Mother-plant-mediated pumping of zinc into the developing seed
Résumé
1 Insufficient intake of zinc and iron from a cereal-based diet is 2 one of the causes of "hidden hunger" (micronutrient 3 deficiency), which affects some two billion people 1,2. 4 Identifying a limiting factor in the molecular mechanism of 5 zinc loading into seeds is an important step towards determin-6 ing the genetic basis for variation of grain micronutrient 7 content and developing breeding strategies to improve this 8 trait 3. Nutrients are translocated to developing seeds at a 9 rate that is regulated by transport processes in source leaves, 10 in the phloem vascular pathway, and at seed sinks. Nutrients 11 are released from a symplasmic maternal seed domain into 12 the seed apoplasm surrounding the endosperm and embryo 13 by poorly understood membrane transport processes 4-6. 14 Plants are unique among eukaryotes in having specific P1B-15 ATPase pumps for the cellular export of zinc 7. In Arabidopsis, 16 we show that two zinc transporting P1B-ATPases actively 17 export zinc from the mother plant to the filial tissues. Mutant 18 plants that lack both zinc pumps accumulate zinc in the 19 seed coat and consequently have vastly reduced amounts of 20 zinc inside the seed. Blockage of zinc transport was observed 21 both at high and low external zinc supplies. The phenotype 22 was determined by the mother plant and is thus due to a 23 lack of zinc pump activity in the seed coat and not in the 24 filial tissues. The finding that P1B-ATPases are one of the 25 limiting factors controlling the amount of zinc inside a seed 26 is an important step towards combating nutritional zinc 27 deficiency worldwide. 28 In the quest to identify the membrane-bound transporters that 29 deliver zinc into seeds, heavy metal-transporting P1B-ATPases, 30 which belong to the family of P-type ATPases that actively trans-31 ports ions and lipids across membranes, are promising candidates. 32 In Arabidopsis thaliana, the P1B-ATPase subfamily consists of 33 eight members, AtHMA1 Q2-8 (ref. 8). Whereas AtHMA2-4 transport 34 divalent zinc and cadmium cations and AtHMA5-8 transport 35 monovalent copper and silver cations, AtHMA1 has been impli-36 cated in the transport of a number of different cations including 37 copper, silver, zinc and cadmium 9,10. AtHMA3 is localized to the 38 tonoplast and is involved in detoxification by sequestering excess 39 amounts of zinc and cadmium into the vacuole 11. AtHMA2 and 40 AtHMA4 are structurally very similar to each other and have over-41 lapping functions. Whereas the hma2 and hma4 single mutants 42 have wild-type growth rates, the hma2,hma4 double mutant 43 shows a stunted growth phenotype that can only be rescued by 44 zinc supplementation. Also, the hma2,hma4 double mutant does 45 not set seed, a phenotype mainly ascribed to a defect in pollen 46 development 12. Both AtHMA2 and AtHMA4 have been localized to 47 the plasma membrane and function as cellular zinc exporters 12-14. 48 Their involvement in xylem loading of zinc in roots is well known. 49 They are both expressed in the root pericycle and xylem parenchyma 50 cells and mutants lacking functional AtHMA2 and AtHMA4 51 accumulate zinc in the pericycle and endodermal cell layers, and 52 exhibit reduced root-to-shoot translocation of zinc 12,13. The 53 finding that the zinc hyperaccumulator Arabidopsis halleri accumu-54 lates high levels of zinc in the shoot mainly because of the strongly 55 enhanced expression of AhHMA4, caused by activating cis-regulat-56 ory mutations and gene copy number expansion 15 , emphasizes the 57 role of HMA4 in the root-to-shoot translocation of zinc. 58 To test whether developing seeds depend on AtHMA2 and 59 AtHMA4 for proper zinc unloading from the seed coat, we analysed 60 the amount of zinc present in the maternal and filial parts of the 61 seed, respectively, in wild-type and mutant seeds lacking functional 62 AtHMA2, AtHMA4, or both. For this purpose, we developed an 63 assay in which mature seeds are separated into seed coat (with the 64 endosperm attached) and embryo fractions, and their respective 65 metal contents are subsequently analysed by inductively coupled 66 plasma mass spectrometry (ICPMS). All seeds used for this analysis 67 were from plants grown side by side in soil watered with 3 mM 68 ZnSO 4 twice a week, as these conditions were needed for the 69 hma2,hma4 double mutant to set seed. At this relatively high zinc 70 supply, trafficking of zinc to the shoot did not differ between geno-71 types (Supplementary Fig. 1). We determined the total amounts of 72 zinc, magnesium, phosphorus and manganese in batches of ten 73 mature seeds, either intact or separated into seed coat and embryo 74 fractions, using ICPMS. This revealed that the content of the differ-75 ent genotypes or between the content of intact seeds and the com-76 bined content of separated seeds for the tested elements showed 77 little variation (Supplementary Fig. 2). As there was no difference 78 between the weight of seeds of different genotypes and the com-79 bined weight of their separated parts (Supplementary Fig. 3a), loss 80 of material during the separation procedure was negligible. 81 However, the weight of any ten seeds was variable, which explains 82 the variance seen for the total amounts of element. Instead of weigh-83 ing samples before ICPMS, which could result in contamination, we 84 used the ratio of seed coat weight to embryo weight as a reference 85 point for normalizing data as it consistently was found to be very 86 similar between and within genotypes (Supplementary Fig. 3b). 87 We found that in wild-type seeds most of the zinc was present in 88 the embryo fraction (Fig. 1a). Conversely, in the hma2,hma4 double
Domaines
Biologie végétale
Origine : Fichiers éditeurs autorisés sur une archive ouverte