Singlet oxygen signatures are detected independent of light or chloroplasts in response to multiple stresses

Abstract

The production of singlet oxygen is typically associated with inefficient dissipation of photosynthetic energy or can arise from light reactions as a result of accumulation of chlorophyll precursors as observed in fluorescent (flu)-like mutants. Such photodynamic production of singlet oxygen is thought to be involved in stress signaling and programmed cell death. Here we show that transcriptomes of multiple stresses, whether from light or dark treatments, were correlated with the transcriptome of the flu mutant. A core gene set of 118 genes, common to singlet oxygen, biotic and abiotic stresses was defined and confirmed to be activated photodynamically by the photosensitizer Rose Bengal. In addition, induction of the core gene set by abiotic and biotic selected stresses was shown to occur in the dark and in nonphotosynthetic tissue. Furthermore, when subjected to various biotic and abiotic stresses in the dark, the singlet oxygen-specific probe Singlet Oxygen Sensor Green detected rapid production of singlet oxygen in the Arabidopsis (Arabidopsis thaliana) root. Subcellular localization of Singlet Oxygen Sensor Green fluorescence showed its accumulation in mitochondria, peroxisomes, and the nucleus, suggesting several compartments as the possible origins or targets for singlet oxygen. Collectively, the results show that singlet oxygen can be produced by multiple stress pathways and can emanate from compartments other than the chloroplast in a light-independent manner. The results imply that the role of singlet oxygen in plant stress regulation and response is more ubiquitous than previously thought.

Figure 1.

Correlation values of select abiotic and biotic stress transcriptomes generated by the ROSMETER platform. Shown is a color-coded heat map of correlation scores between transcriptomes of stresses (y ordinate) with transcriptome indices (abscissa) of the individual ROS-producing treatments compiled in the ROSMETER platform. The predicted ROS type, including ozone (O₃), superoxide (O₂^·−), H₂O₂, and singlet oxygen (¹O₂), and cellular location for each index are indicated at the top of the heat map (Rosenwasser et al., 2011, 2013). The source of transcriptome data used for the bioinformatics analysis and abbreviations are in Supplemental Table S10. The color scale for correlation is shown on the right. Blue, yellow, and white boxes are described in the text.

Figure 2.

Delineation of a CGS for stress and singlet oxygen-related transcripts. A, Venn diagram showing the number of overlapping genes between abiotic stress (wound local and systemic leaves after 10 min, 1 h drought, and 30 min of high light), biotic stress (1 h OGs, Flg22 W.t. 30 min, EF-Tu 30 min, chitin 30 min), and early flu responses 60 min after the dark-to-light transition. B, Venn diagram showing the number of overlapping genes between CGS, rapid wound-responsive (Walley et al., 2007), and common stress-responsive genes (Ma and Bohnert, 2007). RWR, Rapid wound responsive.

Figure 3.

Expression of transcripts after stress treatments. qRT-PCR of CGS transcripts from plants after various treatments. A, Plants were treated by spraying with 0.01% Silwet L-77 as the control, or 0.01% Silwet L-77 and 200 µm Rose Bengal. After 1 h in the darkness, plants were exposed to light intensity of 120 μE m⁻² s⁻¹ for 6 h. B, Plants were sprayed with 0.01% Silwet L-77 as a control or 0.01% Silwet L-77, 0.25, or 25 µm DCMU as indicated. They were then kept in the dark for 9 h. Plants were transferred to daylight with an average of 600 μE m⁻² s⁻¹ for 7.5 h. C, Dark-adapted plants were wounded and processed after 30 min as described in Materials and Methods. D, Plants were grown on vertical agar plates for 7 d and then transferred to liquid medium either supplemented with 1 µm flg22 or not (control). They were subsequently kept in darkness for 140 min before samples were collected. An ANOVA statistical test followed by a contrast Student’s t test and step-up correction for multiple testing was used to determine significance of the difference relative to the control. *P < 0.1; **P < 0.05; ***P < 0.01; ****P < 0.001. Average ± se values (n = 3–4) are presented. [See online article for color version of this figure.]

Figure 4.

Confocal imaging and analysis of SOSG fluorescence in root tissue. A, Root stained simultaneously with 50 µm Rose Bengal and 100 µm SOSG and visualized for Rose Bengal (left) or SOSG (right). Top, The root before 1-s activation by an argon laser at 559 nm (circled area). Bottom, The root after activation. B, Relative quantification of the fluorescent signal emitted by Rose Bengal (red squares) and SOSG (green diamonds) within the activation spot shown in (A). Arrow indicates laser activation. C, Differentiation zone of the root of 6-d-old seedlings expressing the mitochondrial marker Saccharomyces cerevisiae cytochrome c oxidase IV (ScCOX4):mCherry stained with DAPI and SOSG. The last column shows a merge of all channels. Seedlings were either incubated in 1× phosphate-buffered saline to serve as control (top row), or with flg22 for 80 min (bottom row). The arrow points to mitochondrion displaying SOSG fluorescence, whereas the arrowhead points to an area lacking ScCOX4:mCherry fluorescence but containing SOSG fluorescence. D, Differentiation zone of the root of 4-d-old seedling expressing the peroxisomal marker ARABIDOPSIS PEROXIN5 (AtPEX5)-CFP stained with DAPI and SOSG. The last column shows a merge of all channels. The AtPEX5-CFP was color coded red for better visualization of colocalization. The arrow points to peroxisomes displaying SOSG fluorescence, whereas the arrowhead points to an area lacking AtPEX5-CFP fluorescence but containing a SOSG signal. E, Percentage of peroxisomes and mitochondria with a SOSG fluorescent signal that is higher than the nearby surrounding area. a.u., Arbitrary unit; DAPI, 4′,6-Diamino-phenylindole; M, Mitochondria; P, Peroxisome. Bar = 10 μm.

Figure 5.

Generation of singlet oxygen in vivo by wound or by photodynamic activation of Rose Bengal. A, A schematic diagram of the experimental setup used to monitor wound stress (left). Five-day-old seedlings were sandwiched between slides that were separated so as not to evoke mechanical pressure (detailed in “Materials and Methods”). Cotyledons were wounded with a hemostat and imaged within the area shown. Seedlings that were loaded with 100 µm SOSG for 1 h followed by 50 µm Rose Bengal for 15 min, washed, and analyzed in an inverted fluorescence microscope (Olympus Ix71; right. A representative result is shown (n = 4). B, A schematic diagram of the sequential root cell layers measured by confocal microscopy (Olympus Flow View FV1000; left). Measurements started at the surface and reached 20 μm in depth; the area measured is shown with double arrows. Quantification of the SOSG fluorescent intensity in the root tip before and after wounding of cotyledons (right). Measurements were made along the different cell layers depicted on the left at different times. Seedlings were incubated with 100 µm SOSG for 10 to 20 min. The representative result is shown (n = 15). a.u., Arbitrary unit. [See online article for color version of this figure.]

Figure 6.

Analysis of singlet oxygen levels in vivo after drought, flg22, and rotenone treatments. Imaging was performed on the root tip of 5-d-old seedlings. Insets above bars display a representative result for each treatment. Top, Seedlings were incubated with SOSG for 20 min and then transferred to wet (control) or to dry Whatman paper (drought) for 15 min (n = 4). Middle, Seedlings were incubated with 10 µm flg22 for 80 min, and 100 µm SOSG was added 20 min before the end of the incubation time (n = 3). Bottom, Seedlings were incubated with 0.03% chloroform (control) or 0.03% chloroform and 40 µm rotenone for 100 min. SOSG (100 µm) was added 20 min before the end of the incubation time (n = 3 to 4). A two-sided Student’s t test was performed to evaluate the significance. **P < 0.05; ***P < 0.01. a.u., Arbitrary unit. [See online article for color version of this figure.]

Figure 7.

Detection of singlet oxygen during drying treatment by EPR spectroscopy. Arabidopsis leaf discs were incubated with TEMP for 30 min in the dark, rinsed, and left on water or subjected to dehydration treatments in the dark for the indicated times. Tissue was frozen and processed in the dark and measured using EPR spectroscopy as described in Materials and Methods.