Fig 1.
Pollen tube growth is impaired in sss1-D.
(A) Comparison of mature panicle of wild type (WT) and sss1-D. Arrows indicate the sterile spikelets. (B) Seed-setting rate. Data are means ± SD (n > 5). **P<0.01 by the Student’s t test. (C) Floral structure of WT and sss1-D. (D and E) I2-KI staining of pollen grains of WT and sss1-D. (F and G) Aniline blue staining of the WT and sss1-D pollen germination in vitro. (H-K) Aniline blue staining showing normal germination of pollen grains on stigmas in selfed WT (H and I) and sss1-D (J and K) at 5 minutes after pollination (MAP). (L-O) Retardation of pollen tube growth in sss1-D. WT pollen tubes reach the bottom part of the style in most WT pistils at 30 MAP (L and M), whereas in some sss1-D pistils, pollen tubes stay at the stigma–style boundary (N and O). (P-V) Multiple pollen tubes can be observed in the WT ovule at 120 MAP (P), whereas the sss1-D ovule contains fewer or no pollen tubes (Q and R). Pollen tubes stay at the stigma–style boundary (S) or at the middle part of the style in majority of sss1-D pistils (T). (U) Quantification of ovaries with pollen tubes observed in the ovule at 120 MAP, showing defective pollen tube growth in more than half of sss1-D pistils. (V) Verification of the maternal effect of sss1-D for defective pollen tube growth. The sss1-D pollen tubes are capable of growing to ovule at 120 MAP in WT pistils, whereas the WT pollen tubes fail to reach the ovule in more than 50% of sss1-D pistils. Arrows point to pollen tubes in I and K, and pollen tube tips in L-O and S-T. Data in U and V are means ± SD from 3 replicates with > 30 pistils observed per replicate, and different letters indicate a significant difference at P < 0.01 by the Student’s t-test. Scale bars, 5 cm in (A); 2 cm in (C); 100 μm in (D-G,M,O,S,T); 300 μm in (H,J,L,N,P-R); 50 μm in (I and K).
Fig 2.
(A) The sss1-D locus was mapped to a ~52 kb region on the short arm of chromosome 6 between the markers Q-20 and Y-80. (B) Annotated open reading frames (ORF) in the ~52 kb region. Note that the inversion of a 44 kb genomic fragment in sss1-D disrupts both ORF1 and ORF3. (C) Verification of the inversion in sss1-D. Primers used for PCR analysis are shown in B. (D) A diagram of ORF1 and ORF3 in wild type and their mutant version in sss1-D. The inversion results in a truncated ORF1 with a premature stop (mORF1) and a stop codon-less ORF3 (mORF3). Filled box, exon; bold line, intron. (E) qRT-PCR analysis of the transcript levels of ORF1/mORF1 (primer pair 5F/5R in D), ORF2, and ORF3. Data are means ± SD (n = 3).
Fig 3.
Functional complementation and gene knockout analyses.
(A) Genetic scheme of W109 selection. (B-D) The seed-setting rate (B) and qRT-PCR analysis of ORF1 (primer pair 6F/6R shown in Fig 2D) (C) and ORF3 transcript levels (D) in W109 and the transgenic plants. Three independent transgenic lines individually transformed with pORF1::ORF1, 35S::ORF1-GFP, gORF3, and 35S::ORF3-GFP are shown. (E) Observation of pollen tube growth in W109, 35S::ORF1-GFP-1 transgenic plant, and gORF3-1 transgenic plant. Arrow indicates the pollen tube tip. Scale bars, 300 μm. (F) Frequency of the pistils with at least one pollen tube reaching the ovaries at 120 MAP. (G and H) Sketch map of the mutations of ORF1 (G) and ORF3 (H) in knockout lines. The mutation site, the corresponding seed-setting rate, and the percentage of ovules with pollen tube of each line are shown. Minus (-) and plus (+) signs indicate the number of nucleotides deleted and inserted, respectively. Data are means ± SD (n > 5 in B; n = 3 in G and H); data in F are means ± SD from 3 replicates with > 30 pistils observed per replicate. ns, no significance; **P<0.01 by the Student’s t test.
Fig 4.
The expression of OsCNGC13 (ORF1) positively affects the plant seed-setting rate.
(A) qRT-PCR analysis of OsCNGC13 and OsCNGC13-D in WT, sss1-D, and heterozygous plants with the 5F/5R primer pair shown in Fig 2D. Note that the expression of OsCNGC13 and OsCNGC13-D in the heterozygous plants was significantly reduced. (B and C) qRT-PCR analysis of OsCNGC13 and OsCNGC13-D in WT, sss1-D and heterozygous plants with allele-specific primers. Unequal reduction of expression of OsCNGC13 and OsCNGC13-D is observed in the heterozygous plant as normalized to WT (B) and sss1-D (C). (D) Comparison of mature panicles of wild type Kitaake and two independent lines (-1 and -2) of pOsCNGC13::OsCNGC13-D. Scale bars, 5 cm. (E and F) The seed-setting rate (E) and the percentages of ovules with pollen tube (F) in Kitaake and the transgenic plants. Data in E are means ± SD (n > 5). Data in F are means ± SD from 3 replicates with > 30 pistils observed per replicate. Different letters indicate a significant difference at P < 0.01 by the Student’s t-test. (G) qRT-PCR analysis of OsCNGC13 and OsCNGC13-D in Kitaake and the transgenic plants. The primer pairs used in A (5F/5R for both OsCNGC13 and OsCNGC13-D), B (6F/6R specific for OsCNGC13), C (7F/7R specific for OsCNGC13-D), G (6F/6R and 7F/7R) are shown in Fig 2D. Data are means ± SD (n = 3). (H) Positive correlation between the seed-setting rates and expression levels of OsCNGC13 in several RNAi transgenic lines. (I) Positive correlation between the seed-setting rates and the percentages of ovules with pollen tube in several RNAi transgenic lines. r value is based on two-tailed Pearson correlation analyses.
Fig 5.
OsCNGC13 mRNA expression and subcellular localization of its protein.
(A and B) GUS staining of the spikelets after removing the lemma and palea. Mature pistil before flowering (A) and pistil at 30 MAP (B) are shown. (C-L) In situ hybridization analysis. Longitudinal sections of the floral primodia (C and D), stigma (E) and style (F) before flowering, and the style of the pistil at 30 MAP (I) are shown. (D) The magnified image of the selected area in C. Transverse sections of the stigmas (G and J) and styles (H, K and L) of the pistils before flowering (G and H) and the pistils at 30 MAP (J-L) are shown. (L) Negative control with sense probe. (M-P) Localization of GFP protein (M), OsCNGC13-GFP fusion protein (N), and PIP2;1-mCherry fusion protein (O) in rice protoplasts. (P) The merged image of N and O. (Q-T) Localization of GFP protein in the root tip cells of transgenic rice plants expressing 35S promoter-driven GFP (Q) and localization of OsCNGC13-GFP fusion protein in the root cells of the transgenic rice plants expressing 35S::OsCNGC13-GFP (R). (S) Cellular outlines of the root cells were stained with FM4-64 for 5 min on ice. (T) The merged image of R and S. Scale bars, 1 mm in (A and B); 200 μm in (C and D); 50 μm in (E-L); 10 μm in (M-P); 20 μm in (Q-T).
Fig 6.
OsCNGC13 exhibits permeability to inward Ca2+ in HEK293 cells and is required for [Ca2+]cyt accumulation in the style.
(A) Patch-clamp whole-cell recordings of inward Ca2+ currents in HEK293 cells transfected with GFP (as a control), OsCNGC13, or OsCNGC13-D. The voltage protocols, as well as time and current scale bars for the recordings are shown. (B) The I-V relationship of the steady-state whole-cell inward Ca2+ currents. The data are derived from the recordings shown in A, and presented as means ± SD. (C) Style Ca2+ content measurement using SEM-EDX. Increased Ca2+ concentrations are detected in WT but not in sss1-D after pollination. (D-G) Fluo-3/AM showing Ca2+ accumulation in the styles. Images of styles before Fluo-3/AM incubation (D), after incubation in the buffer without Fluo-3/AM (E), and after Fluo-3/AM incubation (F) are shown. Note that the samples before incubation and the samples incubated in the buffer without Fluo-3/AM were performed as the negative controls, both of which showed no fluorescence. (G) Quantification of Fluo-3/AM fluorescence intensity. (H and I) Ca2+ accumulation in the styles indicated by the YC3.6 protein fluorescence. (I) Quantification of YC3.6 fluorescence intensity. Data are means ± SD (n = 4 in C; n = 9 in G; n = 6 in I). *P<0.05 and **P<0.01 by the Student’s t test. Scale bars, 100 μm.
Fig 7.
Defective PCD in styles of sss1-D during pollen tube growth and a proposed model of OsCNGC13 function.
(A-H) Transverse paraffin (A-D) and plastic (E-H) sections at the middle of the style. The boxed regions in A, B, E, and F are magnified in C, D, G, and H, respectively. Intercellular space (red arrows), as a result of PCD, is observed in WT but not in sss1-D. (I and J) TEM images of cells at the bottom of the style, showing a collapsed cell in WT (I) but not in sss1-D (J). Nu, nucleus; Mt, mitochondria. (K-P) TUNEL assay shows that DNA fragmentation signal (white arrow) is visible in WT but not in sss1-D. Longitudinal sections of the bottom (K and L), transverse sections at the middle (M and N) and bottom parts (O and P) of the style are shown. (Q) qRT-PCR analysis of the transcript levels of the genes related to PCD. Pollination-triggered expression of these genes is significantly reduced in sss1-D. All samples used in A-P were collected at 30 MAP. Data in Q are the mean ± SD (n = 3). **P < 0.01 by the Student’s t-test. Scale bars, 50 μm in (A, B, E, F and K-P); 1 μm in (I and J). (R) A sketch illustrating the patterns of pollen tube growth in the pistils of the wild type and the sss1-D mutant. The growth of pollen tube is normal in almost all the wild type pistils, while the pollen tube is blocked in about half of the sss1-D pistils. (S) A proposed model of OsCNGC13 function. OsCNGC13 is localized on the plasma membrane of the STT cells and plays an important role in linking [Ca2+]cyt accumulation in the style after pollination, ECM components modification and PCD in the style to facilitate the penetration of the pollen tube, successful double fertilization, and consequently high seed-setting rate.