The Histone Chaperone HIRA Is a Positive Regulator of Seed Germination

Abstract

Histone chaperones regulate the flow and dynamics of histone variants and ensure their assembly into nucleosomal structures, thereby contributing to the repertoire of histone variants in specialized cells or tissues. To date, not much is known on the distribution of histone variants and their modifications in the dry seed embryo. Here, we bring evidence that genes encoding the replacement histone variant H3.3 are expressed in Arabidopsis dry seeds and that embryo chromatin is characterized by a low H3.1/H3.3 ratio. Loss of HISTONE REGULATOR A (HIRA), a histone chaperone responsible for H3.3 deposition, reduces cellular H3 levels and increases chromatin accessibility in dry seeds. These molecular differences are accompanied by increased seed dormancy in hira-1 mutant seeds. The loss of HIRA negatively affects seed germination even in the absence of HISTONE MONOUBIQUITINATION 1 or TRANSCRIPTION ELONGATION FACTOR II S, known to be required for seed dormancy. Finally, hira-1 mutant seeds show lower germination efficiency when aged under controlled deterioration conditions or when facing unfavorable environmental conditions such as high salinity. Altogether, our results reveal a dependency of dry seed chromatin organization on the replication-independent histone deposition pathway and show that HIRA contributes to modulating seed dormancy and vigor.

Keywords: chromatin; histone chaperones; histone variants; seed dormancy and germination; seed vigor.

Figure 1

Chromatin in dry seeds is enriched in the replacement histone H3.3 variant. (a–f) Relative transcript levels of two genes encoding H3.1 (HTR1 (a), HTR9 (b)) and the middle subunit of the CAF-1 complex (FAS2 (c)), as well as two genes encoding H3.3 (HTR5 (d), HTR8 (e)) and the central subunit of the HIR complex (HIRA (f)) in flowers, siliques (7 to 9 and 12 to 13 days after pollination (DAP)), dry seeds and seeds imbibed for 2 hours (H), 8 h or 32 h as well as two-days-old seedlings. Transcript levels are normalized to MONENSIN SENSITIVITY 1 (MON1, At2g28390). Error bars correspond to the standard error of the mean (SEM) of three biological replicates and statistically significant differences relative to dry seeds were determined using a two-sided Student’s t-test (* p < 0.05; *** p < 0.001). (g–j) Maximum intensity projections of confocal images of embryos 1 h after imbibition (g,i) and two-days-old cotyledons (h,j) of transgenic lines expressing H3.1-GFP (g,h) or H3.3-GFP (i,j).

Figure 2

Loss of HIRA leads to reduced H3 content and MNase hypersensitivity. (a) Representative Western blot of H3 histones in total protein extracts from WT and hira-1 mutant dry seeds. (b) Quantification of H3 band intensities relative to ACTIN from three biological and four technical replicates. One wild type sample was set to 1 in each technical replicate. Statistically significant differences relative to WT were determined using a two-sided Student’s t-test (* p < 0.05). (c) Nuclei from WT or hira-1 mutant seeds were isolated and incubated with MNase for 1, 4, 16 and 64 min. Equal amounts of digested DNA were loaded on an agarose gel and stained with ethidium bromide. One experiment of three biological replicates with similar results is shown.

Figure 3

Seeds lacking HIRA show enhanced seed dormancy. Dormancy defects induced by HIRA loss-of-function (a) Germination of freshly harvested seeds of Col-0 (black) and hira-1 mutants (dark grey) in darkness at 25 °C after one, two and three days of stratification. Percentage of germinating seeds was scored six days after transfer to 25 °C. (b,c) Germination of freshly harvested seeds of Col-0 (black), hira-1 mutants (dark grey) and the complementing line (hira-1 pHIRA::HIRA-GFP; light grey) in darkness at 25 °C without (b) and after three days of stratification at 4 °C (c). Error bars correspond to SD from three biological replicates.

Figure 4

Enhanced seed dormancy of hira-1 mutants is not alleviated upon treatment with ABA, GA or ethylene. Germination of Col-0 (black, square) and hira-1 mutant (dark grey, circle) seeds in presence (filled line) or absence (dashed line) of ABA (1 µM) (a) at 10 °C in darkness or of GA (1 mM) (b) and ethylene (50 ppm) (c) at 25 °C in darkness. Means of biological triplicates with SEM are shown. Statistically significant differences relative to WT at D7 were determined using a two-sided Student’s t-test (* p < 0.05; ** p < 0.01; ns = not significant).

Figure 5

HIRA is epistatic to HUB1 and RDO2/TFIIS. (a,b) Germination of freshly harvested seeds of Col 0 (black), hub1-5 (red), rdo2-2 (dark blue), hira-1 (grey), hub1-5 hira-1 (orange) and rdo2-2 hira-1 (light blue) at 10 °C (seed viability test; a) or at 25 °C (dormancy test; b) in darkness. Means of biological triplicates with SEM are shown. Statistically significant differences relative to dry seeds were determined using a two-sided Student’s t-test (a = p < 0.01 relative to Col 0; b = p < 0.01 relative to hub1-5; c = p < 0.01 relative to rdo2-2).

Figure 6

Germination of hira-1 mutant seeds is affected by salt stress and is sensitive to storage in a high temperature and humidity. (a) Germination of Col 0 (black) and hira-1 (grey) seeds under long-day conditions at 23 °C in presence of 50 mM or 100 mM NaCl after two or three days of stratification at 4 °C in darkness. Means of two biological replicates with SEM are shown. (b) Germination of Col 0 (black) and hira-1 (grey) seeds under long-day condition at 23 °C after exposure of dry seeds for 4, 6, 7 or 10 days to 80% relative humidity at 37 °C. Seeds were stratified for three days at 4 °C in darkness and germinated seeds scored six days after transfer to light. Means of four biological replicates with SEM are shown. Statistically significant differences relative to wild type seeds were determined using a two-sided Student’s t-test (* p < 0.05, ** p < 0.01, *** p < 0.001).