Abstract
Although asymmetric deposition of the plant extracellular matrix is critical for the normal functioning of many cell types, the molecular mechanisms establishing this asymmetry are not well understood. During differentiation, Arabidopsis seed coat epidermal cells deposit large amounts of pectin-rich mucilage asymmetrically to form an extracellular pocket between the plasma membrane and the outer tangential primary cell wall. At maturity, the mucilage expands on contact with water, ruptures the primary cell wall, and extrudes to encapsulate the seed. In addition to polysaccharides, mucilage contains secreted proteins including the β-galactosidase MUCILAGE MODIFIED 2 (MUM2). A functional chimeric protein where MUM2 was fused translationally with Citrine yellow fluorescent protein (Citrine) indicated that MUM2-Citrine fluorescence preferentially accumulates in the mucilage pocket concomitant with mucilage deposition and rapidly disappears when mucilage synthesis ceases. A secreted form of Citrine, secCitrine, showed a similar pattern of localization when expressed in developing seed coat epidermal cells. This result suggested that both the asymmetric localization and rapid decrease of fluorescence is not unique to MUM2-Citrine and may represent the default pathway for secreted proteins in this cell type. v-SNARE proteins were localized only in the membrane adjacent to the mucilage pocket, supporting the hypothesis that the cellular secretory apparatus is redirected and targets secretion to the outer periclinal apoplast during mucilage synthesis. In addition, mutation of ECHIDNA, a gene encoding a TGN-localized protein involved in vesicle targeting, causes misdirection of mucilage, MUM2 and v-SNARE proteins from the apoplast/plasma membrane to the vacuole/tonoplast. Western blot analyses suggested that the disappearance of MUM2-Citrine fluorescence at the end of mucilage synthesis is due to protein degradation and because several proteases have been identified in extruded seed mucilage. However, as mutation of these genes did not result in a substantial delay in MUM2-Citrine degradation and the timing of their expression and/or their intracellular localization were not consistent with a role in MUM2-Citrine disappearance, the mechanism underlying the abrupt decrease of MUM2-Citrine remains unclear.