The Role of Persulfide Metabolism During Arabidopsis Seed Development Under Light and Dark Conditions

Abstract

The sulfur dioxygenase ETHE1 oxidizes persulfides in the mitochondrial matrix and is involved in the degradation of L-cysteine and hydrogen sulfide. ETHE1 has an essential but as yet undefined function in early embryo development of Arabidopsis thaliana. In leaves, ETHE1 is strongly induced by extended darkness and participates in the use of amino acids as alternative respiratory substrates during carbohydrate starvation. Thus, we tested the effect of darkness on seed development in an ETHE1 deficient mutant in comparison to the wild type. Since ETHE1 knock-out is embryo lethal, the knock-down line ethe1-1 with about 1% residual sulfur dioxygenase activity was used for this study. We performed phenotypic analysis, metabolite profiling and comparative proteomics in order to investigate the general effect of extended darkness on seed metabolism and further define the specific function of the mitochondrial sulfur dioxygenase ETHE1 in seeds. Shading of the siliques had no morphological effect on embryogenesis in wild type plants. However, the developmental delay that was already visible in ethe1-1 seeds under control conditions was further enhanced in the darkness. Dark conditions strongly affected seed quality parameters of both wild type and mutant plants. The effect of ETHE1 knock-down on amino acid profiles was clearly different from that found in leaves indicating that in seeds persulfide oxidation interacts with alanine and glycine rather than branched-chain amino acid metabolism. Sulfur dioxygenase deficiency led to defects in endosperm development possibly due to alterations in the cellularization process. In addition, we provide evidence for a potential role of persulfide metabolism in abscisic acid (ABA) signal transduction in seeds. We conclude that the knock-down of ETHE1 causes metabolic re-arrangements in seeds that differ from those in leaves. Putative mechanisms that cause the aberrant endosperm and embryo development are discussed.

Keywords: amino acids; cellularization; cysteine degradation; embryogenesis; mitochondria; persulfide signaling.

FIGURE 1

Seed development of Arabidopsis thaliana wild type and ethe1-1 seeds under light and dark conditions. (A) Representative seeds and embryos of wild type (wt) and ethe1-1 from 1 to 9 days after pollination (DAP) grown under light and dark conditions, bars = 100 μm. (B) Progression of embryo development in wt and ethe1-1 seeds. A total of 180 siliques and 7781 seeds between 1 and 9 DAP were analyzed of wild type and ethe1-1 grown under light and dark conditions. Blue arrows indicates stages investigated by metabolomics and proteomics.

FIGURE 2

Seed weight, seed size and biomass composition of mature seeds. (A) Seed size and weight of 100 mature seeds from wild type and ethe1-1 plants grown under light and dark conditions. Asterisks indicate significant changes compared to wtL based on a Student’s t-test (p-value ≤ 0.05). (B) Storage compounds and amino acids were extracted from mature seeds from wild type and ethe1-1 grown under light and dark conditions and quantified. wtL, wild type seeds grown under light conditions; wtD, wild type seeds grown under dark conditions; mL, ethe1-1 seeds grown under light conditions; mD, ethe1-1 seeds grown under dark conditions.

FIGURE 3

Amino acid profiles. Relative amino acid contents of wild type and ethe1-1 seeds grown under light and dark conditions at 5 and 9 days after pollination (DAP) and in a mature state. Datasets were normalized to wtL samples of the same age, respectively. wtL, wild type seeds grown under light conditions; wtD, wild type seeds grown under dark conditions; mL, ethe1-1 seeds grown under light conditions; mD, ethe1-1 seeds grown under dark conditions.

FIGURE 4

Principle component analysis and overview of proteomic comparisons. (A) For quality control of the shotgun MS datasets principle component analysis (PCA) was performed using Perseus software. (B) The numbers of significantly increased and decreased proteins for relevant comparisons are indicated (Student’s t-test, p-value < 0.05). Comparisons were wtD/wtL, mL/wtL, mD/mL. (C) Venn diagram illustrating the numbers of proteins detected in the individual sample goups. The total number of detected proteins is listed in the bottom. wtL, wild type seeds grown under light conditions; wtD, wild type seeds grown under dark conditions; mL, ethe1-1 seeds grown under light conditions; mD, ethe1-1 seeds grown under dark conditions, numbers indicate the age of the seed in days after pollination.

FIGURE 5

The effect of darkness and ETHE1 knock-down on the proteome of developing Arabidopsis seeds at 9 DAP. (A) Functional categorization of proteins that were detected in dark but not in light grown seeds according to the MapMan annotation file (Thimm et al., 2004, version Ath_AGI_LOCUS_TAIR10_Aug2012). (B) Log₂-fold changes in protein abundances were calculated for three different comparisons (from left to right): dark vs. light grown wild type seeds, dark vs. light grown ethe1-1 seeds, and ethe1-1 vs. wild type seeds grown under control conditions. The color gradient represents the means of all log₂-ratios (scaling: –1 to +1) for proteins included in the functional categories listed according to the MapMan annotation file (version Ath_AGI_LOCUS_TAIR10_Aug2012). Wt, wild type; m, mutant; PS, photosynthesis; TAG, triacylglyceride; TCA, tricarboxylic acid.

FIGURE 6

ETHE1 deficiency affects endosperm cellularization. (A) Vulcano plot illustrating differences in protein abundance between ethe1-1 and wt seeds at 9 DAP. Solid lines represent the threshold for significance (FDR: 0.05, s0: 0.1). Proteins involved in cell wall synthesis or degradation are highlighted in red, constituents of the cytoskeleton are highlighted in blue, and proteins involved in vesicle traffic are highlighted in purple. Filled circles represent proteins that are consistenly regulated in light and dark grown ethe1-1 compared to wt seeds. (B) Seed sections (7 DAP) of wild type and ethe1-1 seeds grown under light and dark conditions. Sections were stained with Toluidine Blue O to visualize endospserm cellularization, bars = 100 μm. (C) Endosperm cell size of ca. 50 cells is given as area [μm²] obtained from microscopy images using AxioVision software (Version 4.8.1). Asterisks indicate significant changes to wild type light (wtL) based on a Student’s t-test (p-value ≤ 0.05). (D) Electron microsopic analysis of mitochondrial structure from ethe1-1 seeds compared to wild type, bars = 500 nm; wtL, wild type seeds grown under light conditions; wtD, wild type seeds grown under dark conditions; mL, ethe1-1 seeds grown under light conditions; mD, ethe1-1 seeds grown under dark conditions.

FIGURE 7

ETHE1 deficiency affects abscisic acid (ABA) signal transduction. (A) Vulcano plot illustrating differences in protein abundance between light grown ethe1-1 and wt seeds at 9 DAP. Solid lines represent the threshold for significance (FDR: 0.05, s0: 0.1). Proteins induced by ABA or jasmonic acid (JA) are marked in red and blue, respectively (Goda et al., 2008). Filled circles represent proteins that are consistenly regulated in light and dark grown ethe1-1 compared to wt seeds. (B) Phenotype of wild type and ethe1-1 seedlings grown on agar plates containing 0 μM (control) or 1 μM ABA 10 days after transfer to light conditions. Arrows indicate seeds that had germinated but arrested growth after emergence of the radicle. scale bars = 1 cm. (C) Relative growth ratio of wt compared to ethe1-1 seedlings on agar plates containing 1 μM ABA calculated as the percentage of germinated plants that had also developed leaves. All data points were normalized to the growth ratio of ethe1-1 seedlings at 9 days after imbibition. Asterisks indicate significant differences to the wild type based on a Student’s t-test (p-value ≤ 0.05).