Chloroplast ATP-dependent metalloprotease FtsH5/VAR1 confers cold-stress tolerance through singlet oxygen and salicylic acid signaling

Abstract

Cold stress substantially affects plant growth and productivity. Chloroplasts are primary sites for the production of reactive oxygen species (ROS) and biosynthesis of the defense hormone salicylic acid (SA) under environmental stress conditions. However, the mechanisms by which plants integrate ROS and SA signaling to adapt to stress remain elusive. Here, we report that Arabidopsis FILAMENTOUS TEMPERATURE-SENSITIVE H5/YELLOW VARIEGATED1 (FtsH5/VAR1), a thylakoid-localized ATP-dependent zinc metalloprotease, is essential for plant cold-stress tolerance. The var1-1 mutant exhibits pronounced chlorosis and variegation, as well as retarded growth under cold stress conditions. We observed a strong correlation between elevated SA biosynthesis/signaling and the cold-sensitive phenotype of var1. Reducing SA accumulation, either by overexpressing the salicylate hydroxylase gene (NahG) or knocking out SA biosynthesis-related genes (ICS1, EDS1, or PAD4), partially suppressed the chlorosis phenotype of var1. Furthermore, we demonstrated that EXECUTOR1 (EX1)-mediated singlet oxygen (¹O₂) signaling acts upstream of EDS1 to regulate the expression of SA-responsive genes (SARGs) in var1 under cold stress. Notably, we identified a critical role for EX2, in which the mutation in EX2 significantly suppressed the cold-sensitive phenotype of var1, in activating the expression of SARGs while repressing photosynthesis-associated nuclear genes. Collectively, our results suggest a vital role for VAR1 in plant cold tolerance and highlight the tight connection between ¹O₂ and SA signaling, elucidating a previously unheeded function of EX2, which likely operates independently of EX1-mediated ¹O₂ signaling.

Keywords: EX2; VAR1; cold stress; salicylic acid; singlet oxygen.

Figure 1

Disruption of VAR1 leads to reduced cold-stress tolerance in Arabidopsis. (A) The var1-1 mutant is hypersensitive to cold stress. Wild type (WT), var1-1 and var2-15 mutants, and the complementation lines of var1-1 (var1-1 com) and var2-15 (var2-15 com) were germinated and grown under control conditions (22°C) for 11 days or cold-stress conditions (12°C /8°C day/night) for 30 days. To reach a comparable developmental stage, var1-1 plants were grown under cold-stress conditions for 50 days. Scale bar, 1 cm. (B and C) Chlorophyll a fluorescence images (B) showing the Fv/Fm values of different genotypes and corresponding quantification from five biological replicates (C). For var1-1 and var2-15 seedlings grown at low temperature, cotyledons (indicated by arrowheads) were used for measurements because other leaves exhibited chlorosis and yielded no measurable Fv/Fm values. Data are shown as means ± SD. (D) Chlorophyll content measurements of seedlings shown in (A). Data are means ± SD (n = 3 biological replicates). In (C) and (D), statistical significance was determined by one-way ANOVA (p < 0.05) with post hoc Tukey’s honestly significant difference (HSD) test.

Figure 2

Transcriptomic and proteomic analyses demonstrate a critical role of VAR1 under cold stress. (A and B) Volcano plots illustrating differentially expressed genes (DEGs) (A) and differentially accumulated proteins (B) in var1-1 or var2-15 compared to WT under control and cold-stress conditions. PS, photosynthesis; Ribo, ribosomal proteins; ns, nonsignificant; FC, fold change. (C) Histogram showing subcellular localization of protein products of DEGs between cold-stressed WT and var1-1 identified by transcriptomic analyses and differentially accumulated proteins identified by proteomic analyses. Numbers beside each column indicate the total number of DEGs or differentially accumulated proteins identified, with numbers of plastid-localized proteins shown in parentheses. DAP, differentially accumulated protein; DEG, differentially expressed gene; Pt, plastid; Vac, vacuole; PM, plasma membrane; Per, peroxisome; Nuc, nucleus; Mt, mitochondrion; Golgi, Golgi apparatus; Ex, extracellular; ER, endoplasmic reticulum; Cy, cytosol. Subcellular localization was annotated according to SUBAcon from the SUBA4 database. (D) Venn diagrams illustrating overlaps of upregulated (left) or downregulated (right) genes identified by transcriptomic analyses in the var1-1 or var2-15 mutants compared to the WT under control and cold-stress conditions.

Figure 3

Cold sensitivity of var1-1 is highly correlated with SA. (A) GO enrichment analysis of genes specifically upregulated in the var1-1 mutant under cold stress (1334 genes, Figure 2D). Immune/defense-related pathways were highly enriched, particularly the GO term “response to SA”. The top 20 enriched GO terms are shown. (B) Weighted gene coexpression network analysis (WGCNA) of transcriptomic data from WT, var1-1, and var2-15 under control and cold-stress conditions. The hierarchical cluster dendrogram shows consensus gene coexpression modules (CMs). In the dendrogram, each leaf represents a gene, with each module labeled by color (right side). Genes with FPKM > 1 in all three biological replicates for at least one genotype under one condition (18 909 genes in total) were analyzed. CMs, coexpression modules. (C) Correlation analysis of eigengenes from each module with genotypes and growth conditions (WT, var1-1, and var2-15; under control and cold stress). In each heatmap unit, the upper value indicates the correlation coefficient between the module eigengene and trait; the lower value indicates the corresponding p-value. Among 27 identified modules, the turquoise module (marked by an asterisk) showed the highest eigengene correlation with cold-stressed var1-1. (D) Heatmap visualizing the expression of SA-related genes. This gene set represents the overlap among genes in the GO term “response to SA” in (A), genes from the turquoise module in (C), and genes from cluster 1 in Supplemental Figure 2C. Relative gene expression is compared with WT under control conditions. FC, fold change.

Figure 4

Mutation of SA-related genes partially suppresses cold sensitivity in var1-1. (A) Introducing mutations in ICS1, PAD4, or EDS1, or overexpressing the bacterial salicylate hydroxylase gene NahG, substantially suppressed the chlorosis and reduced seedling size phenotypes of var1-1 under cold stress. Seedlings were grown under control (22°C for 11 days) or cold-stress conditions (12°C day/8°C night for 50 days). Scale bars, 1 cm. (B–D) Measurements of fresh weight (B), chlorophyll (Chl) content (C), and Fv/Fm (D) of seedlings shown in (A). Numbers on or above the histogram columns indicate the relative values compared to WT. Data are means ± SD (n = 6 for fresh weight, n = 3 for chlorophyll content, and n = 6 for Fv/Fm). For fresh-weight quantification, each replicate is an average of 10 seedlings. Statistical significance was determined by two-tailed, unpaired Student’s t-tests; corresponding p-values are indicated. FW, fresh weight.

Figure 5

¹O₂ is involved in the expression of the cold-sensitive phenotype of var1. (A and B) RT–qPCR analyses demonstrating strong upregulation of the SORGs SIB1, WRKY33, and WRKY40(A), as well as the oxidative stress-responsive genes ZAT10 and ZAT12(B). Data represent means ± SD from three biological replicates. (C) Venn diagram showing overlap between 168 eSORGs (Dogra et al., 2017) and genes specifically upregulated in var1-1 or var2-15 under cold stress. (D) SOSG staining visualizing ¹O₂ accumulation in leaves. Seedlings were grown under control conditions for 11 days or cold-stress conditions for 50 days. Scale bar, 200 μm. (E) Quantification of SOSG fluorescence intensity from (D). Eight to 10 microscopy images of individual seedlings per genotype were analyzed; approximately 10 cells per image were randomly selected (118–132 cells in total for different genotypes), and fluorescence intensities were quantified using ImageJ software. (F) Phenotypes of var1-1 single mutants and double/triple mutants involving ¹O₂ and SA signaling components under cold stress. Seedlings were grown under control conditions for 11 days or cold-stress conditions for 50 days. Scale bars, 1 cm. (G) Fresh weight measurements of seedlings shown in (F). Numbers on or above histogram columns indicate relative values for each genotype compared to WT. Data represent means ± SD (n = 8); each replicate is an average from 10 seedlings. (A, B, E, and G) Statistical significance was determined by one-way ANOVA (p < 0.05) with post hoc Tukey’s honestly significant difference (HSD) test.

Figure 6

EX2 promotes SA biosynthesis and signaling in cold-stressed var1-1. (A) Cluster of genes (profile 13 from Supplemental Figure 15) whose expression was upregulated in cold-stressed var1-1 and repressed in ics1 var1, eds1 var1, and ex2 var1 double mutants. (B) GO enrichment analysis of the 1,760 genes in profile 13 shown in (A). GO terms with enrichment fold > 2.0 and FDR < 0.05 are shown. (C) Heatmap depicting relative expression of SARGs from profile 13 shown in (A), which were differentially expressed between var1-1 and WT under cold stress. (D and E) RT–qPCR analyses demonstrating strong upregulation of SA signaling (D) and SA biosynthesis (E) genes in cold-stressed var1-1, significantly repressed by the introduction of the EX2 mutation. Relative gene expression was normalized to ACTIN and compared to WT under the control condition. (F) Overaccumulation of SA in cold-stressed var1-1 was largely repressed in the ex2 var1 double mutant. In (D–F), seedlings of different genotypes were grown under control conditions for 11 days or cold stress for 50 days. Data represent means ± SD. Statistical significance was determined by one-way ANOVA (p < 0.05) with post hoc Tukey’s honestly significant difference (HSD) test. n = 3 in (D) and (E), n= 4 in (F).

Figure 7

Critical role of EX2 in suppressing the expression of PhANGs. (A) Venn diagram showing overlap between gene sets C1 (genes differentially expressed in cold-stressed var1-1 restored by ICS1 mutation) and C2 (genes differentially expressed in cold-stressed var1-1 restored by EDS1 mutation) from Supplemental Figure 17A and 17B. (B) GO enrichment analysis of 360 genes specific to set C2 from (A). GO terms with enrichment fold > 2.5 and FDR < 0.05 are shown. (C) Venn diagram showing overlap between gene sets C3 (genes differentially expressed in cold-stressed var1-1 restored by EX1 mutation) and C4 (genes differentially expressed in cold-stressed var1-1 restored by EX2 mutation) from Supplemental Figure 17D and 17E. (D) GO enrichment analysis of 469 genes specific to set C4 from (C). GO terms with enrichment fold > 2.5 and FDR < 0.05 are shown. (E) Heatmap depicting relative expression of PhANGs identified from the GO terms in (B), (D), and Supplemental Figure 17C and 17F (highlighted in green). (F) Schematic illustration of ¹O₂ and SA signaling events in var1 in response to cold stress. Cold stress induces expression of FtsH metalloprotease subunits, which are required for photosystem biogenesis and degradation of damaged photosynthetic complex subunits. When FtsH5/VAR1 is absent, the content of functional FtsHs decreases significantly, impairing photosystem biogenesis and causing imbalanced accumulation of photosynthetic subunits (PSII) encoded by PhANGs and photosynthesis-associated plastid genes, leading to ¹O₂ accumulation. Concurrently, absence of CHLM may cause accumulation of tetrapyrrole intermediates that act as photosensitizers under light, further exacerbating ¹O₂ accumulation. Accumulated ¹O₂ acts as a plastid retrograde signal mediated by EX1/EX2, leading to upregulation of SORGs and thus SARGs, and SA biosynthesis, while repressing PhANG expression. EDS1/PAD4 may amplify this signaling through a feedback loop, resulting in SA overaccumulation in var1-1. These disruptions ultimately impair chloroplast biogenesis and inhibit the growth of var1-1 plants. Notably, EX2 plays a critical role in this process, likely partially independent of EX1. Proto IX, protoporphyrin IX; MgP, Mg-protoporphyrin IX; MgP MME, MgP monomethyl ester.