Nucleus- and plastid-targeted annexin 5 promotes reproductive development in Arabidopsis and is essential for pollen and embryo formation

Abstract

Background: Pollen development is a strictly controlled post-meiotic process during which microspores differentiate into microgametophytes and profound structural and functional changes occur in organelles. Annexin 5 is a calcium- and lipid-binding protein that is highly expressed in pollen grains and regulates pollen development and physiology. To gain further insights into the role of ANN5 in Arabidopsis development, we performed detailed phenotypic characterization of Arabidopsis plants with modified ANN5 levels. In addition, interaction partners and subcellular localization of ANN5 were analyzed to investigate potential functions of ANN5 at cellular level.

Results: Here, we report that RNAi-mediated suppression of ANN5 results in formation of smaller pollen grains, enhanced pollen lethality, and delayed pollen tube growth. ANN5 RNAi knockdown plants also displayed aberrant development during the transition from the vegetative to generative phase and during embryogenesis, reflected by delayed bolting time and reduced embryo size, respectively. At the subcellular level, ANN5 was delivered to the nucleus, nucleolus, and cytoplasm, and was frequently localized in plastid nucleoids, suggesting a likely role in interorganellar communication. Furthermore, ANN5-YFP co-immunoprecipitated with RABE1b, a putative GTPase, and interaction in planta was confirmed in plastidial nucleoids using FLIM-FRET analysis.

Conclusions: Our findings let us to propose that ANN5 influences basal cell homeostasis via modulation of plastid activity during pollen maturation. We hypothesize that the role of ANN5 is to orchestrate the plastidial and nuclear genome activities via protein-protein interactions however not only in maturing pollen but also during the transition from the vegetative to the generative growth and seed development.

Keywords: Accession; Annexin; Arabidopsis; Chlorophyll; Embryo; Nucleoid; Plastid; Pollen grain; Rab GTPase; Seed.

Fig. 1

ANN5 expression profiles. (a) Average expression of ANN5 in different organs of wild-type Arabidopsis Col-0. (b) Average expression of ANN5 in reproductive structures of wild-type Arabidopsis Col-0. (c) Average expression of ANN5 in floral buds and flowers at anthesis collected from different Arabidopsis genotypes. n = 3 biological replicates. Bars represent SD

Fig. 2

Impact of ANN5 expression on mature pollen grain size. (a) Scanning electron micrographs of the pollen grains from wild-type Arabidopsis (Col-0), ANN5-RNAi_15, ANN5-RNAi_13, and ANN5-OE_2. Scale bars = 10 μm. (b) Relative expression of ANN5 in mature pollen grains of wild-type Arabidopsis (Col-0), ANN5-RNAi_15, ANN5-RNAi_13, and ANN5-OE_2. Values above each column are expressed as percentage of the ANN5 expression in comparison to the wild-type (100%). (c) Mean length of mature pollen grains from wild-type Arabidopsis (Col-0), ANN5-RNAi_15, ANN5-RNAi_13, and ANN5-OE_2. n = 50. Asterisks indicate significant difference compared with values for wild-type pollen (one-way ANOVA, Dunnett post hoc test, *p < 0.05; **p < 0.01; ***p < 0.001). Bars represent SD. (d) and (e) Ultrastructure of viable and collapsing pollen grains from Arabidopsis genotypes with altered ANN5 expression. (d) Transmission electron micrographs showing ultrastructural details of viable mature pollen grains, whereas (e) depicts aborted pollen grains isolated during anthesis from wild-type Arabidopsis Col-0, ANN5-RNAi_15, ANN5-RNAi_13, and ANN5-OE_2. See also Additional file 1: Figures S3 and S4. Nu: nucleus, black arrow: plastid. Scale bars = 5 μm

Fig. 3

Pollen tube growth in pistils in ANN5 RNAi-silenced lines. Pollen tubes were fixed and stained with Aniline Blue 6 h after hand-pollination of (a) wild-type Col-0, (b) ANN5-RNAi_15, (c) ANN5-RNAi_13, and (d) ANN5-OE_2 plants. Aniline blue staining of pollen tubes was performed as described by [19]. Yellow arrows indicate pollen tube length measured from the top of style to the front of the longest pollen tube. (e) Average lengths of pollen tubes in pistils. n = 3 independent experiments. Asterisk indicates significant difference compared with the wild type (one-way ANOVA, Dunnett post hoc test, (^#p = 0.0501; *p < 0.05; **p < 0.01; ***p < 0.001). See also Additional file 1: Figure S5. Scale bars = 200 μm

Fig. 4

Impact of ANN5 expression on seed yield. (a) Dry seeds isolated from siliques at positions 36–40 of the main bolt and embryos dissected from rehydrated seeds of wild-type Arabidopsis (Col-0), ANN5-RNAi_15, ANN5-RNAi_13, and ANN5-OE_2. Scale bars = 500 μm. (b) Average sizes of pooled seeds from a single biological replicate. n = 1000. Three independent experiments were performed with similar outcomes. Asterisks indicate significant differences compared with wild-type seeds (one-way ANOVA, Dunnett post hoc test, *p < 0.05; **p < 0.01; ***p < 0.001). Bars represent SD. (c) Average sizes of seeds collected from siliques at specified positions on the main bolt, pooled from a single biological replicate. n = 120–150. Bars represent SD

Fig. 5

Subcellular localization of ANN5 in epidermal cells. Confocal optical sections of N. benthamiana leaf epidermal cells depicting localization of (a) C-terminus tagged ANN5 (35S:ANN5-GFP), (b) ANN5-GFP localization merged with chlorophyll autofluorescence, (c) N-terminus tagged ANN5 (35S:GFP-ANN5), and (d) GFP-ANN5 localization merged with chlorophyll autofluorescence. Scale bars = 10 μm. (e) Confocal optical section of two neighboring epidermal cells revealing different patterns of ANN5-GFP localization within plastids. White asterisks denote plastids containing ANN5-GFP. Scale bar = 10 μm. (f) Percentage of cells showing nucleo-cytoplasmic (N-C) or plastidial (Ch) localization of ANN5-GFP, and GFP-ANN5. The data were obtained in three independent experiments. Bars represent SD

Fig. 6

ANN5 interacts with RABE1b in plastidial nucleoids. Upper panel (a–b): Confocal optical section of N. benthamiana leaf epidermal plastid transiently expressing ANN5-YFP (a) and counterstained with DAPI (1 μg ml-1 for 15 min at room temperature) after fixation with 2% paraformaldehyde (24 h at 4 °C) (b). Pseudocolored fluorescence of (A) ANN5-YFP (yellow), (b) DAPI (magenta), (c) merged channels of ANN5-YFP and DAPI, and (d) overlaid with chlorophyll autofluorescence (blue). Lower panel (E–H): Confocal optical section of N. benthamiana leaf epidermal plastid transiently co-expressing ANN5-YFP (E) and RABE1b-CFP (f). Pseudocolored fluorescence of (e) ANN5-YFP (yellow), (f) RABE1b-CFP (magenta), (g) merged channels of ANN5-YFP and RABE1b-CFP, and (h) overlaid with chlorophyll autofluorescence (blue). See also Additional file 1: Figure S6. Scale bar = 10 μm. (j) FLIM-FRET analysis of interactions between ANN5 and RABE1b in plastidial nucleoids. Average CFP lifetime was measured in the donor leaf samples of N. benthamiana expressing only RABE1b-CFP and in the presence of acceptor in samples co-expressing RABE1b-CFP and ANN5-YFP. n = 7 individual epidermal cells. Measurements were performed on a single plastid per cell, ** indicates statistically significant differences (Student’s t-test, p < 0.05)