Moving magnesium in plant cells
Provide feedback or get help You are viewing our new enhanced HTML article.
If you can't find a tool you're looking for, please click the link at the top of the page to "Go to old article view". Alternatively, view our Knowledge Base articles for additional help. Your feedback is important to us, so please let us know if you have comments or ideas for improvement.
Magnesium (Mg) is among the most abundant mineral elements in plants, yet the knowledge of which genes control its accumulation in specific tissues and organelles lags behind that of many other mineral elements. Only in recent years has identification of important molecular players begun to take shape. In this issue of New Phytologist, Conn et al. (pp. 583–594) shed additional light on two Mg transporters that play important roles in accumulation of Mg in leaf cell vacuoles. Using sub-cellular level ion measurements on leaves, gene expression measurements after single cell sampling, a genetic approach, and clever use of calcium (Ca) and Mg supply to plants or detached leaves, Conn et al. have demonstrated that vacuoles of mesophyll rather than epidermal or bundle sheath cell types of Arabidopsis leaves are the main sites of Mg accumulation. They have also shown the effects of mutations in MRS2-1 and MRS2-5 genes on this accumulation, and a role for these genes in plant adaptation to adverse soil environments.
‘Curiously, unlike transporters for most other mineral elements, short- and long-term Mg deficiency did not induce expression of MRS2 family genes.’
Biological roles for magnesium
Total Mg concentrations in cells range from 15 to 25 mM (Moomaw & Maguire, 2008). However, most Mg ions are bound or incorporated into cellular components, which leaves free cytosolic Mg2+ in the range of 0.4 to 0.5 mM (Karley & White, 2009; Maathuis, 2009). Typically 15–20% of total leaf Mg is in chlorophyll (White & Broadley, 2009), although this percentage can be higher or lower depending on plant Mg status (Marschner, 1995). The largest pool of Mg is devoted to protein synthesis by bridging ribosome subunits. Magnesium also coordinates with nucleotides and nucleic acids, and bridges enzyme and substrate for many types of reactions including phosphorylation and dephosphorylation (Marschner, 1995; Maathuis, 2009). Phloem loading of carbohydrates depends on adequate Mg levels, as Mg deficient leaves accumulate starch and sugars (Marschner, 1995; Karley & White, 2009). Free Mg is primarily stored in leaf cell vacuoles, organelles that are a driving force for cell expansion by providing an osmotic potential. Certain adverse soil environments, known as serpentine soils, have low Ca and high Mg concentrations. Plants species tolerant to these environments typically have lower leaf Mg concentrations, and polymorphisms in some Mg transporters have been associated with increased tolerance to serpentine soils (Turner et al., 2010).
Microbial magnesium transporters
Magnesium transporters were first identified in bacteria in a screen for cobalt (Co) resistance, and were thus named CorA (Moomaw & Maguire, 2008). The CorA protein acts as an ion channel that transports Mg, Co and nickel across membranes with their concentration gradients. The functional CorA protein is a homopentamer, where each subunit has two transmembrane domains, with most of the protein oriented to the inside of the cytosol (Moomaw & Maguire, 2008). CorA proteins form a superfamily with members also present in eukaryotes.
Fungal homologues of CorA were identified in Saccharomyces cerevisiae in screens that were not specifically designed to find Mg transporters. Nontheless, there are five CorA family genes in total that mediate uptake across the plasma membrane, uptake into mitochondria, and efflux from vacuoles. The first yeast Mg transporters were discovered because they conferred aluminum resistance when overexpressed, and were named ALR1 and ALR2 (MacDiarmid & Gardner, 1998). These genes function for Mg uptake across the plasma membrane into the cells. Shortly thereafter, a mitochondrial Mg transporter, MRS2 (Bui et al., 1999), was identified for its ability to correct an RNA splicing defect, demonstrating another important role for Mg in cellular biochemistry. Based on sequence similarity, a second mitochondrial Mg transporter, LPE10 (Gregan et al., 2001), was also characterized. More recently, the fifth yeast CorA homologue, MNR2, has been shown to be necessary for cells to access Mg stored in the vacuole (Pisat et al., 2009). However, the genes that load Mg into the yeast vacuole have not been identified, and are apparently not CorA family members.
Arabidopsis magnesium transporters
Compared to single celled organisms, plants have additional challenges in distributing Mg to required locations. In addition to uptake into root cells, plants will have various other uptake and efflux steps as vascular tissues are loaded and unloaded, and Mg is distributed away from or into xylem and phloem and associated parenchyma. Also, plant cells have plastids, and Mg must be imported into this organelle for chlorophyll synthesis. Arabidopsis thaliana has several CorA/MRS2 homologues, with nine genes (Table 1) and two pseudogenes. The first Arabidopsis CorA homologues were identified in expressed sequence tag (EST) databases by homology to yeast MRS2. Arabidopsis MRS2-1 functionally complemented the yeast mrs2 mutant (Schock et al., 2000). A complementation screen of the yeast alr1/alr2 Mg uptake mutant strain identified MGT10 (also called MRS2-11) as a high affinity Mg transporter (Li et al., 2001), which was later localized to the chloroplast (Drummond et al., 2006). A second family member, MGT1 (MRS2-10) was capable of complementing a bacterial CorA mutant, and was localized to the plasma membrane. Both of these research groups described the remaining family members (Schock et al., 2000; Li et al., 2001). It has now been demonstrated that all the family members can complement yeast mrs2 mutants if the genes are fused to yeast native mitochondrial targeting sequences (Gebert et al., 2009).
Table 1. Some characteristics of the MRS2/MGT gene family of Arabidopsis thaliana
|At1g80900||MGT1, MRS2-10||Y||4, 8||Y||2||PM||2|| || ||261894_at||74.4 (0.7)||Senescing leaves||160||ATGE_25|
|Stage 15 petals||183||ATGE_42|
|At1g16010||MGT2, MRS2-1||Y||1, 4, 8|| || ||Vacuole||9, 10|| || ||261795_at||354.6 (2.7)||Senescing leaves||1011||ATGE_25|
|At2g03620||MGT3, MRS2-5||Y||8|| || ||Vacuole||9, 11|| || ||npc|| || || || |
|At3g19640||MGT4, MRS2-3||Y||8|| || || || || || ||257019_at||298.8 (4.2)||Flower, stage 9||726||ATGE_31|
|Seeds, stage 6||744||ATGE_79|
|At4g28580||MGT5, MRS2-6||Y||8||Y||5||Mitochondria||5||Pollen development defect||5||253781_at||18.2 (0.5)||Mature pollen||57||ATGE_73|
|Flower, stage 10/11||38||ATGE_32|
|At3g58970||MGT6, MRS2-4||Y||8|| || || || || || ||251508_at||147.2 (1.8)||Senescing leaves||407||ATGE_25|
|At5g09690||MGT7, MRS2-7||Y||8||Y||7||Endomembrane system||8||Growth defect on low Mg||8||250487_at||55.5 (1.1)||Root||137||ATGE_99|
|At5g64560||MGT9, MRS2-2||Y||8||Y||6|| || ||Pollen development defect||6||247248_at||132.6 (1.6)||Stamen, stage 15||604||ATGE_43|
|At5g22830||MGT10, MRS2-11||Y||2, 4, 8||Y||5||Chloroplast||4|| || ||249864_at||395.8 (6.0)||Mature pollen||93||ATGE_73|
Most of the AtMRS2 genes are found on the Affymetrix ATH1 microarray, and the transcript expression levels for MRS2 genes have been quantified in major tissues over plant development (Schmid et al., 2005). These studies have given clues to physiological functions of the individual genes of the family. Curiously, unlike transporters for most other mineral elements, short (28 h) term (Hermans et al., 2010a) and long (1 wk) term (Hermans et al., 2010b) Mg deficiency did not induce expression of MRS2 family genes. This lack of upregulation of Mg transport has contributed to the difficulty in identifying root Mg uptake genes. Recently, it was shown that a mutation in MRS2-7 resulted in a growth defect on low Mg nutrient solution (Gebert et al., 2009), implicating this gene in uptake. Consistent with this, developmental expression studies indicated highest expression of MRS2-7 in roots, and lowest in leaves (Table 1). An MRS2-7-GFP fusion indicated localization in the endomembrane system (Gebert et al., 2009). The authors also tested knockouts in MRS2-1, MRS2-5 and MRS2-10, and constructed double mutants of mrs2-5/2-1 and mrs2-5/2-10, none of which showed a growth defect phenotype on low Mg. Knockouts of two genes, MRS2-6 (Li et al., 2008) and MRS2-2 (Chen et al., 2009), were shown to have defects in pollen development, and consistent with this, both have higher than average expression levels in floral parts or pollen. MRS2-6 protein was localized to the mitochondria. MRS2-3 also has high expression in flowers and in developing seeds, and this gene co-localized with a Quantitative Trait Locus (QTL) for seed Mg concentration (Waters & Grusak, 2008a). On the contrary, MRS2-10 and MRS2-11 have quite low expression in pollen. MRS2-11 also has low expression in root tissue, but high expression levels in rosette leaves, which fits well with its chloroplast localization (Drummond et al., 2006). Expression of two family members, MRS2-4 and MRS2-10, are increased in the oldest flower petals, in cauline leaves, and in senescing rosette leaves, which are tissues that supply certain minerals to developing seeds through remobilization (Waters & Grusak, 2008b).
New insights into vacuolar magnesium accumulation and fitness
Conn et al. continue to advance our knowledge of roles of MRS2 genes in Mg localization within leaves of Arabidopsis. Using careful and detailed measurements of Mg in specific cell types in leaves, Conn et al. demonstrated that Mg was at its highest levels in the mesophyll rather than epidermal or bundle sheath cells, and that after increasing supply of Mg to leaves, Mg concentration increases primarily in the vacuoles of palisade and spongy mesophyll cells. This was followed by cell-type specific expression analysis, both by microarray and real-time reverse transcription-polymerase chain reaction (RT-PCR), which established expression levels of specific MRS2 genes in palisade mesophyll and epidermal cells. Two of these genes, MRS2-1 and MRS2-5, were the subject of further study. Both genes were localized to the vacuole, which corresponded with previous proteomic data (Alexandersson et al., 2004; Carter et al., 2004; Whiteman et al., 2008). By using a genetic approach coupled with manipulation of the environment, that is, Ca and Mg supply to cells, Conn et al. were able to demonstrate a subtle phenotype showing that MRS2-1 and MRS2-5 are important for Mg accumulation in vacuoles of mesophyll cells. This becomes more important in serpentine soil environments, when soil Ca levels are greatly reduced relative to Mg. In these conditions, the authors hypothesize that Mg can substitute for Ca as an osmoticum in leaf vacuoles to drive expansion of leaf cells for growth. Indeed, in low Ca solution, vacuole osmolality in mrs2-1 and mrs2-5 was altered and shoot growth rate was reduced, conditions that did not occur under normal Ca supply. Interestingly, when Arabidopsis was grown in low Ca solution, several MRS2 genes had increased expression, a phenomenon not observed under Mg deficiency (Hermans et al., 2010a,b), suggesting that plant cells may compensate for low Ca by increasing Mg transport. The increased MRS2 expression was even more pronounced in the cax1/cax3 Ca transporter double mutant. Mutations in CAX1 were previously shown to confer tolerance to serpentine conditions (Bradshaw, 2005) and result in a higher Mg requirement for optimal growth, and lower leaf Mg concentrations.
A greater understanding of the roles of Mg transporters in plant physiology and development, as well as adaptation to specific environments, could have widespread applications in agriculture. For example, breeding crop varieties that could better grow in adverse environments, such as low Ca soils, could raise yields or allow food to be produced in areas that are not currently usable. In addition, production of new varieties biofortified with minerals such as Mg would have beneficial effects on human health (White & Broadley, 2009).