari and Curie2010) (Figure 1). In planta, this protein also mediates the uptake of Fe with low affinity (Castaings et al., Reference Castaings, Caquot, Loubet and Curie2016), whereas high-affinity Fe uptake is pursued by iron-regulated transporter 1 (AtIRT1), a member of the Zrt/Irt-like family (Vert et al., Reference Vert, Grotz, Dédaldéchamp, Gaymard, Guerinot, Briat and Curie2002). Apart from Fe, AtIRT1 transports a range of other metal ions, including Mn (Korshunova et al., Reference Korshunova, Eide, Clark, Guerinot and Pakrasi1999). However, AtIRT1 does not contribute to Mn acquisition under conditions of sufficient Fe supply. After uptake, Mn may move symplastically from cell to cell via plasmodesmata towards the root vasculature and be released into the xylem apoplast, whereby the mechanism of xylem loading is still obscure.
Unlike AtIRT1 in Arabidopsis, the barley ortholog HvIRT1 is of high importance for the uptake and root-to-shoot translocation of Mn but is not involved in Fe acquisition (Long et al., Reference Long, Persson, Duan, Jorgensen, Yuan, Schjoerring and Pedas2018). In Oryza sativa (rice), the import of Mn into exo- and endodermal cells is primarily mediated by OsNRAMP5 (Sasaki et al., Reference Sasaki, Yamaji, Yokosho and Ma2012), an ortholog of AtNRAMP1, as well as additional transporters of lower relevance, such as OsNRAMP1 (Chang et al., Reference Chang, Huang, Yamaji, Zhang, Ma and Zhao2020). Release of Mn from those cells into the stellar apoplast is mediated by a transporter of the metal tolerance protein (MTP) family, OsMTP9 (Ueno et al., Reference Ueno, Sasaki, Yamaji, Miyaji, Fujii, Takemoto, Moriyama, Che, Moriyama, Iwasaki and Ma2015). OsNRAMP5 and OsMTP9 are polarly localised at the distal and proximal side of the cells, respectively, thus forming a one-way road for Mn. Polar localisation has not yet been observed for AtNRAMP1.
4. Mn allocation, storage and sensing in the vasculature
Following its transfer into the root vascular system, Mn needs to be allocated to plant parts demanding the metal, while preventing critically high concentrations under high load. In rice, the OsNRAMP3 protein functions as molecular points in the node, distributing Mn to either young leaves and panicles or to old leaves, depending on the Mn status (Yamaji et al., Reference Yamaji, Sasaki, Xia, Yokosho and Ma2013). This is brought about by OsNRAMP3-mediated loading of xylem transfer cells in enlarged vascular bundles as well as the subsequent loading of phloem parenchyma cells, also facilitated by OsNRAMP3. Under excessive Mn supply, OsNRAMP3 is degraded, allowing Mn to follow the main transpiration stream to old leaves. Xylem unloading of translocated Mn has recently been demonstrated to be a second task of OsNRAMP5 besides its function in root Mn uptake (Huang, Konishi, et al., Reference Huang, Konishi, Yamaji and Ma2024). In the sheath of rice leaves, this transporter is polarly localised in xylem parenchyma cells to supply Mn to this part of the leaf and restrict further translocation to the leaf blade.
Apart from OsNRAMP3 and 5, our picture of vascular Mn translocation in rice is sketchy, and it is even more so in Arabidopsis. Genes encoding several transport proteins permeating Mn are primarily or exclusively expressed in its vascular system, including AtNRAMP2 (Alejandro et al., Reference Alejandro, Cailliatte, Alcon, Dirick, Domergue, Correia, Castaings, Briat, Mari and Curie2017; Gao et al., Reference Gao, Xie, Yang, Xu, Li, Wang, Chen and Huang2018), AtNRAMP3 and 4 (Lanquar et al., Reference Lanquar, Lelièvre, Bolte, Hamès, Alcon, Neumann, Vansuyt, Curie, Schröder, Krämer, Barbier-Brygoo and Thomine2005; Lanquar et al., Reference Lanquar, Schnell Ramos, Lelièvre, Barbier-Brygoo, Krieger-Liszkay, Krämer and Thomine2010), AtNRAMP6 (Cailliatte et al., Reference Cailliatte, Lapeyre, Briat, Mari and Curie2009; Li et al., Reference Li, Zhu, Liao, Yang, Fan, Zhang, Yamaji, Dirick, Ma, Curie and Huang2022), yellow stripe like 1 (AtYSL1) and 3 (Waters et al., Reference Waters, Chu, DiDonato, Roberts, Eisley, Lahner, Salt and Walker2006), as well as the Ca2+/Mn2+-ATPase AtECA3 (Li et al., Reference Li, Chanroj, Wu, Romanowsky, Harper and Sze2008). Deletion mutants in these proteins have Mn-related phenotypes, but their role in stellar Mn allocation is largely unclear, as the specific vascular cell types expressing the transporters have not been well identified. However, genetic interactions of some of the transporters have been observed. For instance, AtNRAMP2, located in Trans Golgi Network (TGN)/Early Endosomes (EE), is required for Mn supply of chloroplasts (Alejandro et al., Reference Alejandro, Cailliatte, Alcon, Dirick, Domergue, Correia, Castaings, Briat, Mari and Curie2017), which was shown before to also depend on Mn release from the vacuole by AtNRAMP3 and 4 (Lanquar et al., Reference Lanquar, Schnell Ramos, Lelièvre, Barbier-Brygoo, Krieger-Liszkay, Krämer and Thomine2010). Accordingly, triple mutant analyses indicated that AtNRAMP2 and AtNRAMP3/4 operate successively. The mechanism that links AtNRAMP2/3/4 activity in the vascular system with chloroplast supply in mesophyll cells is unclear. A further player, the trans-Golgi-localised P2A-type ATPase AtECA3, is expressed exclusively in root and shoot vasculature and is required to maintain growth under low Mn supply by an unknown mechanism (Li et al., Reference Li, Chanroj, Wu, Romanowsky, Harper and Sze2008); growth defects of nramp2 and eca3 mutants were additive (Farthing et al., Reference Farthing, Henbest, Garcia-Becerra, Peaston and Williams2023).
Abolition of AtNRAMP6, located in plasma membrane and secretory pathway in an Mn-dependent manner, similarly led to reduced Mn translocation to young leaves and reduced root growth in Mn-deficient conditions (Li et al., Reference Li, Zhu, Liao, Yang, Fan, Zhang, Yamaji, Dirick, Ma, Curie and Huang2022). However, for unknown reasons, this defect became only apparent if AtNRAMP1 was also knocked out. Moreover, grafting experiments indicated that AtNRAMP1 and 6 were required in both root and shoot for the maintenance of root growth and the translocation of Mn, which might indicate shoot-to-root signalling of Mn status involving AtNRAMP6 (Li et al., Reference Li, Zhu, Liao, Yang, Fan, Zhang, Yamaji, Dirick, Ma, Curie and Huang2022).
An AtNRAMP2 homolog has recently been characterised in Zea mays (maize). In contrast to AtNRAMP2, ZmNRAMP2 is targeted to the tonoplast of xylem parenchyma cells and is required for root-to-shoot transfer of Mn (Guo et al., Reference Guo, Long, Chen, Dong, Liu, Chen, Wang and Yuan2022). Knockout plants showed decreased Mn concentration in the xylem sap with a concomitant decrease in photosynthesis, suggesting a role of ZmNRAMP2 in Mn translocation from roots to shoots by release of vacuolar Mn into xylem parenchyma cells.
The expression of a diverse set of Mn transporters predominantly or exclusively in the vasculature of roots and shoots indicates the importance of vascular tissues as a site of storage and remobilisation and probably in sensing Mn status. It will be highly informative to identify the specific vascular cell types in which individual transporters are expressed in order to elucidate the mechanisms and pathways underlying vascular Mn homeostasis. Based on its subcellular localisation, it may be hypothesised that AtNRAMP6 facilitates Mn uptake into the Arabidopsis vasculature. Under conditions of Mn deficiency, redistribution may occur through the release from vacuoles by AtNRAMP3/4, export into the Golgi by AtECA3, and further transport into the TGN/EE via vesicular trafficking, before distribution from the TGN/EE via AtNRAMP2, with YSL transporters subsequently mediating the transfer into mesophyll cells.
5. Organellar Mn homeostasis
5.1. The vacuole – Mn pantry and dump
On a cellular level, the vacuole, as the plant’s by far largest compartment, may contain high amounts of Mn. Vacuolar exporters, such as AtNRAMP3/4 in Arabidopsis (Lanquar et al., Reference Lanquar, Schnell Ramos, Lelièvre, Barbier-Brygoo, Krieger-Liszkay, Krämer and Thomine2010) or ZmNRAMP2 in maize (Guo et al., Reference Guo, Long, Chen, Dong, Liu, Chen, Wang and Yuan2022), mediate the supply of Mn-demanding cells and organelles by Mn release from the vacuole. Mutant analyses indicate that interim storage in the vacuole is the default pathway, and that vacuolar release is thus essential to maintain optimum plant growth under limiting Mn availability (Lanquar et al., Reference Lanquar, Schnell Ramos, Lelièvre, Barbier-Brygoo, Krieger-Liszkay, Krämer and Thomine2010). In poplar, a duplication of the NRAMP3 ortholog was identified recently, of which only one functions as a vacuolar Mn/Fe exporter like in Arabidopsis. The second one was localised to the secretory pathway and involved in intercellular translocation in the vascular system, thereby altering plant-wide Mn homeostasis (Pottier et al., Reference Pottier, Thi, Primard-Brisset, Marion, Bianchi, Victor, Déjardin, Pilate and Thomine2022).
Vacuolar Mn sequestration by AtMTP8 is employed to stock Mn in developing embryos of Arabidopsis (Chu et al., Reference Chu, Car, Socha, Hindt, Punshon and Guerinot2017; Eroglu et al., Reference Eroglu, Giehl, Meier, Takahashi, Terada, Ignatyev, Andresen, Küpper, Peiter and von Wirén2017). The transporter mediates the storage of the metal in subepidermal cells at the abaxial side of cotyledons and in cortical cells of the hypocotyl, which is required for efficient germination if the Mn supply of the mother plant is limited (Eroglu et al., Reference Eroglu, Giehl, Meier, Takahashi, Terada, Ignatyev, Andresen, Küpper, Peiter and von Wirén2017). Knockout of AtMTP8 causes a reallocation of Mn to provascular strands, mediated by the vacuolar Fe/Mn transporter AtVIT1 (Kim et al., Reference Kim, Punshon, Lanzirotti, Li, Alonso, Ecker, Kaplan and Guerinot2006). Intriguingly, the ubiquitous overexpression of AtMTP8 did not change the tissue distribution of Mn within the embryo, indicating that this pattern is determined by either post-translational regulation mechanisms or upstream processes that need yet to be resolved (Höller et al., Reference Höller, Küpper, Brückner, Garrevoet, Spiers, Falkenberg, Andresen and Peiter2022).
Vacuolar sequestration also prevents Mn from interfering with critical processes upon Mn overload. Plants employing an acidification/reduction-based strategy of Fe mobilisation face the problem of excessive Mn uptake, since alongside Fe, Mn becomes more available in the acidified rhizosphere, and Mn is an alternative substrate of the Fe uptake transporter IRT1. Under these conditions, in Arabidopsis, expression of AtMTP8 is strongly induced in cells expressing AtIRT1. Knock-out of AtMTP8 renders plants hypersensitive to Fe-deficiency-induced chlorosis (Chu et al., Reference Chu, Car, Socha, Hindt, Punshon and Guerinot2017; Eroglu et al., Reference Eroglu, Meier, von Wirén and Peiter2016; Giehl et al., Reference Giehl, Flis, Fuchs, Gao, Salt and von Wirén2023). This vacuolar sequestration of Mn is required for the functioning of the Fe acquisition machinery, with ferric chelate reductase likely to be affected by elevated cytosolic Mn (Eroglu et al., Reference Eroglu, Meier, von Wirén and Peiter2016).
Intriguingly, in Beta vulgaris ssp. vulgaris (sugar beet) Fe deficiency did not induce the expression of any Mn-sequestrating BvMTP and provoked a lower Mn accumulation in roots compared to Arabidopsis, indicating functional differences in their Fe acquisition machinery (Alejandro et al., Reference Alejandro, Meier, Hoang and Peiter2023). In Lupinus albus (white lupin), which accumulates massive amounts of Mn due to its P-mobilising activity, the role of MTP8-type transporters has been extended to Mn sequestration in leaves (Olt et al., Reference Olt, Alejandro, Fermum, Ramos, Peiter and Ludewig2022). Similarly, in rice as a monocot able to grow on submerged soils with very high Mn availability, mechanisms to cope with high Mn loads involve vacuolar sequestration in the shoot by MTP8-type transporters (Chen et al., Reference Chen, Fujii, Yamaji, Masuda, Takemoto, Kamiya, Yusuyin, Iwasaki, Kato, Maeshima, Ma and Ueno2013).
5.2. The chloroplast – Mn driving photosynthesis
The majority of Mn in the chloroplast is bound to PSII, where it is incorporated in the Mn4CaO5 cluster of the OEC, accepting electrons from H2O. On the luminal side of PSII, three extrinsic proteins (PsbO, PsbP and PsbQ) protect the OEC (Schmidt et al., Reference Schmidt, Jensen and Husted2016). Deficiency of Mn causes the dissociation of PsbP and PsbQ, leading to Mn release into the thylakoid lumen and ROS production, damaging light-harvesting pigments of PSII (Lilay et al., Reference Lilay, Thiébaut, du Mee, Assuncao, Schjoerring, Husted and Persson2024). Mn-deficient plants of Marchantia polymorpha, an emerging model to study metal homeostasis, also showed increased non-photochemical quenching (NPQ) by increased cyclic electron flow to protect PSII against photoinhibition. Additionally, chloroplast ultrastructure was altered under Mn deficiency, reflected by disorganised thylakoids (Messant et al., Reference Messant, Hani, Hennebelle, Guérard, Gakière, Gall, Thomine and Krieger-Liszkay2023).
Mn supply of chloroplasts is conferred by transporters that were grouped in the uncharacterized protein family 0016 (UPF0016). In plants, this family has been assigned various names, including photosynthesis-affected mutant 71 (PAM71), PAM71-homolog (PAM71-HL), chloroplast manganese transporter (CMT), chloroplast-localised Ca2+/H+ antiporter (CCHA), photosynthesis-affected mutant 71-Like (PML) and bivalent cation transporter (BICAT). In this review, we employ the latter terminology, as it embraces all family members and reflects their function (He et al., Reference He, Rössner, Hoang, Alejandro and Peiter2021). Proteins of the BICAT family are related to Gcr1-dependent translation factor 1 (GDT1) in Saccharomyces cerevisiae, which was initially identified as a Ca2+ transporter localised in the Golgi (Demaegd et al., Reference Demaegd, Foulquier, Colinet, Gremillon, Legrand, Mariot, Peiter, Van Schaftingen, Matthijs and Morsomme2013). Later work demonstrated it also transports Mn (Thines et al., Reference Thines, Deschamps, Sengottaiyan, Savel, Stribny and Morsomme2018). This dual functionality was also observed for its human homolog, TMEM165 (Stribny et al., Reference Stribny, Thines, Deschamps, Goffin and Morsomme2020). In Arabidopsis, AtBICAT2/PAM71-HL/CMT1 translocates Mn across the inner envelope membrane, and mutants display symptoms that resemble Mn deficiency, including decreased photosynthetic efficiency and altered chloroplast ultrastructure (Eisenhut et al., Reference Eisenhut, Hoecker, Schmidt, Basgaran, Flachbart, Jahns, Eser, Geimer, Husted, Webers, Leister and Schneider2018; Zhang et al., Reference Zhang, Zhang, Liu, Jing, Wang, Jin, Yang, Fu, Shi, Zhao, Lan and Luan2018). Moreover, in addition to Mn, AtBICAT2/PAM71-HL/CMT1 transports Ca2+ like its yeast and human counterparts, and is required for Ca2+ elevations in the chloroplast stroma induced by the onset of darkness (Frank et al., Reference Frank, Happeck, Meier, Hoang, Stribny, Hause, Ding, Morsomme, Baginsky and Peiter2019). From the stroma, Mn is further translocated into the lumen by AtBICAT1/PAM71 localised in the thylakoid membrane (Schneider et al., Reference Schneider, Steinberger, Herdean, Gandini, Eisenhut, Kurz, Morper, Hoecker, Ruhle, Labs, Flugge, Geimer, Schmidt, Husted, Weber, Spetea and Leister2016). Hence, the supply of the OEC with Mn is hampered in bicat1/pam71 mutants. Complementation of a bicat1/pam71 mutant demonstrated its functional similarity to human TMEM165 as well as the cyanobacterial Mn exporter MNX (Hoecker et al., Reference Hoecker, Hennecke, Schrott, Marino, Schmidt, Leister and Schneider2021). Intriguingly, AtBICAT1/PAM71 also determines Ca2+ homeostasis in the chloroplast stroma, with its deletion augmenting the darkness-induced [Ca2+]stroma transient (Frank et al., Reference Frank, Happeck, Meier, Hoang, Stribny, Hause, Ding, Morsomme, Baginsky and Peiter2019).
While constitutive Mn supply of the thylakoid lumen is essential to drive the photosynthetic light reaction, and, as recent work showed, Mn activates protein kinases in the stroma (Espinoza-Corral et al., Reference Espinoza-Corral, Schwenkert and Schneider2023), Ca2+ regulates various other processes in the chloroplast, including carbon assimilation and protein import (He et al., Reference He, Rössner, Hoang, Alejandro and Peiter2021). The requirement of uncoupling Ca2+ and Mn homeostasis raises the question, if and how selectivity of BICAT proteins is accomplished in planta. Unfortunately, neither free Mn ([Mn]) levels in chloroplast and cytosol nor the kinetics and selectivity of BICAT1/2 are known. Yeast GDT1 heterologously expressed in Lactococcus lactis has a Km of 15 and 83 μM for Ca2+ and Mn, respectively (Thines et al., Reference Thines, Deschamps, Sengottaiyan, Savel, Stribny and Morsomme2018), while TMEM165 transports Ca2+ and Mn with a Km of 21 and 170 μM, respectively (Stribny et al., Reference Stribny, Thines, Deschamps, Goffin and Morsomme2020). If plant BICATs operate in a similar concentration range, their Km for Ca2+ would be around 100-fold higher than steady-state [Ca2+]cyt. Despite a lower affinity for Mn than for Ca2+, such a transporter may still preferentially transport Mn when free Mn exceeds free Ca2+.
Transporters of the GDT1 family presumably function as H+ antiporters, and the activity of GDT1 and TMEM165 has been demonstrated to alter organellar pH homeostasis in yeast and human cells (Demaegd et al., Reference Demaegd, Foulquier, Colinet, Gremillon, Legrand, Mariot, Peiter, Van Schaftingen, Matthijs and Morsomme2013; Deschamps et al., Reference Deschamps, Thines, Colinet, Stribny and Morsomme2023; Wang et al., Reference Wang, Yang, Huang, He, Li, Zhang, Zhang and Tang2020). Similarly, BICATs may thus impact pH in chloroplast compartments, which is relevant because the proton motive force across the thylakoid membrane drives ATP synthesis, and luminal pH regulates NPQ. Indeed, electrochromic shift measurements indicated an altered pH gradient in bicat1/pam71 mutants (Schneider et al., Reference Schneider, Steinberger, Herdean, Gandini, Eisenhut, Kurz, Morper, Hoecker, Ruhle, Labs, Flugge, Geimer, Schmidt, Husted, Weber, Spetea and Leister2016). In this respect, it is notable that the H+ antiport activity of yeast GDT1, when expressed in Lactococcus lactis is reversible (Deschamps et al., Reference Deschamps, Thines, Colinet, Stribny and Morsomme2023). There are hints from yeast complementation experiments that this is also the case for plant BICATs: Besides complementing yeast mutants defective in Ca2+ and Mn efflux into organelles (Eisenhut et al., Reference Eisenhut, Hoecker, Schmidt, Basgaran, Flachbart, Jahns, Eser, Geimer, Husted, Webers, Leister and Schneider2018; Frank et al., Reference Frank, Happeck, Meier, Hoang, Stribny, Hause, Ding, Morsomme, Baginsky and Peiter2019; He et al., Reference He, Yang, Hause, Rössner, Peiter-Volk, Schattat, Voiniciuc and Peiter2022; Schneider et al., Reference Schneider, Steinberger, Herdean, Gandini, Eisenhut, Kurz, Morper, Hoecker, Ruhle, Labs, Flugge, Geimer, Schmidt, Husted, Weber, Spetea and Leister2016; Wang et al., Reference Wang, Xu, Jin, Zhang, Lai, Zhou, Zhang, Liu, Duan, Wang, Peng and Yang2016), BICATs also restore Mn influx in a yeast mutant devoid of the Mn uptake transporter Smf1 (Xu et al., Reference Xu, Wu, Liu, Tai, Zha, Gong, Zou, Zhang, Zhang and Chen2023; Zhang et al., Reference Zhang, Zhang, Liu, Jing, Wang, Jin, Yang, Fu, Shi, Zhao, Lan and Luan2018; Zhang, Fu et al., Reference Zhang, Fu, Sun, Ju, Miao, Wang, Xie, Ma, Gong and Wang2021). Hence, the BICAT proteins are likely capable of mediating transport in both directions, as governed by the electrochemical potential gradient of the transported ions.
5.3. The secretory pathway – an emerging Mn hub
The secretory pathway links diverse cellular organelles and membranes by vesicular trafficking and fusion events. These processes may thus allocate Mn contained in vesicles to other compartments or the apoplast for utilisation or detoxification of the metal. Furthermore, numerous enzymatic processes within compartments of the secretory pathway are Mn-dependent. The absence of the P2A-type ATPase AtECA3, located in the Golgi apparatus of Arabidopsis, causes growth defects under low Mn supply (Farthing et al., Reference Farthing, Henbest, Garcia-Becerra, Peaston and Williams2023; Mills et al., Reference Mills, Doherty, López-Marqués, Weimar, Dupree, Palmgren, Pittman and Williams2008) and also renders plants sensitive to high levels of Mn (Li et al., Reference Li, Chanroj, Wu, Romanowsky, Harper and Sze2008). The mechanistic basis of these phenotypes is unclear. Apart from ECA3, the Golgi apparatus of plants, like that of yeast and humans, contains proteins of the BICAT family. In Arabidopsis, AtBICAT3/PML3 has been shown to supply Mn for enzymatic activities in the trans-Golgi and to be localised primarily in these cisternae (He et al., Reference He, Yang, Hause, Rössner, Peiter-Volk, Schattat, Voiniciuc and Peiter2022). Import of Mn into Golgi cisternae is likely energised by the H+ gradient, based on a slightly more acidic pH by 0.5 units in the Golgi as compared to the cytosol (Shen et al., Reference Shen, Zeng, Zhuang, Sun, Yao, Pimpl and Jiang2013). Knockout of the ubiquitously expressed AtBICAT3/PML3 impairs cell expansion and thus reduces plant growth (He et al., Reference He, Yang, Hause, Rössner, Peiter-Volk, Schattat, Voiniciuc and Peiter2022; Yang et al., Reference Yang, Wang, Singh, Fan, Liu, Zhao, Cao, Xie, Yang and Huang2021). This was linked to an aberrant synthesis of cell wall matrix polysaccharides, in particular a severely decreased abundance of galactose moieties and 4-Gal linkages, which points to an affected β-1,4-galactan side chain substitution of rhamnogalacturonan I (He et al., Reference He, Yang, Hause, Rössner, Peiter-Volk, Schattat, Voiniciuc and Peiter2022). The glycosyltransferases mediating this reaction, AtGALS1 to 3, are localised in the trans-Golgi and Mn-dependent (He et al., Reference He, Rössner, Hoang, Alejandro and Peiter2021). This specific glycosylation defect suggests that Mn supply of the Golgi may be cisternae-specific, with AtBICAT3/PML3 supplying the trans subcompartment. However, this model is challenged by another study that claimed AtBICAT3/PML3 to be present primarily in the cis-Golgi and bicat3/pml3 mutants to be altered in protein glycosylation, leading to defects in cellulose synthesis and to display severe root tip degeneration under Mn deficiency (Yang et al., Reference Yang, Wang, Singh, Fan, Liu, Zhao, Cao, Xie, Yang and Huang2021). In addition, that study reported a hypersensitivity of bicat3/pml3 mutants to Mn toxicity, unlike the findings of He et al. (Reference He, Yang, Hause, Rössner, Peiter-Volk, Schattat, Voiniciuc and Peiter2022). The discrepancies between the two studies need to be resolved, and may hint at a dynamic regulation of AtBICAT3/PML3 localisation and function.
An ortholog of BICAT3/PML3 has been characterised in rice (Xu et al., Reference Xu, Wu, Liu, Tai, Zha, Gong, Zou, Zhang, Zhang and Chen2023). The consequences of a disrupted Mn supply of the Golgi in mutants for this transporter differ from those in Arabidopsis, with a major effect on hemicellulose biosynthesis. This is expected, as the cell wall composition of monocots differs from that of dicots.
A further role of AtBICAT3/PML3 was evident from the decreased seed setting of the mutants, caused by a male gametophyte defect (He et al., Reference He, Yang, Hause, Rössner, Peiter-Volk, Schattat, Voiniciuc and Peiter2022; Zhang, Zhang et al., Reference Zhang, Zhang, Liu, Fu and Luan2021). Tip-growing pollen tubes of bicat3/pml3 showed abnormal growth, which went along with a severely diminished deposition of partially methyl-esterified homogalacturonan (HG) in their cell walls (He et al., Reference He, Yang, Hause, Rössner, Peiter-Volk, Schattat, Voiniciuc and Peiter2022). HG is synthesised in the Golgi by proteins of the galacturonosyltransferase family. These harbour the DxD Mn-binding motif and are absolutely dependent on Mn (Amos et al., Reference Amos, Pattathil, Yang, Atmodjo, Urbanowicz, Moremen and Mohnen2018).
Intriguingly, chloroplastic Mn concentrations of Mn-deficient plants were increased in bicat3/pml3 mutants compared to the wild type, accompanied by a concomitant increase in photosynthetic activity (He et al., Reference He, Yang, Hause, Rössner, Peiter-Volk, Schattat, Voiniciuc and Peiter2022). This alteration may be explained by competition between Golgi and chloroplasts for Mn, indicating that the activity of AtBICAT3/PML3 prevents Mn to be allocated to other compartments. It remains to be studied if this effect is harnessed by the plant to efficiently allocate the metal according to subcellular requirements. On the tissue level, Mn accumulation was also altered in bicat3/pml3. The mutant showed an increased Mn translocation to the shoot, interestingly caused by its absence in the shoot, as determined in reciprocal grafting experiments (He et al., Reference He, Yang, Hause, Rössner, Peiter-Volk, Schattat, Voiniciuc and Peiter2022). This unexpected finding led to the hypothesis that Golgi Mn homeostasis may be involved in the regulation of Mn transport at the plasma membrane (Wege, Reference Wege2022).
Efflux transporters or channels for Mn have not been identified in the Golgi yet. The metal may thus be translocated by vesicular trafficking, potentially followed by exocytosis, as suggested for Mn exclusion involving the Golgi-localised Mn transporter AtMTP11 (Peiter et al., Reference Peiter, Montanini, Gobert, Pedas, Husted, Maathuis, Blaudez, Chalot and Sanders2007). Recently, relocalisation of AtMTP11 but not AtBICAT3/PML3 to the plasma membrane has been demonstrated upon inhibition of clathrin-mediated endocytosis, which supports an MTP11-dependent mechanism of vesicular Mn export by exocytosis (Vetal et al., Reference Vetal, Jaskolowski and Poirier2025). In addition, Mn may be released from the Golgi by yet unidentified transport mechanisms.
The TGN/EE-localised transporter AtNRAMP2 is required for Mn supply of chloroplasts, albeit it is primarily expressed in the vasculature (Alejandro et al., Reference Alejandro, Cailliatte, Alcon, Dirick, Domergue, Correia, Castaings, Briat, Mari and Curie2017). As suggested by Kosuth, Leskova, Castaings et al. (Reference Kosuth, Leskova, Castaings and Curie2023), the interplay of AtNRAMP2 with AtECA3, AtBICAT3/PML3 and AtMTP11 may determine the cellular fate of Mn, whereby the release of Mn from the secretory pathway by AtNRAMP2 prevents its further traffic and replenishes cytosolic Mn. The testing of this attractive model requires the identification of the missing players and their functional interactions, as well as new methods to quantify Mn levels with subcellular resolution.
6. Distribution control – the multilayered regulation of Mn transport
6.1. Transcriptional regulation of Mn transport
The highly dynamic availability of Mn in the soil and the differing Mn demand and sensitivity of tissues require the regulation of Mn transport activities. However, the restriction of Mn uptake under overload conditions is limited, largely due to the low selectivity of the transport systems. For example, the expression of AtIRT1, which imports Mn alongside Fe, is under Fe-dependent transcriptional control and is even increased by excessive Mn availability. Increased activity of AtIRT1 under Fe deficiency requires Mn detoxification, which in Arabidopsis is conferred by AtMTP8 sequestering Mn in the vacuole (Eroglu et al., Reference Eroglu, Meier, von Wirén and Peiter2016). Massive transcriptional upregulation of AtMTP8 is thereby controlled by FER-like iron deficiency-induced transcription factor (AtFIT), a central regulator of Fe deficiency-induced genes. Additionally, AtMTP8 is under developmental control in embryos of developing seeds and becomes activated at late developmental stages (Eroglu et al., Reference Eroglu, Giehl, Meier, Takahashi, Terada, Ignatyev, Andresen, Küpper, Peiter and von Wirén2017). Apart from its regulation by Fe and seed developmental status, direct transcriptional regulation of AtMTP8 by Mn has not been described. This is also true for many other Mn transporter genes, including AtMTP11 (Delhaize et al., Reference Delhaize, Gruber, Pittman, White, Leung, Miao, Jiang, Ryan and Richardson2007), AtBICAT1/PAM71 (Eisenhut et al., Reference Eisenhut, Hoecker, Schmidt, Basgaran, Flachbart, Jahns, Eser, Geimer, Husted, Webers, Leister and Schneider2018), or AtBICAT3/PML3 (He et al., Reference He, Yang, Hause, Rössner, Peiter-Volk, Schattat, Voiniciuc and Peiter2022), which are unaffected by Mn availability. In contrast, AtBICAT2/PAM71-HL/CMT1 is mildly downregulated by high Mn availability, which may serve to protect the chloroplast from Mn overload (Eisenhut et al., Reference Eisenhut, Hoecker, Schmidt, Basgaran, Flachbart, Jahns, Eser, Geimer, Husted, Webers, Leister and Schneider2018). Contrarily, Mn deficiency provokes a slight upregulation of AtNRAMP1 and AtNRAMP2 (Alejandro et al., Reference Alejandro, Cailliatte, Alcon, Dirick, Domergue, Correia, Castaings, Briat, Mari and Curie2017; Cailliatte et al., Reference Cailliatte, Schikora, Briat, Mari and Curie2010). However, directly Mn-responsive transcriptional networks and Mn sensors are unknown so far.
6.2. Post-transcriptional regulation of Mn transport
There is increasing evidence that, as with animals, alternative pre-mRNA splicing also plays an important role in diversifying the properties of proteins encoded by individual genes in plants. In rice, the alternative splicing of a large number of genes was found to be provoked by the deficiency of nutrients, including Mn (Dong et al., Reference Dong, He, Berkowitz, Liu, Cao, Tang, Shi, Wang, Li, Shen, Whelan and Zheng2018). Information on the role of this regulatory layer in Mn handling is limited. In Arabidopsis, the AtNRAMP6 gene has been shown to express a partially spliced variant, which, however, appears to be truncated and non-functional (Cailliatte et al., Reference Cailliatte, Lapeyre, Briat, Mari and Curie2009). In contrast, work on sugar beet demonstrated functional alterations of Mn transporters by differential pre-mRNA processing (Alejandro et al., Reference Alejandro, Meier, Hoang and Peiter2023) (Figure 2). Sugar beet produces two splice variants of BvMTP11, BvMTP11α and ß, with distinct localisation. BvMTP11ß localised to the Golgi, as described before for AtMTP11, whereas BvMTP11α was targeted to the vacuolar membrane. Both variants, which show different expression patterns, complement an Mn-sensitive yeast mutant and the Arabidopsis Atmtp11 mutant under excess Mn (Alejandro et al., Reference Alejandro, Meier, Hoang and Peiter2023). Alternative splicing may thus serve to direct intracellular Mn fluxes to different compartments, and even to alter the Mn distribution within the plant. In addition to a change in protein localisation, alternative splicing was shown to modulate the substrate spectrum of another Mn transporter in sugar beet. Sugar beet BvMTP9, the role of which is still unknown also in Arabidopsis, exhibits two splice variants, BvMTP9α and BvMTP9ß. The N-terminus of BvMTP9ß contains an additional 23 amino acids, which confer it with the ability to restore growth of an Fe-sensitive yeast strain on high-Fe medium, next to the complementation of an Mn-sensitive yeast, as found for BvMTP9α (Alejandro et al., Reference Alejandro, Meier, Hoang and Peiter2023). The N-terminus of BvMTP9ß contains a D/ExxD/E motif identified before to enable Fe transport by Mn-MTPs (Chu et al., Reference Chu, Car, Socha, Hindt, Punshon and Guerinot2017). The functional relevance of the regulation of selectivity or localisation by alternative splicing is unknown so far but potentially very significant.
Figure 2. Alternative splicing of BvMTP9 and BvMTP11 from sugar beet. The blue boxes, red boxes and black lines represent untranslated regions, exons and introns, respectively. The additional 23 amino acids in the N-terminus of BvMTP9ß contain a D/ExxD/E (DITE) motif (orange area), which enables BvMTP9ß to transport Fe next to Mn. An N-terminal dileucine residue (red area) in BvMTP11α results in its targeting the vacuole, whereas BvMTP11ß localises to the secretory pathway similar to Arabidopsis AtMTP11. Data on protein structures were obtained using alphafold3 (https://alphafoldserver.com/, Abramson et al. (Reference Abramson, Adler, Dunger, Evans, Green, Pritzel, Ronneberger, Willmore, Ballard, Bambrick, Bodenstein, Evans, Hung, O'Neill, Reiman, Tunyasuvunakool, Wu, Zemgulyte, Arvaniti, Beattie, Bertolli, Bridgland, Cherepanov, Congreve, Cowen-Rivers, Cowie, Figurnov, Fuchs, Gladman, Jain, Khan, Low, Perlin, Potapenko, Savy, Singh, Stecula, Thillaisundaram, Tong, Yakneen, Zhong, Zielinski, Zídek, Bapst, Kohli, Jaderberg, Hassabis and Jumper2024)), and models were created using PyMol (https://www.schrodinger.com/platform/products/pymol/).
6.3. Post-translational regulation of Mn transporter localisation
The subcellular localisation determines the physiological role of a transport protein. Trafficking of Mn transporters to and from the plasma membrane may thus regulate the Mn influx. Indeed, AtNRAMP1, the principal high-affinity uptake transporter, has been shown to cycle between the plasma membrane and endosomes (Castaings et al., Reference Castaings, Alcon, Kosuth, Correia and Curie2021). Localisation in the late endosome/multivesicular body compartment thereby depended on phosphatidylinositol 3-phosphate (PI3P) binding to the pleckstrin homology domain protein AtPH1 that localises in these vesicles (Agorio et al., Reference Agorio, Giraudat, Bianchi, Marion, Espagne, Castaings, Lelièvre, Curie, Thomine and Merlot2017). The absence of AtPH1 caused an accumulation of AtNRAMP1 at the vacuolar membrane. Another study indicated an involvement of choline in the plasma membrane insertion of AtNRAMP1 by regulating endocytosis, possibly through phospholipase D activity (Gao et al., Reference Gao, Chen, Chen, An, Lv, Han, Wang, Salt and Chao2017). Potentially toxic levels of Mn evoke the internalisation of AtNRAMP1 by clathrin-mediated endocytosis (Castaings et al., Reference Castaings, Alcon, Kosuth, Correia and Curie2021) (Figure 3). This response is dependent on the phosphorylation of Ser20 in the protein’s N-terminus. Phospho-dead mutations of this residue render AtNRAMP1 unable to undergo internalisation, and plants carrying this variant are hypersensitive to high Mn concentrations (Castaings et al., Reference Castaings, Alcon, Kosuth, Correia and Curie2021).
Figure 3. Schematic representation of phosphorylation processes in response to varying Mn levels. a. Under Mn deficiency, a delayed increase in [Ca2+]cyt occurs with oscillatory kinetics, likely activating AtCPK21; AtCPK23 is constitutively active. These kinases phosphorylate the Thr498 residue of AtNRAMP1, resulting in its activation and subsequent Mn uptake. b. In conditions of Mn excess, a rapid and transient elevation in [Ca2+]cyt activates the AtCBL1/9-AtCIPK23 module, leading to the phosphorylation of Ser20/22/24 of AtNRAMP1. This may trigger clathrin-mediated endocytosis of AtNRAMP1, thereby preventing further Mn uptake. c. Additionally, elevated [Ca2+]cyt activates AtCPK4/5/6/11, which phosphorylates Ser31/32 of AtMTP8, promoting Mn export into the vacuole. Subsequently, the AtCBL2/3-AtCIPK3/9/26 module phosphorylates Ser35, resulting in the deactivation of AtMTP8.
Two families of protein kinases play a principal role in the phosphorylation of transport proteins: calcium-dependent protein kinases (CPKs) and CBL-interacting protein kinases (CIPKs). AtCIPK23 was identified to phosphorylate several target sites in the N- and C-termini of AtNRAMP1, including Ser20 (Kosuth, Leskova, Ródenas et al., Reference Kosuth, Leskova, Ródenas, Vert, Curie and Castaings2023; Zhang et al., Reference Zhang, Fu, Xie, Wang, Zhao, Ma, Huang, Ju and Wang2023). One study indicated that high Mn-induced AtNRAMP1 internalisation was abolished in an Atcipk23 mutant, corresponding to Mn hypersensitivity of Atcipk23 (Zhang et al., Reference Zhang, Fu, Xie, Wang, Zhao, Ma, Huang, Ju and Wang2023). Phosphorylation of AtNRAMP1 by AtCIPK23 and the Mn hypersensitivity of an Atcipk23 mutant were also shown in another work (Kosuth, Leskova, Ródenas et al., Reference Kosuth, Leskova, Ródenas, Vert, Curie and Castaings2023). However, in this case, the AtCIPK23 knockout did not have an effect on AtNRAMP1 internalisation, indicating a more complex regulatory mechanism, which is not completely unexpected. First, AtCIPK23 regulates a plethora of ion transporters, including AtIRT1 that also transports Mn. Second, the Ser20 site has been shown before to be also phosphorylated by AtCPK21 and 23 without promoting internalisation (Fu et al., Reference Fu, Zhang, Wallrad, Wang, Höller, Ju, Schmitz-Thom, Huang, Wang, Peiter, Kudla and Wang2022), indicating that multiple signalling pathways may converge at this point. The regulation of AtNRAMP1 internalisation demands further attention.
Permeation of Mn by the primary Fe uptake transporter AtIRT1 may lead to excessive Mn accumulation. Similar to AtNRAMP1, the protein cycles between the plasma membrane and TGN/EE vesicles, thereby undergoing clathrin-mediated endocytosis and monoubiquitination (Barberon et al., Reference Barberon, Zelazny, Robert, Conéjéro, Curie, Friml and Vert2011). The internalisation is dependent on PI3P and mediated by non-Fe substrates (Barberon et al., Reference Barberon, Dubeaux, Kolb, Isono, Zelazny and Vert2014). Sensing of these ‘secondary’ metals by a histidine-rich domain triggers phosphorylation by AtCIPK23 and subsequent polyubiquitination by AtIDF1 to promote vacuolar targeting (Dubeaux et al., Reference Dubeaux, Neveu, Zelazny and Vert2018). AtIRT1 thus functions as a transceptor, both transporting and directly sensing non-Fe metal cations to avoid their overaccumulation (Cointry & Vert, Reference Cointry and Vert2019). The affinity of the metal-sensing motif in IRT1 varies for different metals, being lower for Cd compared to Zn or Mn (Spielmann et al., Reference Spielmann, Cointry, Devime, Ravanel, Neveu and Vert2022).
6.4. Post-translational regulation of Mn transporter function
Apart from regulating the trafficking of transporters, post-translational modifications may regulate their function. AtNRAMP1 is targeted by both CPKs and CIPKs, whereby phosphorylation by AtCPK21 and AtCPK23 promotes its activity by a yet unknown mechanism (Fu et al., Reference Fu, Zhang, Wallrad, Wang, Höller, Ju, Schmitz-Thom, Huang, Wang, Peiter, Kudla and Wang2022) (Figure 3). Both kinases phosphorylate Thr498, in addition to Ser20 and Ser22. Mutation of AtCPK21 and AtCPK23, as well as their target Thr498, render the plant hypersensitive to low Mn availability, and the phospho-dead nramp1T498A mutant is affected in Mn transport, as determined by yeast complementation (Fu et al., Reference Fu, Zhang, Wallrad, Wang, Höller, Ju, Schmitz-Thom, Huang, Wang, Peiter, Kudla and Wang2022). As described above, Ser20 is also targeted by AtCIPK23, which may trigger its internalisation under certain conditions (Kosuth, Leskova, Ródenas et al., Reference Kosuth, Leskova, Ródenas, Vert, Curie and Castaings2023; Zhang et al., Reference Zhang, Fu, Xie, Wang, Zhao, Ma, Huang, Ju and Wang2023). In addition, Ser499 is phosphorylated by AtCIPK23 (Kosuth, Leskova, Ródenas et al., Reference Kosuth, Leskova, Ródenas, Vert, Curie and Castaings2023), and the AtCPK21/23-targeted Thr498 might be as well (Zhang et al., Reference Zhang, Fu, Xie, Wang, Zhao, Ma, Huang, Ju and Wang2023). Further work is necessary to disentangle potential interactions of the phosphorylation sites and/or the kinases and to resolve the apparently conflicting outcomes of different studies.
In addition to AtNRAMP1, the vacuolar Mn transporter AtMTP8 has been reported to be under post-translational control (Figure 3). Phosphorylation in its N-terminal domain by four CPKs promotes its activity, and mutation of the phosphosites or knockout of these CPK genes renders the plants Mn-hypersensitive (Zhang, Fu et al., Reference Zhang, Fu, Sun, Ju, Miao, Wang, Xie, Ma, Gong and Wang2021). As described above, AtMTP8 plays a role specifically under Fe deficiency and in developing seeds, and it is only expressed under these conditions (Eroglu et al., Reference Eroglu, Meier, von Wirén and Peiter2016; Eroglu et al., Reference Eroglu, Giehl, Meier, Takahashi, Terada, Ignatyev, Andresen, Küpper, Peiter and von Wirén2017). Intriguingly, its function under those circumstances was absolutely dependent on the phosphorylation of Ser31 and Ser32 by AtCPK4, 5, 6 and/or 11 (Zhang, Fu et al., Reference Zhang, Fu, Sun, Ju, Miao, Wang, Xie, Ma, Gong and Wang2021). It remains to be determined whether individual CPKs operate in specific tissues expressing AtMTP8. Phospho-mimetic mutation of AtMTP8 caused an overaccumulation of Mn in root vacuoles similar to a several thousand-fold overexpression (Eroglu et al., Reference Eroglu, Meier, von Wirén and Peiter2016; Zhang, Fu et al., Reference Zhang, Fu, Sun, Ju, Miao, Wang, Xie, Ma, Gong and Wang2021).
AtMTP8 is furthermore a target of AtCIPK3, 9 and 26, which phosphorylate it primarily at Ser35 (Ju et al., Reference Ju, Zhang, Deng, Miao, Wang, Wallrad, Javed, Fu, Zhang, Kudla, Gong and Wang2022). This modification negatively regulates its activity, and the deletion of the CIPKs or their interacting calcineurin B-like protein 2 (AtCBL2) and 3 renders plants more Mn-tolerant. Similar to the regulation by CPKs, the antagonistic regulation by CIPKs affected all MTP8-related phenotypes described so far, and its abolishment drastically increased Mn concentrations in root vacuoles (Ju et al., Reference Ju, Zhang, Deng, Miao, Wang, Wallrad, Javed, Fu, Zhang, Kudla, Gong and Wang2022). After exposure of plants to toxic Mn levels, phosphorylation by AtCPK5 and AtCIPK26 was induced successively, leading to the assumption that AtMTP8 activated by CPKs upon short-term Mn stress is again inactivated by CIPKs during long-term Mn exposure (Ju et al., Reference Ju, Zhang, Deng, Miao, Wang, Wallrad, Javed, Fu, Zhang, Kudla, Gong and Wang2022). It remains to be elucidated how such a mechanism is integrated into long-term housekeeping functions of AtMTP8, such as Mn loading of developing embryos.
6.5. Calcium signals – a new determinant of Mn homeostasis
Protein kinases of the CPK and CIPK families are activated by cytosolic free Ca2+ ([Ca2+]cyt), the former by direct binding of Ca2+ and the latter by interaction with Ca2+-binding CBL proteins. If Ca2+ is bound with a high affinity in the range of the steady-state concentration, kinase activity occurs without a further elevation of [Ca2+]cyt. This is the case for CPK23 that activates NRAMP1 (Fu et al., Reference Fu, Zhang, Wallrad, Wang, Höller, Ju, Schmitz-Thom, Huang, Wang, Peiter, Kudla and Wang2022) and may serve to unblock the transporter once it has reached its target membrane. Kinases binding Ca2+ with lower affinity require an elevation of [Ca2+]cyt for activation, which was observed under both Mn deficiency and surplus. Depletion of Mn triggered unique, long-lasting, and very slow [Ca2+]cyt oscillations in Arabidopsis roots (Fu et al., Reference Fu, Zhang, Wallrad, Wang, Höller, Ju, Schmitz-Thom, Huang, Wang, Peiter, Kudla and Wang2022), which initiated in the centre of the elongation zone after 2 hours of Mn depletion, and thereafter expanded to cell layers of the meristematic and differentiation zones. The oscillations continued for at least 6 hours at a frequency of 30 min (Fu et al., Reference Fu, Zhang, Wallrad, Wang, Höller, Ju, Schmitz-Thom, Huang, Wang, Peiter, Kudla and Wang2022). High-resolution imaging revealed that those Ca2+ elevations originated in cells of the epidermis, extended to cortical cells within 5 min, and continued with increasing intensity in the stele. These signals may be decoded by CPK21 to activate NRAMP1, which operates on the background of constitutive activation by CPK23.
High Mn availability also triggered a [Ca2+]cyt elevation but with markedly different kinetics. Upon exposure to 1.5 mM Mn, a single increase was observed, either instantaneously or slowly rising within several minutes, as detected by aequorin luminescence and GCamP6f fluorescence, respectively (Fu et al., Reference Fu, Zhang, Wallrad, Wang, Höller, Ju, Schmitz-Thom, Huang, Wang, Peiter, Kudla and Wang2022; Zhang, Fu et al., Reference Zhang, Fu, Sun, Ju, Miao, Wang, Xie, Ma, Gong and Wang2021). The Ca2+ signal in response to excessive Mn was initiated in cells of the elongation zone, finding its maximum in the cortical cell layer. This Ca2+ elevation may be decoded by AtCBL1/9 that interacts with AtCIPK23, thereby inactivating and/or internalising AtNRAMP1 and AtIRT1 (Kosuth, Leskova, Ródenas et al., Reference Kosuth, Leskova, Ródenas, Vert, Curie and Castaings2023; Zhang et al., Reference Zhang, Fu, Xie, Wang, Zhao, Ma, Huang, Ju and Wang2023). Intriguingly, the AtCBL1/9-AtCIPK23 module also regulates K+ uptake by AtAKT1 (Xu et al., Reference Xu, Li, Chen, Wang, Liu, He and Wu2006) and NO3− uptake by AtNRT1.1 (Ho et al., Reference Ho, Lin, Hu and Tsay2009). It is still a largely open question how the specificity of AtCIPK23 activity is conferred. In addition, AtCPK4/5/6/11 and the AtCBL2/3-AtCIPK3,9,26 module are likely activated by the high-Mn-triggered [Ca2+]cyt transient to regulate AtMTP8 function (Ju et al., Reference Ju, Zhang, Deng, Miao, Wang, Wallrad, Javed, Fu, Zhang, Kudla, Gong and Wang2022; Zhang, Fu et al., Reference Zhang, Fu, Sun, Ju, Miao, Wang, Xie, Ma, Gong and Wang2021). However, as mentioned above, one role of AtMTP8 lies in the Mn loading of vacuoles in developing embryos, which likely demands a constitutive activity of AtMTP8 that is difficult to reconcile with Ca2+ signal-mediated activation.
The mechanisms generating low- and high-Mn-induced Ca2+ signals, as well as the Mn sensors upstream of those signals, are unknown.
7. Outlook
Over the last few years, we have seen a substantial progress in our understanding of Mn transport and its regulation in plants, in particular regarding the supply of organellar compartments and the post-transcriptional regulation of transporters. However, the co-operation of transport proteins operating in the same membrane or different organelles is still not well understood, albeit a picture is beginning to unfold, with the secretory pathway playing a central role. On the whole-plant level, the allocation of Mn and its handling in the vascular system is another field requiring attention, also with respect to the improvement of crop performance.
Our understanding of Mn homeostasis has primarily been derived from work on Arabidopsis in addition to rice, which is a monocot adapted to submerged soils with high Mn availability. The sketchy evidence available in a few other species already shows that there is a marked variability in Mn handling and that the toolbox of Mn transporters is used in a highly versatile manner, which calls for more extensive mechanistic studies in non-model species.
Our quantitative understanding of the feedback regulation of free and total Mn concentrations at different scales is still in its infancy. Advances in this area will require new methodological capabilities, such as Mn reporters, to assess the relevance, cooperation and regulation of transport proteins in planta. The design of specific reporters for Mn is challenging due to its low ligand affinity, but progress has recently been made in this area for a bacterial Mn reporter (Park et al., Reference Park, Cleary, Li, Mattocks, Xu, Wang, Mukhopadhyay, Gale and Cotruvo2022). Finally, current discrepancies between studies from different laboratories may indicate a physiologically relevant adaptive capacity of the Mn-handling machinery, which can only be elucidated through a collaborative effort among the manganese community.
Data availability statement
No datasets were generated or analysed in this article.
Author contributions
E.P. conceived the article. B.M., O.M. and E.P. wrote the first draft and designed the figures. All authors contributed to the editing of the manuscript and approved its final version.
Funding statement
This work was supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) within the Research Training Group 2498 (E.P., grant number 400681449) and the Collaborative Research Centre 1664 (E.P., grant number 514901783) and by COST Action CA 19116 (‘Trace metal metabolism in plants – PLANTMETALS’ to E.P.).
Competing interest
The authors declare none.
Open peer review
To view the open peer review materials for this article, please visit http://doi.org/10.1017/qpb.2025.10012.
Open access
Comments
Dear Ingo,
please find attached our manuscript entitled “Manganese handling in plants: advances in the mechanistic and functional understanding of transport pathways”, which is a revision of our previous manuscript “Manganese in plants - its journey from soil to cell”. We submit this revised manuscript as a solicited Review Article to be considered for publication in the focus issue “Quantitative approaches to cellular aspects of plant ion homeostasis” of Quantitative Plant Biology. We apologize for the delay of our submission.
We thank the two anonymous reviewers for their valuable comments to the manuscript. In response to those suggestions, we have modified and clarified the text at many instances. All alterations are highlighted in the marked version of the manuscript. A new citation has been included providing novel insight into regulation of the Mn transporter MTP11 (Vetal et al., 2025). There is very little quantitative information on plant Mn homeostasis in the literature. We have therefore included the kinetics of yeast and mammalian BICAT homologs, as information of the plant members is lacking. In addition, recent progress in the development of selective Mn reporters is now mentioned (Park et al., 2022).
Implementing all of the changes suggested by referee 1 would have largely changed the focus of the review in a way that seemed undesirable to us. Our approach of devoting individual chapters to processes at different levels of scale (in different depths) received very positive and confirmatory comments from the second referee, and restructuring the manuscript is unlikely to be seen as an improvement. We hope that our revisions and amendments are satisfactory.
We confirm that this manuscript has not been published elsewhere and is not under consideration elsewhere, including the internet. All authors have approved the manuscript and agree with its submission to Quantitative Plant Biology.
With best wishes, Edgar