The effects of H3.3 chaperone ASF1A/1B, HIRA and ATRX on Arabidopsis microgamete development. (A) Mature siliques collected prior to dehiscence and cleared in 70% ethanol for several days to reveal seeds. Scale bar: 5 mm. (B) Seed setting rate per silique. n ≥ 16 siliques from ≥3 plants for Col-0, atrx-1, hira-1 and asf1a1b mutants. Statistical significance relative to Col-0 was determined by two-tailed Student’s t test (****P < 0.01). (C) Morphology of representative flowers, stamens and pistils of Col-0, atrx-1, hira-1 and asf1a1b mutants. Scale bar, 1 mm. (D) Morphology of representative anthers of Col-0, atrx-1, hira-1 and asf1a1b mutants. Scale bar, 200 μm. (E) The transcription levels of several microgametogenesis-associated genes in Col-0 and atrx-1, hira-1 and asf1a1b mutants. (F) Global levels of HTR4-GFP in Col-0, pHTR4-HTR4-GFP, pHTR4-HTR4-GFP hira-1 and pHTR4-HTR4-GFP asf1a1b plants. Total protein extracts from young inflorescences were analyzed by protein immunoblots using specific antibodies that recognize GFP or actin as indicated. (G) ChIP analysis of HTR4-GFP deposition at specific genes in young inflorescences of Col-0, pHTR4-HTR4-GFP, pHTR4-HTR4-GFP hira-1 and pHTR4-HTR4-GFP asf1a1b plants. ChIP samples were analyzed by quantitative PCRs on the region represented in Fig. 5C of each gene. Statistical significance of the HTR4-GFP level in pHTR4-HTR4-GFP was determined by two-tailed Student’s t test (*P < 0.05, **P < 0.01, ***P < 0.001).