Polarized subcellular activation of Rho proteins by specific ROPGEFs drives pollen germination in Arabidopsis thaliana

Abstract

During plant fertilization, excess male gametes compete for a limited number of female gametes. The dormant male gametophyte, encapsulated in the pollen grain, consists of two sperm cells enclosed in a vegetative cell. After reaching the stigma of a compatible flower, quick and efficient germination of the vegetative cell to a tip-growing pollen tube is crucial to ensure fertilization success. Rho of Plants (ROP) signaling and their activating ROP Guanine Nucleotide Exchange Factors (ROPGEFs) are essential for initiating polar growth processes in multiple cell types. However, which ROPGEFs activate pollen germination is unknown. We investigated the role of ROPGEFs in initiating pollen germination and the required cell polarity establishment. Of the five pollen-expressed ROPGEFs, we found that GEF8, GEF9, and GEF12 are required for pollen germination and male fertilization success, as gef8;gef9;gef12 triple mutants showed almost complete loss of pollen germination in vitro and had a reduced allele transmission rate. Live-cell imaging and spatiotemporal analysis of subcellular protein distribution showed that GEF8, GEF9, and GEF11, but not GEF12, displayed transient polar protein accumulations at the future site of pollen germination minutes before pollen germination, demonstrating specific roles for GEF8 and GEF9 during the initiation of pollen germination. Furthermore, this novel GEF accumulation appears in a biphasic temporal manner and can shift its location laterally. We showed that the C-terminal domain of GEF8 and GEF9 confers their protein accumulation and demonstrated that GEFs locally activate ROPs and alter Ca2+ levels, which is required for pollen tube germination. We demonstrated that not all GEFs act redundantly during pollen germination, and we described for the first time a polar domain with spatiotemporal flexibility, which is crucial for the de novo establishment of a polar growth domain within a cell and, thus, for pollen function and fertilization success.

Fig 1. GEF8, GEF9, and GEF11 specifically accumulate at the pollen germination site before germination initiation.

(A) Localization of mCit-GEF8, mCit-GEF9, mCit-GEF11, and mCit-GEF12 under their respective promoters during pollen germination. Timepoint 0 corresponds to the beginning of pollen tube emergence, arrowheads mark the site of pollen emergence, and asterisks mark mCit-GEF9 localization around the sperm cells. (B) Kymographs of time-lapse images corresponding to (A) along a line crossing the pollen through the germination site (left) or around the pollen grain (right). The dotted line indicates time point 0 of pollen tube emergence; the dotted line box in the right kymographs encloses the pollen germination site; arrowheads highlight protein accumulations at the pollen germination site. (C–E) Relative fluorescence intensity profiles at the pollen germination site of mCit-GEF8 (yellow, n = 22), mCit-GEF9 (yellow, n = 14), and mCit-GEF11 (yellow, n = 17), all in comparison to mCit-GEF12 (black, n = 17). Thin lines show individual measurements, and thick lines represent the average of all samples. (F, G) Quantification of measurements shown in C–E). (F) Maximum polarization value per pollen grain before germination. Letters above plots show significantly different groups according to a one-way ANOVA with Tukey’s test (p < 0.05). (G) Portion of pollen with a polarized signal (polarization frequency) before germination. The asterisk indicates a significant difference according to a Χ²-test (p < 0.05). (H, I) Colocalization (left) and intensity profiles along a line, as indicated in the merged image, across the pollen grain through the pollen germination site (right) of mCit-GEF8 with mSct-GEF9 (E) and mCit-GEF12 with mSct-GEF9 (F) expressed under their respective promoters. All scale bars represent 10 µm. For underlying data of all quantification see S1 Data.

Fig 2. GEF8, GEF9, and GEF12 are necessary for pollen germination and male fertility.

(A) Pollen germination efficiency of in vitro germinated pollen 4 h after imbibition on pollen germination media. Each point represents one replicate with more than 150 pollen grains. Groups of statistically significant differences according to a one-way ANOVA with Tukey’s test (p < 0.05) are indicated with letters. (B) Representative images of in vitro germinated pollen 4 h after imbibition on PGM. (C) Quantification of mutant allele frequency in F1 generation of reciprocal crosses with Col-0 as female and the indicated mutants as pollen donors. The heterozygous allele of each genotype is indicated in bold. Asterisks indicate a significant difference in the allele frequency from the expected 50% according to Χ²-test (p < 0.05). For underlying data of all quantification see S1 Data.

Fig 3. The C-terminus of GEF8 and GEF9 is necessary and sufficient for protein accumulation.

(A) AlphaFold2 protein structure prediction of full-length GEF8 and ROP1 (Rank 1, predicted separately), matched on the PRONE8-ROP4 double-dimer structure (PDB: 2NTY). Different angles are shown, and the terminal regions are indicated by different colors. (B) Schemes of GEF protein structure and investigated truncation constructs, with variable N and C terminal domains indicated as in (A). (C–E) Protein localization of different mutant constructs during pollen germination. Timepoint 0 corresponds to the beginning of pollen tube emergence, arrowheads mark the site of pollen emergence, and asterisks mark localization around the sperm cells. (C) Truncation constructs mCit-GEF8^ΔN, mCit-GEF8^ΔC, mCit-GEF9^ΔN, and mCit-GEF9^ΔC in Col-0. (D) Domain swap constructs of GEF12 with alternative C-terminal domain, mCit-GEF12^GEF8C in gef8-cΔ1 and mCit-GEF12^GEF9C in Col-0. (E) Phosphorylation site mutations of GEF8-S518 to a phospho-mimic (mCit-GEF8^S518D) and phospho-dead variant (mCit-GEF8^S518A) variant expressed in gef8-cΔ1. All scale bars represent 10 µm. (F, G) Quantification of measurements shown in C–E). (F) Maximum polarization value per pollen grain before germination. Letters above plots show significantly different groups according to a one-way ANOVA with Tukey’s test (p < 0.05) with mCit-GEF8 (dotted line, group a) as a reference. (G) Portion of pollen with a polarized signal (polarization frequency) before germination. The dotted line indicates mCit-GEF8 frequency. The asterisks indicate a significant difference to mCit-GEF8 or mCit-GEF9 according to a Χ²-test (p < 0.05). (H) Pollen germination efficiency of in vitro germinated pollen 4 h after imbibition on pollen germination media. Each point represents one replicate with more than 50 pollen grains. Groups of statistically significant differences according to a one-way ANOVA with Tukey’s test (p < 0.05) are indicated with letters. For underlying data of all quantification see S1 Data.

Fig 4. GEF8 is required for polar ROP activation and Ca²⁺ signaling.

(A) Localization of mCit-ROP1, mCit-ROP3, and mCit-ROP5 under their respective promoter during pollen germination. Timepoint 0 corresponds to the beginning of pollen tube emergence, and arrowheads mark the site of pollen emergence. (B) Kymographs of time-laps images corresponding to (A) along a line crossing the pollen through the germination site (left) or around the pollen grain (right). The dotted line indicates time point 0 of pollen tube emergence; the dotted box in the right kymographs encloses the pollen germination site; arrowheads highlight protein accumulations at the pollen germination site. (C) Localization of the ROP activity indicator GEF12p::CRIB4-mCit in Col-0 and gef8-cΔ1 backgrounds. (D) Kymographs of time-laps images corresponding to (C) along a line crossing the pollen through the germination site (left) or around the pollen grain (right). Arrowheads highlight protein accumulations at the pollen germination site. (E, F) Quantification of measurements in germinating pollen grains expressing GEF12p::CRIB4-mCit in Col-0 (n = 12) and gef8-cΔ1 (n = 12). (E) Portion of pollen with a polarized signal (polarization frequency) before germination. No significant difference was found according to a Χ²-test (p < 0.05). (F) Maximum polarization value per pollen grain before germination. p-values of a Student t test is shown. (G) LAT52p::RGeco1 Ca²⁺ biosensor in Col-0 and gef8-cΔ1 background. (H) Kymographs of time-laps images corresponding to (E) along a line crossing the pollen through the germination site. Arrowhead highlights signal increases at the pollen germination site during germination initiation, which are represented by images. (I) Quantification of the number of detected large Ca²⁺ elevations p-values of a Student t-test is shown. All scale bars represent 10 µm. For underlying data of all quantification see S1 Data.