Fertility control: the role of magnesium transporters in pollen development

Main

Magnesium (Mg²⁺) is essential for many plant processes. It is particularly important for photosynthesis, where it is the central element of the chlorophyll molecules. Magnesium is also a constituent of many enzymes and serves as an activator of others, including DNA polymerases, protein kinases, and phosphatases. Intracellular Mg²⁺ concentration is precisely controlled by Mg²⁺ uptake, translocation across intracellular membranes and efflux ¹. Surprisingly, however, the molecular mechanisms underlying these processes in higher plants are still largely unknown. Recently, two reports ^{2, 3} disclosed functions of two Mg²⁺ transporters in pollen development. Pollen grains are the male gametes in flowering plants.

Mg²⁺ transporters of the CorA family appear to be the primary transport systems mediating both influx and efflux of Mg²⁺. CorA-type transporters are widely distributed in bacteria, fungi, animals and in plants (see below). The crystal structure of the CorA transporter from the bacterium Thermotoga maritima has recently been elucidated revealing a unique structure with a large, acidic N-terminal domain and two transmembrane domains in the C-terminal region ⁴. The currently best studied eukaryotic homologues are those from budding yeast (Saccharomyces cerevisiae), where the MRS2 and LPE10 proteins are targeted towards the inner mitochondrial membrane, while ALR1, ALR2 and MNR2 transporters are located in the plasma membrane ⁵.

Higher plant transporters of the CorA/MRS2 family were first identified in 2000, when Schock et al. ⁶ reported a gene family of ten members in Arabidopsis thaliana, named AtMRS2-1 to AtMRS2-10. These authors also demonstrated that AtMRS2-1 functionally complements Mrs2- deficient yeast restoring intramitochondrial Mg²⁺ concentration to the wild-type level.

Functional evidence for Mg²⁺ transport was also provided by Li et al. ⁷ who in 2001 discovered the same gene family, but named it as AtMGT (for A rabidopsis t haliana magnesium transporter). Most members of the AtMGT gene family were shown by RT-PCR to be expressed throughout the plant, with the exception of AtMGT5 which was only expressed in flowers, and AtMGT8 that was not expressed in stem tissue. AtMGT1 functionally complemented a bacterial (Salmonella typhimurium) mutant called MM281 that lacks three Mg²⁺ transport systems (CorA, MgtA and MgtB). AtMGT1 turned out to be a high-affinity transporter for Mg²⁺. Among divalent cations it was highly selective for Mg²⁺. AtMGT10 complemented a yeast mutant, lacking both ALR1 and ALR2 genes ⁷, although it localizes to the chloroplast envelope membrane when expressed in plant cells, indicating a role in Mg²⁺ uptake and translocation into chloroplasts ⁸. Recently the first low-affinity Mg²⁺ transporter, AtMGT7a, was identified in higher plants. Two different transcripts are produced from the AtMGT7 gene; AtMGT7a transcript is expressed in all tissues analyzed, whereas the slightly shorter AtMGT7b transcript is present in roots and flowers ⁹. Only AtMGT7a complemented the Salmonella MM281 mutant, whereas AtMGT7b, which encodes a protein lacking a 15-amino acid region, did not.

The biological roles of all above mentioned Mg²⁺ transporters are still unknown although ectopic expression of AtMGT1 in Nicotiana benthamiana conferred improved tolerance to aluminum ¹⁰. However, a physiological function was recently discovered for AtMGT5 which represents a dual-functional Mg²⁺ transporter that mediates uptake and efflux of Mg²⁺ in a concentration-dependent manner when expressed in the MM281 strain ². At low (i.e., micromolar) concentration of Mg²⁺ the AtMGT5 protein serves as an importer, whereas at high (millimolar) Mg²⁺ levels it facilitates efflux of the divalent cation. Expression of AtMGT5-GFP fusion protein in transgenic Arabidopsis plants labeled mitochondria (verified by MitoTracker Red CMXRos staining), suggesting that it shuttles Mg²⁺ between the mitochondrial compartment and the cytoplasm. Promoter-reporter (β-glucuronidase) fusion studies revealed exclusive expression of AtMGT5 in anthers at earlier stages of flower development, in both gametophytic and sporophytic tissues ². Interestingly, when AtMGT5 was mutated by T-DNA insertion, pollen formation was strongly impaired in hemizygous mutants; homozygous atmgt5 mutants were never obtained, indicating lethality of a complete knock-out of the AtMGT5 gene.

The story became even more fascinating when Chen et al. ³ described a second Mg²⁺ transporter, AtMGT9, which also affected pollen viability. AtMGT9 supported the growth of the bacterial MM281 mutant at 500 μM Mg²⁺, but not at 10 μM Mg²⁺, reminiscent of a low-affinity transporter. Tracer inhibition studies, that utilize the fact that Mg²⁺ and Ni²⁺ can be transported by the same CorA-type transporters with similar kinetics, revealed that AtMGT9 is able to transport Mg²⁺ at physiological concentrations normally found in plant cells, whereas the transport of other divalent ions – trace elements – appeared unlikely due to their much lower levels prevalent in plants. Transcriptional activity of AtMGT9 in anthers followed a distinct temporal and spatial pattern. Its expression was evenly distributed throughout the anther early in development and was then concentrated in the tapetum, the tissue nursing pollens during their formation. AtMGT9 expression vanished when the tapetum was degraded and resumed again during pollen maturation. As with AtMGT5, disruption of AtMGT9 led to pollen abortion, and only hemizygous mutants survived. Crossing experiments proved that the mutated atmgt9 allele could not be transmitted through the pollen due to lack of fertility ³.

The works by Luan and co-workers ^{2, 3} show a tight connection between Mg²⁺ transporter function and pollen development for the first time, although we do not know in detail how the two transporters sustain pollen formation and thus fertility. One model supposes that both proteins are located in mitochondria (though the localization of AtMGT9 still has to be investigated) but function independently to contribute to mitochondrial Mg²⁺ homeostasis. Alternatively, both proteins may constitute subunits of a hetero-oligomeric transporter. However, as both AtMGT5 and AtMGT9 seemingly form functional transporters when individually expressed in bacterial cells, it appears likely that functional (homo-oligomeric) transporters can also form in planta. Another scenario is that AtMGT9 resides in a non-mitochondrial membrane to exert its transport function jointly with mitochondrial AtMGT5. None of these models can be excluded at present.

A further interesting asset comes from the fact that both magnesium transporter genes follow distinct, partially overlapping tempo-spatial activity patterns during anther development. Identifying the transcription factors that govern their expression will contribute to our understanding of the regulatory networks that control pollen viability through the function of AtMGT5 and AtMGT9 transporters.