Organelles contribute differentially to reactive oxygen species-related events during extended darkness

Abstract

Treatment of Arabidopsis (Arabidopsis thaliana) leaves by extended darkness generates a genetically activated senescence program that culminates in cell death. The transcriptome of leaves subjected to extended darkness was found to contain a variety of reactive oxygen species (ROS)-specific signatures. The levels of transcripts constituting the transcriptome footprints of chloroplasts and cytoplasm ROS stresses decreased in leaves, as early as the second day of darkness. In contrast, an increase was detected in transcripts associated with mitochondrial and peroxisomal ROS stresses. The sequential changes in the redox state of the organelles during darkness were examined by redox-sensitive green fluorescent protein probes (roGFP) that were targeted to specific organelles. In plastids, roGFP showed a decreased level of oxidation as early as the first day of darkness, followed by a gradual increase to starting levels. However, in mitochondria, the level of oxidation of roGFP rapidly increased as early as the first day of darkness, followed by an increase in the peroxisomal level of oxidation of roGFP on the second day. No changes in the probe oxidation were observed in the cytoplasm until the third day. The increase in mitochondrial roGFP degree of oxidation was abolished by sucrose treatment, implying that oxidation is caused by energy deprivation. The dynamic redox state visualized by roGFP probes and the analysis of microarray results are consistent with a scenario in which ROS stresses emanating from the mitochondria and peroxisomes occur early during darkness at a presymptomatic stage and jointly contribute to the senescence program.

Figure 1.

Cell death and chlorophyll content during extended dark treatment. Detached rosettes of 5-week-old Arabidopsis plants were kept in darkness, and the seventh leaf was examined following the transition to darkness. Chlorophyll content and cell death were determined in discs of this leaf as described in “Materials and Methods.” Leaves at 0 and 5 d of darkness are shown. O.D., Optical density. [See online article for color version of this figure.]

Figure 2.

Temporal changes of global gene expression following the transition to darkness. Hierarchical clustering of up-regulated or down-regulated genes during darkness is shown. Gene expression levels were normalized to the level of nondarkened leaves and are displayed as the log₂ ratio. Only genes that showed greater than 2-fold changes, in at least one time point, were included in the analysis.

Figure 3.

Histogram of the distribution of ROS-related gene expression signatures under extended dark treatment. The transcripts with 2-fold change were divided into five groups of ROS-specific transcriptome signatures that were defined by Gadjev et al. (2006) and appear on the x axis. Transcript levels for the second, third, and fifth days of these genes were normalized to the nondarkened leaves. The proportion of up-regulated or down-regulated transcripts for each of the groups was determined and calculated as the percentage of total genes modified in each of the categories. Chl, Chloroplast; Cyt, cytoplasm; Mit, mitochondria; Per, peroxisomes; Pl, plastid. The definitions of the five groups were based on data from the flu mutant (singlet oxygen in plastid), knockdown of SOD (superoxide in chloroplast), MV (superoxide in chloroplast or mitochondria), knockout of APX1 (H₂O₂ in cytoplasm), and CAT-deficient plants (H₂O₂ in peroxisomes). Numbers below each bar represent the total number of genes that were modified at the indicated times.

Figure 4.

ROS-related transcriptome signatures under extended darkness. The correlation values between gene expression profiles of extended darkness and ROS-related experiments were calculated as described in “Materials and Methods” and are presented as a heat map. The table on the left describes the localization of the possible ROS forms in the 10 different ROS-related experiments, and the color-coded results for each vector on the various days of the dark treatment are shown on the right. Localization abbreviations are as in Figure 3. Correlation values are between 1 (complete positive correlation; red) and −1 (highest possible negative correlation; green). The actual correlation values are tabulated in Supplemental Table S4. Data for the second day (d2) were normalized to those of nondarkened tissue, and data for the third (d3) and fifth (d5) days were normalized to either that of nondarkened tissue or tissue subjected to darkness for 2 d (d2) or 3 d (d3), respectively. The data sets were obtained from KO-APX1 (exposed to high light [HL]; Davletova et al., 2005), flu mutants (op den Camp et al., 2003), KD-SOD (Rizhsky et al., 2003), fnr1 (see “Materials and Methods”), treatment with MV (Kilian et al., 2007), rotenone treatment (an inhibitor of complex 1 in mitochondrial electron transport; Garmier et al., 2008), AOX-MLD (Giraud et al., 2008), treatment with 3-aminotriazole (AT), a potent inhibitor of CAT (Gechev et al., 2005), and CAT2HP1 (Vanderauwera et al., 2005). Time indicates the hours following each treatment.

Figure 5.

Changes in the degree of oxidation of roGFP in various cellular compartments during extended dark treatment. A, Relative oxidation state in transgenic Arabidopsis lines expressing roGFP2 in plastids and mitochondria and roGFP1 in mitochondria. B, Relative oxidation state in transgenic Arabidopsis lines expressing GRX1-roGFP2 in the peroxisomes and cytosol. Fluorescence related to roGFP was detected at 515 nm following excitation at 400 and 485 nm using a fluorometer as described in “Materials and Methods.”

Figure 6.

The effect of Suc treatment on roGFP2 oxidation in mitochondria. Rosettes of Arabidopsis plants that express roGFP2 in mitochondria were floated on either 60 mm Suc or 60 mm sorbitol and kept in the dark for 1 d. Control at 0 d is of nondarkened leaves.