Fig. 3: PXL1 is required for normal pollen exine formation.

a–e Phenotypic analyses of anther and pollen grains of the WT, DN-CLE19, DM#8, DM#25, DE#12, DE#20, DM#8 pxl1+/−, DE#12 pxl1-1−/− transgenic plants. Alexander-stained anthers and SEM images of pollen are shown. Bar = 100 μm for the anthers and 10 μm for the SEM images. f Quantification of pollen amounts per anther in the indicated genotypes. Three independent biological replicates were performed, and ten anthers were used for analysis for each replicate. The data are shown as the mean ± SD, each dot shows the average value for one biological replicate. Different letters represent significant difference between each other, p < 0.05, one-way ANOVA with Tukey multiple comparison test. Exact p values are 1.66e⁻⁶ for DN-CLE19 vs. WT, 5.13e⁻⁷ for DM#8 vs. WT, 3.94e⁻⁷ for DM#25 vs. WT, 1.15e⁻⁵ for DE#12 vs. WT, 1.26e⁻⁶ for DE#20 vs. WT, 9.53e⁻⁷ for DM#8 pxl1-1+/− vs. WT, 1.44e⁻⁷ for DE#12 pxl1-1−/− vs. WT. g Proportion of pollen with normal, moderate-D and severe-D exine defects of various genotypes. The statistical tests were performed between normal and the sum of moderate-D and severe-D. Letters were assayed based on calculation by chi-square. N indicates the number of anthers counted for each genotype. Exact p values are 6.91e⁻⁶¹ for DN-CLE19 vs. WT, 1.10e⁻²⁹ for DM#8 vs. WT, 2.79e⁻⁵⁸ for DM#25 vs. WT, 1.47e⁻⁴⁰ for DE#12 vs. WT, 2.49e⁻²⁷ for DE#20 vs. WT, 1.35e⁻³² for DM#8 pxl1-1+/− vs. WT, 3.83e⁻⁸¹ for DE#12 pxl1-1−/− vs. WT. The relative expression of DYT1, AMS, MYB103, MS1 (h) and of CYP98A8, CYP86C3, AT5G55320, UGT72E2, PAL4, AT1G76470 and ACOS5 (i) in WT, DM#25 and DE#12, as evaluated by qRT–PCR. ACTIN was used as the internal control. Three biological replicates were performed. Each dot shows the result for one biological replicate. Data are shown as the mean ± SD. p values were calculated by Student’s t test, two sided for (h, i).