The DnaJ proteins DJA6 and DJA5 are essential for chloroplast iron–sulfur cluster biogenesis

Article15 April 2021free access Source DataTransparent process The DnaJ proteins DJA6 and DJA5 are essential for chloroplast iron–sulfur cluster biogenesis Jing Zhang Jing Zhang orcid.org/0000-0002-2921-5860 Key Laboratory of Photobiology, Institute of Botany, Photosynthesis Research Center, Chinese Academy of Sciences, Beijing, China State Key Laboratory of Crop Stress Adaption and Improvement, School of Life Sciences, Henan University, Kaifeng, China University of Chinese Academy of Sciences, Beijing, China Search for more papers by this author Zechen Bai Zechen Bai Key Laboratory of Photobiology, Institute of Botany, Photosynthesis Research Center, Chinese Academy of Sciences, Beijing, China University of Chinese Academy of Sciences, Beijing, China Search for more papers by this author Min Ouyang Min Ouyang State Key Laboratory of Biocatalysis and Enzyme Engineering, School of Life Sciences, Hubei University, Wuhan, China Search for more papers by this author Xiumei Xu Xiumei Xu State Key Laboratory of Crop Stress Adaption and Improvement, School of Life Sciences, Henan University, Kaifeng, China Search for more papers by this author Haibo Xiong Haibo Xiong Key Laboratory of Photobiology, Institute of Botany, Photosynthesis Research Center, Chinese Academy of Sciences, Beijing, China University of Chinese Academy of Sciences, Beijing, China Search for more papers by this author Qiang Wang Qiang Wang orcid.org/0000-0002-7388-4703 State Key Laboratory of Crop Stress Adaption and Improvement, School of Life Sciences, Henan University, Kaifeng, China Search for more papers by this author Bernhard Grimm Bernhard Grimm orcid.org/0000-0002-9730-1074 Institute of Biology/Plant Physiology, Humboldt-Universität zu Berlin, Berlin, Germany Search for more papers by this author Jean-David Rochaix Jean-David Rochaix Departments of Molecular Biology and Plant Biology, University of Geneva, Geneva, Switzerland Search for more papers by this author Lixin Zhang Corresponding Author Lixin Zhang [email protected] orcid.org/0000-0003-0268-6774 State Key Laboratory of Crop Stress Adaption and Improvement, School of Life Sciences, Henan University, Kaifeng, China Search for more papers by this author Jing Zhang Jing Zhang orcid.org/0000-0002-2921-5860 Key Laboratory of Photobiology, Institute of Botany, Photosynthesis Research Center, Chinese Academy of Sciences, Beijing, China State Key Laboratory of Crop Stress Adaption and Improvement, School of Life Sciences, Henan University, Kaifeng, China University of Chinese Academy of Sciences, Beijing, China Search for more papers by this author Zechen Bai Zechen Bai Key Laboratory of Photobiology, Institute of Botany, Photosynthesis Research Center, Chinese Academy of Sciences, Beijing, China University of Chinese Academy of Sciences, Beijing, China Search for more papers by this author Min Ouyang Min Ouyang State Key Laboratory of Biocatalysis and Enzyme Engineering, School of Life Sciences, Hubei University, Wuhan, China Search for more papers by this author Xiumei Xu Xiumei Xu State Key Laboratory of Crop Stress Adaption and Improvement, School of Life Sciences, Henan University, Kaifeng, China Search for more papers by this author Haibo Xiong Haibo Xiong Key Laboratory of Photobiology, Institute of Botany, Photosynthesis Research Center, Chinese Academy of Sciences, Beijing, China University of Chinese Academy of Sciences, Beijing, China Search for more papers by this author Qiang Wang Qiang Wang orcid.org/0000-0002-7388-4703 State Key Laboratory of Crop Stress Adaption and Improvement, School of Life Sciences, Henan University, Kaifeng, China Search for more papers by this author Bernhard Grimm Bernhard Grimm orcid.org/0000-0002-9730-1074 Institute of Biology/Plant Physiology, Humboldt-Universität zu Berlin, Berlin, Germany Search for more papers by this author Jean-David Rochaix Jean-David Rochaix Departments of Molecular Biology and Plant Biology, University of Geneva, Geneva, Switzerland Search for more papers by this author Lixin Zhang Corresponding Author Lixin Zhang [email protected] orcid.org/0000-0003-0268-6774 State Key Laboratory of Crop Stress Adaption and Improvement, School of Life Sciences, Henan University, Kaifeng, China Search for more papers by this author Author Information Jing Zhang1,2,3, Zechen Bai1,3, Min Ouyang4, Xiumei Xu2, Haibo Xiong1,3, Qiang Wang2, Bernhard Grimm5, Jean-David Rochaix6 and Lixin Zhang *,2 1Key Laboratory of Photobiology, Institute of Botany, Photosynthesis Research Center, Chinese Academy of Sciences, Beijing, China 2State Key Laboratory of Crop Stress Adaption and Improvement, School of Life Sciences, Henan University, Kaifeng, China 3University of Chinese Academy of Sciences, Beijing, China 4State Key Laboratory of Biocatalysis and Enzyme Engineering, School of Life Sciences, Hubei University, Wuhan, China 5Institute of Biology/Plant Physiology, Humboldt-Universität zu Berlin, Berlin, Germany 6Departments of Molecular Biology and Plant Biology, University of Geneva, Geneva, Switzerland *Corresponding author: Tel: +86 371 22919961; E-mail: [email protected] The EMBO Journal (2021)40:e106742https://doi.org/10.15252/embj.2020106742 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Fe–S clusters are ancient, ubiquitous and highly essential prosthetic groups for numerous fundamental processes of life. The biogenesis of Fe–S clusters is a multistep process including iron acquisition, sulfur mobilization, and cluster formation. Extensive studies have provided deep insights into the mechanism of the latter two assembly steps. However, the mechanism of iron utilization during chloroplast Fe–S cluster biogenesis is still unknown. Here we identified two Arabidopsis DnaJ proteins, DJA6 and DJA5, that can bind iron through their conserved cysteine residues and facilitate iron incorporation into Fe–S clusters by interactions with the SUF (sulfur utilization factor) apparatus through their J domain. Loss of these two proteins causes severe defects in the accumulation of chloroplast Fe–S proteins, a dysfunction of photosynthesis, and a significant intracellular iron overload. Evolutionary analyses revealed that DJA6 and DJA5 are highly conserved in photosynthetic organisms ranging from cyanobacteria to higher plants and share a strong evolutionary relationship with SUFE1, SUFC, and SUFD throughout the green lineage. Thus, our work uncovers a conserved mechanism of iron utilization for chloroplast Fe–S cluster biogenesis. Synopsis The identity of the iron donor during chloroplast iron-sulfur cluster biogenesis has remained unknown. Here, two Arabidopsis DnaJ-family chaperone proteins are found to promote iron incorporation into chloroplast Fe–S cluster proteins. Arabidopsis DJA6 and DJA5 are essential for photosynthesis, chloroplast development, and plant viability. DJA6 and DJA5 bind iron through their conserved cysteine residues DJA6 and DJA5 facilitate iron delivery to the SUFBC2D complex for Fe–S cluster assembly. DJA6 and DJA5 are highly conserved in photosynthetic organisms. Expression of cyanobacterial DJA6 rescues dja6 dja5 mutant phenotype in Arabidopsis. Introduction Some of the most ancient and important natural prosthetic groups are iron–sulfur clusters that are essential in all living organisms. They act as modular cofactors consisting of the transition Fe cations that are directly coordinated to inorganic sulfides (S2−) or the cysteinyl sulfurs of the associated proteins (Balk & Schaedler, 2014; Lill & Freibert, 2020; Braymer etal, 2021). As Fe can exist in either ferric or ferrous form under different redox conditions, the unique features of these versatile cofactors make Fe–S proteins well-suited for electron transfer and redox state sensing (Balk & Schaedler, 2014; Braymer etal, 2021). Fe–S proteins are involved in nearly all fundamental metabolic processes, such as electron transport, enzyme catalysis, regulation of gene expression, and sensing of environmental stimuli (Balk & Schaedler, 2014; Przybyla-Toscano etal, 2018; Lill & Freibert, 2020). Considering the extreme toxicity of excess free iron cations and sulfides to cells, the storage and delivery of these compounds mobilized for Fe–S cluster assembly needs to be tightly controlled (Braymer etal, 2021; Talib & Outten, 2021). Moreover, its reactive nature makes the nascent Fe–S cluster highly susceptible to cellular oxygen during biosynthesis (Mettert & Kiley, 2015; Mühlenhoff etal, 2020). Therefore, a sophisticated system has been established in living organisms to synthesize and deliver these cofactors. Bacteria have developed three distinct enzyme-mediated systems to synthesize Fe–S clusters, termed ISC (iron–sulfur cluster), SUF (sulfur utilization factor), and NIF (nitrogen fixation) system (Braymer etal, 2021). During evolution, the former two systems were transferred by endosymbiosis to eukaryotes to form the mitochondrial ISC and plastid SUF assembly systems, respectively (Zimorski etal, 2014; Braymer etal, 2021). Eukaryotes also developed the CIA (cytosolic iron–sulfur protein assembly) machinery for the biogenesis of cytosolic and nuclear Fe–S proteins (Netz etal, 2014; Paul & Lill, 2015; Ciofi-Baffoni etal, 2018; Tsaousis, 2019; Lill, 2020). The process of Fe–S cluster biogenesis in these systems can be commonly divided into three steps: 1. iron acquisition and sulfur mobilization; 2. assembly of iron and sulfur into a cluster; and 3. insertion into the target proteins (Braymer etal, 2021). To date, extensive studies have provided deep insights into the mechanisms of sulfur mobilization and the latter two steps of the assembly process (Przybyla-Toscano etal, 2018; Lill & Freibert, 2020; Talib & Outten, 2021). In the ISC system, class I cysteine desulfurase and sulfur transferase catalyze the pyridoxal 5'-phosphate (PLP)-dependent conversion of L-cysteine to generate the persulfide species and then deliver it to the IscU-type scaffold (Lill & Freibert, 2020; Braymer etal, 2021). Once the iron and electrons are available, the core ISC complex and the late-acting ISC machinery accomplish the building and targeting of the [2Fe–2S] and [4Fe–4S] clusters (Melber etal, 2016; Puglisi & Pastore, 2018; Gervason etal, 2019; Patra & Barondeau, 2019; Baussier etal, 2020; Lill & Freibert, 2020; Weiler etal, 2020). Additionally, the ISC components also export a sulfur-containing factor (X–S) to the cytosol for the CIA machinery (Lill etal, 2015; Ciofi-Baffoni etal, 2018; Pandey etal, 2018, 2019; Kassube & Thomä, 2020; Lill & Freibert, 2020; Braymer etal, 2021). In the SUF system, a persulfide is generated by a class II cysteine desulfurase (bacterial SufS or plant NFS2) with the activation from sulfur transferase (bacterial SufE or plant SUFE1-3) and then transferred to the SufBCD scaffold complex (Balk & Schaedler, 2014; Pérard & Ollagnier de Choudens, 2018; Blahut etal, 2019; Braymer etal, 2021). In the presence of iron and electrons, the newly made Fe–S cluster is assembled on the SUFB/D dimer interface with ATP hydrolysis catalyzed by the ABC-type ATPase SUFC subunit and is then released and delivered to apo-proteins through various carrier proteins, such as A-type carriers (bacterial SufA or plant SUFA), NifU-like proteins (bacterial NfuA or chloroplast NFUs), glutaredoxins (plant GRX14/16), and plant HCF101 (Balk & Schaedler, 2014; Hirabayashi etal, 2015; Hu etal, 2017; Yuda etal, 2017; Przybyla-Toscano etal, 2018; Garcia etal, 2019; Berger etal, 2020; Blahut etal, 2020; Roland etal, 2020). Although the process of sulfur mobilization for the Fe–S cluster assembly has been well-described, the nature of the iron donor remains unclear. The iron storage protein, ferritin, was previously assumed to be an iron donor. However, ferritin accumulates mainly in non-green plastids, and loss of ferritin did not have an impact on the accumulation of plastid Fe–S proteins in the Arabidopsis triple ferritin mutant (atfer1-3-4) (Ravet etal, 2009; Briat etal, 2010; Connorton etal, 2017). Since the triple mutant exhibits increased sensitivity to excess iron, it seems that ferritin serves to sequester iron in order to prevent oxidative damage (Ravet etal, 2009; Arosio etal, 2015; Reyt etal, 2015; Bradley etal, 2016). Overall, the factors involved in iron supply to the chloroplast SUF system are still unknown. To identify the iron donor for chloroplast Fe–S cluster assembly, we performed a co-expression analysis of members of the chloroplast SUF apparatus in Arabidopsis and identified two novel components for the assembly of chloroplast Fe–S clusters, AtDJA6 and AtDJA5, based on their potential functions derived from our biochemical analysis. Our results demonstrate that DJA6 and DJA5 can bind iron through their conserved cysteine residues and deliver iron to the SUFBC2D complex by interacting with components of the chloroplast SUF machinery. Loss of DJA6 and DJA5 in Arabidopsis leads to a lethal phenotype, together with a large decrease in the accumulation of Fe–S proteins and a marked iron overload in chloroplasts. Thus, our studies provide evidence that DJA6 and DJA5 may be involved in iron delivery during chloroplast Fe–S cluster assembly in Arabidopsis, a process that is highly conserved in photosynthetic organisms. Results Co-expression analysis of the components of the Arabidopsis chloroplast SUF system Five Arabidopsis genes have been reported to be directly involved in chloroplast Fe–S cluster assembly. They include AtNFS2 (encoding cysteine desulfurase), AtSUFE1 (encoding the activator of NFS2), AtSUFB, AtSUFC, and AtSUFD (encoding three subunits of the scaffold complex). To search for new components participating in iron supply for the assembly of iron–sulfur clusters, we took advantage of the public transcriptome database ATTED-II (version 9.2) (Obayashi etal, 2018) and queried the lists of co-expressed genes. Interestingly, two DnaJ type-I genes DJA6 (At2g22360) and DJA5 (At4g39960) were significantly co-expressed with each of the 5 genes mentioned above, with a high significance score (P < 0.001) (Fig EV1 and Appendix Tables S1-S5). Click here to expand this figure. Figure EV1. Correlations of expression patterns between AtDJA6/5 and genes encoding Arabidopsis chloroplast SUF components were retrieved from ATTED-II (version 9.2) Both axes are relative gene expression values in base-2 logarithm against the averaged expression levels of each gene. AtDJA6, At2g22360; AtDJA5, At4g39960; cysteine desulfurase AtNFS2, At1g08490; the activator of cysteine desulfurase AtSUFE1, At4g26500; the scaffold complex subunits AtSUFB, At4g04770; AtSUFC, At3g10670; AtSUFD, At1g32500. Download figure Download PowerPoint Histochemical analysis of DJA6 and DJA5promoter:GUS-expressing transgenic lines revealed DJA6 and DJA5 promoter activity throughout the entire development (Appendix Fig S1). Fluorescence analysis localized the DJA6- and DJA5-GFP fusion proteins in chloroplasts (Appendix Fig S2A). The fluorescence signals of GFP-fused DJA6 and DJA5 displayed punctate patterns which resembled those of GFP-fused chloroplast Fe–S cluster assembly proteins SUFB, SUFC, SUFD, SUFA, GRXS14, and HCF101 (Appendix Fig S2B; Abdel-Ghany etal, 2005; Hjorth etal, 2005). Moreover, we also co-transformed DJA6-GFP or DJA5-GFP fusion proteins with FRO1-mCherry (mitochondrial-localized FROSTBITE1) (Lee etal, 2002) to demonstrate a clear difference in signals for chloroplast and mitochondrial localization (Appendix Fig S2C). DJA6 and DJA5 are essential for plant viability and chloroplast development To gain insights into the physiological role of these two DJAs, we ordered the Arabidopsis T-DNA insertion line which carries a T-DNA insertion in the eighth exon of At2g22360 (Appendix Fig S3A and C). Neither DJA6 transcripts nor an immuno-reactive band was detected in this dja6 mutant (Appendix Fig S3B and D) indicating that it is a knockout mutant. A knockout mutant of DJA5 was generated by using the CRISPR/Cas9-based genome editing approach (Xing etal, 2014). This mutant was found to contain an additional nucleotide “T”- insertion at position +1544 (Appendix Fig S3A). Moreover, neither DJA6 nor DJA5 was detectable in the dja6dja5 double mutant (Appendix Fig S3D). The dja6 and dja5 single mutants appear to be phenotypically normal under standard growth conditions, whereas the homozygous dja6dja5 double mutant is seedling-lethal on soil (Fig 1A–C). When grown heterotrophically, dja6dja5 displayed an albino phenotype containing only 0.5% of WT chlorophyll (Fig 1A, B, and D). During further vegetative growth, leaves of dja6dja5 were strongly notched and the seedlings developed tiny pale yellow floral stalks producing only a few sterile flowers as siliques did not develop further (Fig 1B) suggesting an essential function of DJA6 and DJA5 for plant viability. The deficiency of one DJA variant is compensated most likely by the other protein, reflecting their functional redundancy. Complementation of the dja6dja5 double mutant with DJA6 or DJA5 was able to restore the wild-type phenotype (Fig 1B and Appendix Table S6). Analysis of the ultrastructure of plastids by transmission electron microscopy revealed normally differentiated and crescent-shaped plastids in WT, dja6, and dja5 leaves containing thylakoid membranes with grana stacks, while dja6dja5 contained smaller and irregularly shaped plastids with many vesicles of various sizes in the stroma, but no thylakoid membranes (Fig 1E), indicating that loss of both DJA6 and DJA5 leads to a severe disturbance of plastid biogenesis. Figure 1. dja6dja5 and RNAi mutants display albino and variegation phenotypes, respectively, and defects in chloroplast development A. Phenotypes of WT, dja6, dja5, dja6dja5, and RNAi seedlings on MS medium with 2% sucrose. Scalebars = 0.5 cm. B, C. Phenotypes and chlorophyll fluorescence images of 6-week-old WT, dja6dja5 and complemented plants (B) on MS medium with 2% sucrose, scalebars = 0.5 cm; WT, dja6, dja5, and RNAi mutants (C) grown on soil, scalebars = 1 cm. Fluorescences in (B) and (C) were measured with the FluorCam700MF and visualized using a pseudocolor index as indicated on the right. D. Chlorophyll content of total leaves in 6-week-old WT, dja6, dja5, dja6dja5, and RNAi seedlings. FW, fresh weight. Data are the means ± SEM (n = 7 biological replicates). E. The ultrastructure of plastids from WT, dja6, dja5, dja6dja5, and RNAi mutants. T, thylakoids. Scalebars = 0.2 μm. Source data are available online for this figure. Source Data for Figure 1 [embj2020106742-sup-0008-SDataFig1.zip] Download figure Download PowerPoint DJA6 and DJA5 mainly affect the accumulation of chloroplast Fe–S proteins Next, we analyzed the abundance of chloroplast Fe–S proteins, representative proteins of the photosynthetic electron transport chain and of the chloroplast SUF apparatus in dja6dja5. The core subunits of PSII (D1 and CP43), cytochrome b6f complex (cytochrome f, cytochrome b6, and Rieske), PSI (PsaA/B and PsaC), and Fe–S cluster-containing proteins (SUFA; HCF101, HIGH CHLOROPHYLL FLUORESCENCE 101; PAO, pheophorbide a oxygenase) were not detectable in dja6dja5. Moreover, other photosynthetic proteins including PSII antenna protein LHCII (light-harvesting complex II) and oxygen-evolving complex subunit PsbO, as well as the components of the chloroplast SUF system (SUFB, SUFD and SUFE1), were significantly reduced in the dja6dja5 double mutant (Fig 2A). Figure 2. Loss of DJA6 and DJA5 mainly affects the accumulation of the Fe–S cluster-containing proteins in chloroplasts A, B. Total protein was extracted from 4-week-old WT, dja6, dja5, dja6dja5, and RNAi seedlings, respectively. WT and dja6dja5 used in (A) were grown on MS medium with 2% sucrose, while WT, dja6, dja5, and RNAi mutants were used in (B) were grown on soil. 10 μg protein was loaded, except for detection with anti-SUFE1 and anti-SUFA antiserum, for which 25 μg was loaded. Specific bands were identified by immunoblotting and by their molecular weight, and are indicated by asterisks. Designations of photosynthetic protein complexes and their diagnostic components are labeled on the left and right, respectively. The Fe–S types of Fe–S proteins are shown in parentheses. Actin served as controls to normalize protein levels. For each protein, three independent biological replicates were performed and a representative one is shown. Source data are available online for this figure. Source Data for Figure 2 [embj2020106742-sup-0009-SDataFig2.pdf] Download figure Download PowerPoint To explore in more detail the fate of chloroplast Fe–S proteins in seedlings deficient in DJA6 and DJA5, the alternative approach of RNA interference (RNAi) was used to suppress the expression of DJA5 in the dja6 mutant background. Three representative RNAi transgenic lines had gradually reduced DJA5 mRNA levels (with ~10, ~30, and up to ~70% of WT levels) and corresponding lower DJA5 protein contents (Appendix Fig S3B and D). Cotyledons were very pale or variegated, while chlorosis and variegation were first visible in the younger true leaves and became more pronounced upon further growth (Fig 1A and C). Furthermore, the plastid morphology of the RNAi mutants resembled that of the dja6dja5 mutant (Fig 1E). In the RNAi mutants, the content of Fe–S proteins PsaA/B, PsaC, Rieske, SUFA, HCF101, and PAO were significantly decreased (Fig 2B). In contrast, the content of non-Fe–S subunits of PSII (D1, CP43, LHCII and PsbO) and components of the chloroplast SUF machinery (SUFB, SUFD and SUFE1) were only moderately decreased (Fig 2B), implying that the Fe–S protein levels of DJAs-deficient plants are drastically reduced, and the lower amount of these proteins often affects the accumulation of other proteins belonging to a common complex. These severe impairments in DJAs-deficient mutants suggested a block in photosynthetic electron transport. As expected, the ratio of variable to maximal fluorescence (Fv/Fm) and photochemical quenching (qP) was drastically decreased in dja6dja5 and RNAi mutants compared with WT plants (Fig 1C, Appendix Fig S4A and Table S6), which is indicative of impaired electron transport in PSII. The magnitude of reduction of oxidized cytochrome f was also diminished (Appendix Fig S4B). Moreover, no absorbance changes or a large decrease in the photooxidation of P700 was detected in dja6dja5 and RNAi mutants, suggesting a severe impairment of the PSI reaction center, while illumination with a saturating light pulse with far-red light background led to a complete re-reduction of P700, indicating sufficient electron supply from PSII for the re-reduction of impaired PSI in these mutants (Appendix Fig S4C). Taken together, these spectroscopic analyses indicate that electron transports through cytochrome b6f complex and PSI is strongly disturbed in the dja6dja5 and RNAi mutants, which correlates with the loss of Fe–S proteins of these two photosynthetic protein complexes. DJA6 and DJA5 play key roles in the maintenance of cellular iron homeostasis To further investigate the cellular functions of DJA6 and DJA5 in assembly of chloroplast Fe–S clusters, we measured the metal concentrations in dja6dja5 and RNAi mutants by inductively coupled plasma mass spectrometry (ICP-MS). Notably, both the dja6dja5 and RNAi plants accumulated much higher levels of Fe in total seedlings than WT, whereas there were no marked differences in Co, Cu, and Mn content, and a moderate increase of Zn content (Figs 3A and B, and EV2A and B). We also isolated chloroplasts to measure the iron concentration in chloroplasts from RNAi plants and found that the chloroplasts of these plants accumulated five times more Fe than WT (Figs 3C, and EV2C). However, in mitochondria, there was no change in the accumulation of metals including Co, Cu, Mn, Zn, and Fe between WT and RNAi plants (Fig EV2D). Figure 3. DJA6 and DJA5 are critical for the maintenance of cellular iron homeostasis A–C. Metal concentrations in the WT, dja6dja5, and RNAi mutants. Fe concentrations in total leaves (A and B) and total chloroplast (C) from 4-week-old WT, dja6dja5, and RNAi mutants were quantified by inductively coupled plasma mass spectrometry and are shown as means ± SEM (n = 3 biological replicates) of μg g−1 dry weight (DW). WT and RNAi mutants used in (A and C) were grown on soil, while WT and dja6dja5 seedlings used in (B) were grown on MS medium with 2% sucrose. Similar results were obtained in two additional independent biological experiments. Asterisks indicate significant differences from the value of WT (two-sample Student’s t-test; *P < 0.05; **P < 0.01; ***P < 0.001). D. Volcano plot of the result from quantitative transcriptome profiling in WT and dja6dja5 mutant. Statistical analysis was performed using Student’s t-test. Data on the X-axis represent the Log2-ratio of FC (fold change) between the absolute abundances of transcripts identified from dja6dja5 against WT. The Y-axis represents the Log10 of FDR (false discovery rate). All DEGs (differentially expressed genes) with FDR < 0.01 and FC > 2 are marked in gray, while all iron-related DEGs with FDR < 0.01 and FC > 2 are marked in red. The iron-related DEGs labeled with red dots within the black boxes are referred to in the text. All data are based on three biological replicates. Source data are available online for this figure. Source Data for Figure 3 [embj2020106742-sup-0010-SDataFig3.xlsx] Download figure Download PowerPoint Click here to expand this figure. Figure EV2. Metal concentrations in the WT, dja6dja5, RNAi mutants, and SynDJA6-FLAG/dja6dja5 transgenic plants A–F. Metal concentrations in total leaves (A, B, and E), total chloroplasts (C and F), and total mitochondria (D) from 4-week-old WT, dja6dja5, RNAi mutants, and 35Spro:SynDJA6-FLAG/dja6dja5#24 transgenic plants were quantified by inductively coupled plasma mass spectrometry and are shown as means ± SEM (n = 3 biological replicates) of μg g−1 dry weight (DW). WT, RNAi mutants, and SynDJA6-FLAG/dja6dja5#24 transgenic plants used in (A, C, D, E, and F) were grown on soil, while WT and dja6dja5 seedlings used in (B) were grown on MS medium with 2% sucrose. Similar results were obtained in two additional independent biological experiments. Asterisks indicate significant differences from the value of WT (two-sample Student’s t-test; *P < 0.05). Source data are available online for this figure. Download figure Download PowerPoint To elucidate the probable role of DJA6 and DJA5 in the maintenance of iron homeostasis, we compared gene expression profiles of WT and dja6dja5, and found 1,136 differentially expressed genes (DEGs) (Fig 3D, Dataset EV1). GO analysis revealed that these DEGs were enriched in pathways linked to the responses to chemicals and to oxygen-containing compounds and to the regulation of iron transport (Appendix Fig S5). Consistent with our ICP-MS data, dja6dja5 showed significantly decreased gene expression levels of transcription factors responding to iron starvation (bHLH38, bHLH39, and bHLH101), various iron and iron–sulfur cluster binding proteins and iron–sulfur cluster assembly factor (SUFE2), while gene expression levels of chloroplast iron superoxide dismutase (FSD3) and iron transporter (IRT3) were largely increased compared to WT (Fig 3D) (Murthy etal, 2007; Myouga etal, 2008; Yuan etal, 2008; Lin etal, 2009; Wang etal, 2013). Moreover, there were no significant differences in the expression levels of SUF components (NFS2, SUFE1, SUFB, SUFC, and SUFD) (Dataset EV1). These results suggest that DJA6/5 may be involved in the maintenance of cellular iron homeostasis. DJA6 and DJA5 bind iron in vitro and in vivo To investigate the biochemical function of DJA6 and DJA5, recombinant MBP-His-tagged DJA6 and DJA5 proteins were purified. Interestingly, after incubation with the protein extracts and wash, the Ni-NTA His-bound resin slowly turned to purple upon exposure to air. During elution, the purple color diminished, and the purified DJA6 and DJA5 were found to be red-colored (Fig 4A). To identify the chromophore of the recombinant proteins, UV-visible absorption spectral analyses were performed. Both DJA6 and DJA5 showed prominent absorbance maxima at 365 and 485 nm, and a broader peak at 570 nm (Fig 4A). These spectral properties are similar to classical rubredoxin (Rd) fold-containing proteins which are known to bind iron (Proudfoot etal, 2008). To distinguish the chromophore of DJA6 and DJA5 from that of other plastid Fe–S proteins, the UV-visible absorption spectra of recombinant GRXS14, HCF101, and NFU2 were characterized. Each protein was brownish with a typical peak at 420 nm (Fig EV3A and B), indicating that the absorbance spectrum of DJA6 and DJA5 was derived from bound iron, but not from an iron–sulfur center (Yabe etal, 2004; Bandyopadhyay etal, 2008; Schwenkert etal, 2009; Gao etal, 2013). A resonance at g ~ 4.3 was observed in DJA6 and DJA5 proteins by electron paramagnetic resonance spectroscopy (EPR) (Fig 4B), characteristic of a high spin ferric iron in tetrahedral geometry of oxidized rubredoxins (Rao etal, 1972; Arnold etal, 2016). Figure 4. DJA6 and DJA5 bind iron invitro and invivo UV-visible absorption spectra of recombinant DJA6 and DJA5. Recombinant DJA6 after elution from Ni-NTA Sepharose beads follow