Arabidopsis pollen tube integrity and sperm release are regulated by RALF-mediated signaling

Abstract

In flowering plants, fertilization requires complex cell-to-cell communication events between the pollen tube and the female reproductive tissues, which are controlled by extracellular signaling molecules interacting with receptors at the pollen tube surface. We found that two such receptors in Arabidopsis, BUPS1 and BUPS2, and their peptide ligands, RALF4 and RALF19, are pollen tube-expressed and are required to maintain pollen tube integrity. BUPS1 and BUPS2 interact with receptors ANXUR1 and ANXUR2 via their ectodomains, and both sets of receptors bind RALF4 and RALF19. These receptor-ligand interactions are in competition with the female-derived ligand RALF34, which induces pollen tube bursting at nanomolar concentrations. We propose that RALF34 replaces RALF4 and RALF19 at the interface of pollen tube-female gametophyte contact, thereby deregulating BUPS-ANXUR signaling and in turn leading to pollen tube rupture and sperm release.

Fig. 1. Pollen-expressed BUPS1 and BUPS2 control pollen tube integrity

(A and B) Expression pattern of BUPS1 and BUPS2, as shown by BUPS1p::GUS (A) and BUPS2p::GUS (B) in flowers. (C and D) Localization of BUPS1-GFP (C) and BUPS2-GFP (D) in pollen tubes. Graphs show fluorescence intensity (FI) 10 µm from the apex (denoted by white dashed line); the x-axis values represent the distance from the first dash at the left (in pixels) along the dashed line. Arrowheads indicate the fluorescence enrichment on the plasma membrane. (E) Mutant alleles of bups1 and bups2. Double backslash (\\) denotes localization of single-guide RNAs; single backslash (\) denotes the mutation site; dashed lines indicate large chromosomal deletion. SP, signal peptide; TM, transmembrane domain. (F and G) Silique length (F) and number of seeds per silique (G) of the wild type (WT) and various mutants. Data are means ± SD (n = 20 siliques); **P < 0.005, ***P < 0.001 (Student t test). (H) Transmission efficiency of bups mutants. ***P < 0.001 (χ² test); n.s., not significant. (I) Pollen germination assay of bups1 and bups1 bups2 pollen grains. Pictures were collected after 7 hours of incubation. Arrows indicate the position of rupture. (J) Aniline Blue staining of WT, bups1, and bups1 bups2 pollen tubes in wild-type pistils after 20 hours of pollination. Arrows show the areas where the bups pollen tubes made contact. Scale bars, 5 mm [(A), (B)], 10 µm [(C), (D)], 50 µm (I), 500 µm (J).

Fig. 2. RALF4 and RALF19 are pollen tube–expressed and, like BUPS1/2, are required for pollen tube integrity

(A and B) Expression pattern of RALF4p::GUS (A) and RALF19p::GUS (B) in flowers. (C and D) Localization of RALF4-GFP (C) and RALF19-GFP (D) in pollen tubes. (E) Mutant alleles of ralf4 and ralf19. Denotations are as described in Fig. 1E. (F and G) Silique length (G) and number of seeds per silique (F) of WT and various mutants. Data are means ± SD (n = 20 siliques); ***P < 0.001 (Student t test). (H) Seed number analysis in reciprocal crosses between WT and homozygous double mutants. ***P < 0.001 (Student t test). (I) Pollen germination assay of ralf4 ralf19 pollen grains. Pictures were collected after 7 hours incubation. (J) Aniline Blue staining of WT and ralf4 ralf19 pollen tubes in WT pistils at 20 hours after pollination. Arrow shows the area where the mutant pollen tubes made contact. Scale bars, 5 mm [(A), (B)], 10 µm [(C), (D)], 50 µm (I), 500 µm (J).

Fig. 3. BUPS1/2, ANX1/2, and RALF4/19 interact with each other

(A) Pull-down assay between Flag-tagged ectodomains of BUPS1/2 (purified from tobacco leaves) and biotinylated RALF4/19. (B) Pull-down assay between insect cell–expressed His-tagged ectodomains of BUPS1/2 and biotinylated RALF4/19. (C) Binding affinity between BUPS1/2 and RALF4/19 by MST analysis; K_d values are as indicated. The published RALF23-FER K_d value is 0.31 µM (24). ΔFNorm, change in fluorescence. (D) Dual-membrane yeast two-hybrid assays of BUPS1/2 and ANX1/2. (E) Interaction of BUPS1/2 and ANX1/2 through their ectodomains, as assessed by coimmunoprecipitation assay. The ectodomains of BUPS1/2 and ANX1/2 were coexpressed in tobacco leaf cells. (F and G) ANX1/2-Flag (F) and ANX1/2-His (G) interact with biotinylated RALF4/19 by pull-down assays. (H) MST analysis of the binding affinity between ANX1/2 ectodomains and RALF4/19. Error bars in (C) and (H) are SEM of two independent experiments.

Fig. 4. RALF34 triggers pollen tube burst and interacts with and competes for the ANX-BUPS receptor complex

(A to F) Time-course imaging of pollen tube discharge induced by RALF34 treatment; 20 µM peptide was used to test the effect on growing wild-type pollen tubes. The pollen tube discharge occurs less than 10 s after RALF34 treatment. (G) Quantitative analysis of pollen tube rupture treated with different concentrations of RALF4, RALF23, and RALF34. Data are means ± SD (n = 230 pollen tubes of each treatment for two independent experiments). (H) RALF34p::GUS expressed predominantly in mature ovules. (I) RALF34-GFP protein expressed in ovules. (J to M) Pull-down assays between biotinylated RALF8/34 and Flag- or His-tagged ectodomains of BUPS1/2 [(J) and (K)] or between biotinylated RALF8/34 and Flag- or His-tagged ectodomains of ANX1/2 [(L) and (M)]. (N and O) RALF34 peptide outcompetes biotinylated RALF4/19 interaction with Flag-tagged ectodomain of BUPS1 or ANX1. Scale bars, 50 µm [(A) to (F), (H)], 20 µm (I).