Extrachloroplastic PP7L Functions in Chloroplast Development and Abiotic Stress Tolerance1[OPEN] Duorong Xu,a Giada Marino,a Andreas Klingl,b Beatrix Enderle,c Elena Monte,d Joachim Kurth,b Andreas Hiltbrunner,c,e Dario Leister,a and Tatjana Kleine a,2,3 a Plant Molecular Biology (Botany), Department Biology I, Ludwig-Maximilians-Universität München, 82152 Planegg-Martinsried, Germany b Plant Development, Department Biology I, Ludwig-Maximilians-Universität München, 82152 Planegg-Martinsried, Germany c Institute of Biology II, Faculty of Biology, University of Freiburg, 79104, Freiburg, Germany d Plant Development and Signal Transduction Program, Center for Research in Agricultural Genomics Consejo Superior de Investigaciones Científicas-Institute of Agrifood Research and Technology-Universidad Autonoma de Barcelona-Universidad de Barcelona, 08193 Barcelona, Spain e Centre for Biological Signalling Studies (BIOSS), University of Freiburg, 79104 Freiburg, Germany ORCID IDs: 0000-0002-1525-4244 (G.M.); 0000-0002-7340-9355 (E.M.); 0000-0003-0438-5297 (A.H.); 0000-0003-1897-8421 (D.L.); 0000-0001-6455-3470 (T.K.). Chloroplast biogenesis is indispensable for proper plant development and environmental acclimation. In a screen for mutants affected in photosynthesis, we identified the protein phosphatase7-like (pp7l) mutant, which displayed delayed chloroplast development in cotyledons and young leaves. PP7L, PP7, and PP7-long constitute a subfamily of phosphoprotein phosphatases. PP7 is thought to transduce a blue-light signal perceived by crys and phy a that induces expression of SIGMA FACTOR5 (SIG5). We observed that, like PP7, PP7L was predominantly localized to the nucleus in Arabidopsis (Arabidopsis thaliana), and the pp7l phenotype was similar to that of the sig6 mutant. However, SIG6 expression was unaltered in pp7l mutants. Instead, loss of PP7L compromised translation and ribosomal RNA (rRNA) maturation in chloroplasts, pointing to a distinct mechanism influencing chloroplast development. Promoters of genes deregulated in pp7l-1 were enriched in PHYTOCHROME-INTERACTING FACTOR (PIF)-binding motifs and the transcriptome of pp7l-1 resembled those of pif and CONSTITUTIVE PHOTOMORPHOGENESIS1 (COP1) signalosome complex (csn) mutants. However, pif and csn mutants, as well as cop1, cryptochromes (cry)1 cry2, and phytochromes (phy)A phyB mutants, do not share the pp7l photosynthesis phenotype. PhyB protein levels were elevated in pp7l mutants, but phyB overexpression plants did not resemble pp7l. These results indicate that PP7L operates through a different pathway and that the control of greening and photosystem biogenesis can be separated. The lack of PP7L increased susceptibility to salt and high-light stress, whereas PP7L overexpression conferred resistance to high- light stress. Strikingly, PP7L was specifically recruited to Brassicales for the regulation of chloroplast development. This study adds another player involved in chloroplast biogenesis. Chloroplasts are the site of photosynthesis and are 1 essential for plant growth and development. They are This work was supported by the Deutsche Forschungsgemein- the prototypes of a diverse family of organelles, the schaft (project C01 no. KL 2362/1-1 to T.K. and project C05 no. plastids, which are found in plants and various algae TRR175 to D.L.). 2 Author for contact: tatjana.kleine@lmu.de. (Keeling, 2010). As descendants of cyanobacteria-like 3 Senior author. progenitors, plastids of higher plants are semiautono- The author responsible for distribution of materials integral to the mous organelles that have retained a reduced genome findings presented in this article in accordance with the policy de- of ;100 genes (Kleine et al., 2009). However, they scribed in the Instructions for Authors (www.plantphysiol.org) is: also contain ;2,000–3,000 nuclear-encoded proteins Tatjana Kleine (tatjana.kleine@lmu.de). (Abdallah et al., 2000; Yu et al., 2008). Consequently, Research was designed by T.K. and D.X.; the pp7l:En-1 mutant was chloroplast multiprotein complexes, such as the pho- confirmed by J.K.; most of the experiments were performed by D.X.; tosynthetic apparatus and the chloroplast gene ex- G.M. realized the shot-gun proteomics experiment; A.K. provided pression machinery, comprise a mixture of proteins TEM pictures; B.E. performed phyB western-blot analysis; T.K., D. encoded by chloroplast and nuclear genes, and chlo- X., G.M., D.L., A.H., and E.M. analyzed the data; T.K. wrote the manuscript with input primarily from D.L., A.H., and E.M., and then roplast function therefore requires stringent coordina- all authors; the whole study was supervised by T.K.; all authors read tion of the expression of the two genomes (Woodson and approved the article. and Chory, 2008). This is achieved by bidirectional ex- [OPEN] Articles can be viewed without a subscription. change of information between the organelles. During www.plantphysiol.org/cgi/doi/10.1104/pp.19.00070 retrograde signaling, chloroplasts send out signals that Plant PhysiologyÒ, May 2019, Vol. 180, pp. 323–341, www.plantphysiol.org Ó 2019 American Society of Plant Biologists. All Rights Reserved. 323 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Role of PP7L in Chloroplast Development inform the nucleus about their developmental and RNA polymerase (PEP; Pfannschmidt et al., 2015). For physiological status so that nuclear gene expression can its function, the multisubunit enzyme PEP requires a set be adjusted accordingly (Bobik and Burch-Smith, 2015; of polymerase-associated proteins and sigma factors Chan et al., 2016; Kleine and Leister, 2016). Because (SIGs), all of which are encoded in the nucleus (Chi most chloroplast proteins are encoded in the nucleus, et al., 2015; Pfannschmidt et al., 2015). In Arabidopsis the nucleus exercises anterograde control over the (Arabidopsis thaliana), there are six different SIGs, des- plastids by regulating various functions, in particular ignated as SIG1–SIG6. Depending on their promoter chloroplast gene expression (Tiller and Bock, 2014; selectivity, SIGs may have overlapping functions as Schmitz-Linneweber et al., 2015; Shikanai, 2015). well as specialized roles (Chi et al., 2015). SIG5, for ex- In true leaves, chloroplasts develop from meriste- ample, effectively mediates blue-light signaling by ac- matic proplastids when the leaf primordia emerge; in tivating transcription from the “blue-light–responsive” cotyledons, chloroplasts develop from proplastids/ psbD promoter (Nagashima et al., 2004), as its own ex- eoplasts or etioplasts present in mesophyll tissue within pression is strictly dependent on blue light. The re- the embryo (Waters and Langdale, 2009; Liebers et al., sponse to blue-light signals in the nucleus is assumed to 2017). These precursors are rapidly converted into involve the perception of blue light by crys and phyA, chloroplasts upon exposure to light. Thus, the devel- and transduction of the signal by the nucleus-localized opment of photosynthetically competent chloroplasts protein phosphatase PP7, which in turn leads to the in angiosperm seedlings is strictly dependent on ex- induction of SIG5 expression (Møller et al., 2003; Chi posure to light (Pogson and Albrecht, 2011) and in- et al., 2015). PP7 is a member of the family of Ser/Thr- volves the so-called chloroplast biogenic signaling specific phosphoprotein phosphatases (PPPs; Farkas pathway, which triggers massive changes in the nuclear et al., 2007). The Arabidopsis PPP family comprises 26 transcriptome, with at least 20% of the genome showing members, which can be assigned to seven subfamilies. light-regulated differential expression during seedling PPPs have roles in abscisic acid, auxin, and brassinos- development (Jiao et al., 2007; Jarvis and López-Juez, teroid signaling, phototropism, regulating the target of 2013). Light is perceived mainly by two families rapamycin pathway, cell stress responses. and flower- of photoreceptors—the phytochromes (phys) and the ing time (Uhrig et al., 2013; Lillo et al., 2014). For ex- cryptochromes (crys)—which are sensitive to red/ ample, phy-associated Ser/Thr protein phosphatase, far-red and UVA/blue light, respectively, and undergo which belongs to the type-6 subfamily, dephosphory- conformational changes upon light perception that lates phyA in vitro and delays flowering (Kim et al., enable them to interact with downstream signaling 2002). Members of the PPP family are present in all partners (Jiao et al., 2007). The latter include PHYTO- eukaryotes. However, the type-7 subfamily is unique to CHROME-INTERACTING FACTOR (PIF)-like tran- plants. In Arabidopsis, this subfamily contains three scription factors of the basic helix–loop–helix (bHLH) members—PP7, “long PP7,” and “inactive PP7” family and basic Leu zipper (bZIP) transcription fac- (Farkas et al., 2007; Uhrig et al., 2013). Only PP7 has tors, such as ELONGATED HYPOCOTYL5 (HY5) and been characterized in detail, and shown to regulate HY5-like HYH (Jiao et al., 2007; Leivar and Monte, blue-light (Møller et al., 2003) as well as red/far-red 2014). PIFs mainly repress photomorphogenesis genes light signaling (Genoud et al., 2008). Long PP7 is also in the dark, whereas HY5 and HYH activate those same designated “MAINTENANCE OF MERISTEMS-LIKE3” genes upon illumination. Hence, PIFs and HY5/HYH and encodes a protein bearing a putative aminotrans- have antagonistic functions in the regulation of e.g. ferase domain in addition to the PP7 domain. The phe- protochlorophyllide production and reactive oxygene notype of mail3-2, in which a transfer-DNA (T-DNA) is species (ROS)-responsive genes during the dark-to- inserted in the PP7 domain, is indistinguishable from light transition (Chen et al., 2013). In the dark, HY5 that of wild-type plants (Ühlken et al., 2014). is targeted for degradation by the ubiquitin ligase Here, we characterize PP7-like (pp7l) mutants, the CONSTITUTIVE PHOTOMORPHOGENESIS1 (COP1) first of which was identified in a screen for mutants in the nucleus (Osterlund et al., 2000). Therefore, COP1 defective in photosynthesis. The PP7L gene encodes the suppresses photomorphogenic development when PP7 homolog PP7L (previously designated as “inactive translocated to the nucleus (Yi and Deng, 2005) and PP7”; Farkas et al., 2007). PP7L is localized to the nu- its nuclear localization is regulated by a component cleus, and is a positive regulator of protein synthesis in of the COP9 signalosome complex (CSN; Wang et al., the developing chloroplast. However, it does not act by 2009). Conversely, upon light exposure, phys trigger modulating SIG factor gene expression like PP7. In- PIF degradation and COP1 inactivation (Gommers stead, the pp7l mutant is shown here to be defective in and Monte, 2018). Strikingly, the HEMERA (HMR)/ chloroplast ribosomal RNA (rRNA) maturation, and transcriptionally active chromosome protein 12 (pTAC12) consequently in mRNA translation. Promoter analysis not only contributes to phy signaling in the nucleus, but of genes deregulated in the pp7l-1 mutant and database also forms part of the chloroplast’s pTAC and is involved analysis of conditions or mutations associated with in plastid gene expression (Chen et al., 2010). gene expression changes similar to those seen in pp7l- In higher plants, transcription of plastid DNA is 1 suggested a tentative association of PP7L with PIFs carried out by a nuclear-encoded phage-type RNA and other light signaling components, but neither pif, polymerase and a plastid-encoded prokaryotic-like csn, cop1, cry1 cry2, or phyA phyB mutants display a 324 Plant Physiol. Vol. 180, 2019 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Xu et al. photosynthesis phenotype. Moreover, although phyB this vector contains the 35S promoter on the left border levels are enhanced in pp7l mutants, overexpression of site (Baulcombe et al., 1986) and can potentially activate phyB does not induce a photosynthesis phenotype. Seed flanking genomic sequences (Ulker et al., 2008). How- germination of pp7l mutants was reduced by exposure to ever, because of these differences in PP7L transcript salt and high light, whereas overexpression of PP7L ren- accumulation, all pp7l mutant lines were transformed dered 4-week–old plants more tolerant to high light. with a genomic DNA fragment comprising the coding sequence from the start to the stop codon of AT5G10900 fused upstream of the enhanced green fluorescence protein (eGFP) reporter gene, which was placed under RESULTS the control of the Cauliflower Mosaic Virus 35S pro- Identification and Phenotypic Analysis of Mutants for the moter. Independent transgenic lines for all three mu- PP7L Locus tants were obtained (overexpression of PP7L [oe-PP7L] ppl7l-1 to -3) and their phenotypic and chlorophyll Screening of an Arabidopsis mutant collection car- (Chl) fluorescence analyses showed that the wild- rying insertions of the maize transposable element En- type function had been restored in all oe-PP7L lines 1 (Wisman et al., 1998) for lines that show alterations in (Supplemental Fig. S2A). This shows that the pp7l the effective quantum yield of PSII, designated FII, phenotype is indeed caused by the insertions within the resulted in the recovery of a set of mutants with defects AT5G10900 gene. The striking dark-green coloration of in photosynthesis (Varotto et al., 2000). In one of these pp7l leaves (Fig. 1A) was found not to be caused by a (ZIGIA line V2-880), the effective (FII) and maximum higher overall Chl content. In fact both (Chl a+b) and the (Fv/Fm) quantum yields of PSII were significantly re- Chl a/b ratio were slightly, but not significantly, duced in emerging leaves of 3-week–old plants com- pared to the wild type (Col-0), implying a defect in energy transfer within PSII. Isolation of the genomic sequence flanking the En-1 transposon enabled identi- fication of the insertion site in the second exon of the gene AT5G10900 (Supplemental Fig. S1A). AT5G10900 is listed in the UniProtKB database (http://www. uniprot.org/uniprot/Q9LEV0) as “Ser/Thr-protein phosphatase 7 (PP7) inactive homolog.” Thus, AT5G10900 was designated “PP7L” and the mutant was named “pp7l:En-1.” Because the pp7l:En-1 mutant germinated poorly, further mutant lines were identi- fied with the SIGnAL T-DNA Express Arabidopsis Gene Mapping Tool (http://signal.salk.edu/cgi-bin/ tdnaexpress). In these lines, named “pp7l-1” (SALK_018295), “pp7l-2” (SALK_033071), and “pp7l-3” (SALK_022053), the pROK2 T-DNA was inserted in exon 4, intron 5, and exon 7 of PP7L at positions 651, 1951, and 2251 relative to the start codon, respec- tively (Supplemental Fig. S1, A and B). In all identi- fied pp7l mutants, both overall growth rates and Fv/Fm values were reduced in emerging leaves of 3-week– old soil-grown plants compared to the wild type (Supplemental Fig. S1A). According to The Arabidopsis Information Resource genome annotation 10, AT5G10900 is a single-copy gene with three predicted transcript splice forms AT5G10900.1, AT5G10900.2, and AT5G10900.3, which differ only at their 39 ends (Supplemental Fig. S1B). To Figure 1. Phenotypic characterization of pp7l T-DNA insertion mu- confirm that the altered expression of PP7L was re- tants. A, Phenotypes of 3-week–old wild-type (Col-0) and the different sponsible for the mutant phenotype, reverse transcrip- pp7l mutant plants grown in LD (16-h light/8-h dark) conditions. The tion quantitative PCR (RT-qPCR) was conducted. We maximum quantum yield of PSII (Fv/Fm) was measured with an imaging Chl fluorometer (Imaging PAM). Scale bar = 1 cm. B, The abaxial surface found that the PP7L transcript was barely detectable in of leaves of plants grown in LD conditions reveals enhanced anthocy- the pp7l-2 mutant and undetectable in pp7l-3, but its anin accumulation in pp7l plants. Anthocyanins were extracted and level was elevated in pp7l-1 (Supplemental Fig. S1C). quantified photometrically, and amounts are reported in arbitrary units The overexpression of the 39 segment of PP7L in the (a.u.). Data are shown as mean values 6 SD from three different plant pp7l-1 allele can be explained by the orientation of the pools. Each pool contained .20 plants. Significant differences between T-DNA integration in the pROK2 vector in the 59RB the data pairs were identified by Tukey’s test, and significant differences (right border)–T-DNA–left border 39 direction, because (P , 0.05) with respect to Col-0 are indicated by asterisks. Plant Physiol. Vol. 180, 2019 325 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Role of PP7L in Chloroplast Development reduced compared to the wild type (Supplemental Fig. was observed during seedling development. In 4-, 5-, 6- S2B). Instead, the darker color of pp7l leaves, which is , and 7-d–old seedlings, Fv/Fm was significantly de- most prominent on the abaxial side, can be attributed to pressed in all pp7l mutants to ;0.41, 0.54, 0.62, and 0.71, the approximately eightfold-higher content of antho- respectively, compared to the respective wild-type cyanins (compared to wild-type levels) found in these values of 0.75, 0.78, 0.79, and 0.80 (Fig. 3A), and this mutants (Fig. 1B). depressed Fv/Fm recovered completely after 10 d (Fig. 3B). Interestingly, this phenotype is reminiscent of those of sig2 and sig6 seedlings (Fig. 3B) and of young PP7L Is Localized to the Nucleus and the Cytosol leaves of the sig6 mutant (Fig. 3C). It was previously shown that sig2 and sig6 mutant seedlings are pale The pp7l mutants show a photosynthesis phenotype, green and accumulate fewer plastid- and nuclear- which at first sight suggests chloroplast localization of encoded transcripts for photosynthesis proteins, the PP7L protein. However, computational prediction of whereas sig1, sig3, sig4, and sig5 seedlings appear more the subcellular localization of PP7L by the algorithms like wild type (Woodson et al., 2013); for the sig5 mu- implemented in SUBA4 (Hooper et al., 2017; http:// tant, see Figure 3C. Because PP7 has been implicated in suba.plantenergy.uwa.edu.au) did not result in a the induction of SIG5 expression (reviewed in Chi et al., chloroplast localization, although various predictors 2015), we asked whether PP7L might similarly act assigned PP7L either to the nucleus, cytosol, or mito- through SIG-mediated gene expression changes. Be- chondria, and the SUBAcon algorithm predicted a cause sig2 and sig6 seedlings show a transient photo- nuclear localization for PP7L. A putative nuclear lo- synthesis phenotype that resembles the pp7l phenotype, calization signal (NLS) was indeed found by the cNLS SIG2 and SIG6 mRNA levels were studied in 4- and mapper (Kosugi et al., 2009; http://nls-mapper.iab. 5-d–old wild-type and pp7l-1 seedlings by RNA gel- keio.ac.jp/cgi-bin/NLS_Mapper_form.cgi), which pre- blot analysis. In fact, we found that levels of SIG2 and dicts a bipartite NLS (ISKRVLDSKLEACKFAFLKL- SIG6 transcripts were increased and not decreased in SAVKTTRMK) spanning the region from amino acids 6 pp7l-1 (Fig. 3D). to 34. To experimentally examine the subcellular lo- To obtain an overall insight into the RNA expression calization of PP7L, protoplasts were transformed with pattern of nuclear- and organelle-encoded genes when the 35S:PP7L-eGFP construct that had been used for PP7L function is compromised, RNAs isolated from 4- complementation of the pp7l alleles. Visualizing eGFP d–old wild-type and pp7l-1 mutant seedlings grown in fluorescence in the protoplasts confirmed that PP7L- long-day (LD; 16-h light/8-h dark) conditions were eGFP was targeted to the nucleus (Fig. 2). Moreover, subjected to long non-coding (lnc)RNA sequencing. several protoplasts displayed eGFP fluorescence in the With this method, the elevated levels of SIG2 (1.8-fold) cytosol, indicating a dual (nuclear and cytosolic) local- and SIG6 (1.7-fold) mRNA were confirmed. Also, SIG1, ization of PP7L. SIG3, and SIG4 transcripts were slightly increased, whereas SIG5 mRNA levels were almost unchanged (0.9-fold) in pp7l-1 (Fig. 3E; Supplemental Table S2). PP7L Is a Positive Regulator of Photosynthesis at Young Moreover, comparison of whole genome transcriptome Developmental Stages changes in pp7l-1 with previously published mRNA- sequencing (Seq) data for sig6 showed that pp7l-1 and In addition to the drop in Fv/Fm seen in emerging sig6 mutants do not provoke similar gene expression leaves of pp7l mutants, reduced photosynthetic activity changes (Fig. 3F). In summary, PP7L does not positively regulate SIG gene expression. Therefore, PP7L likely acts through a mechanism that is different from that of PP7. PP7L Is Involved in the Regulation of Chloroplast Gene Expression The lncRNA-Seq analysis revealed substantial changes in gene expression in 4-d–old pp7l-1 seedlings, specifically 1,198 and 1,004 genes whose mRNA levels were signifi- cantly (more than twofold) reduced or elevated, respec- tively (Supplemental Table S2). Genes encoding enzymes of the phenylpropanoid pathway leading to anthocyanin Figure 2. PP7L is localized to the nucleus and the cytosol. Fluorescence like DIHYDROFLAVONOL4-REDUCTASE (TRANS- microscopy of Arabidopsis protoplasts transiently expressing PP7L PARENT TESTA3), ANTHOCYANIDIN SYNTHASE, fused to eGFP. The eGFP fluorescence (green) and chloroplast auto- and the key enzyme CHALCONE SYNTHASE were fluorescence (red) are shown together in the overlay picture. Scale upregulated 14-, seven-, and fourfold, respectively, a bars = 10 mm. finding that is consistent with higher anthocyanin 326 Plant Physiol. Vol. 180, 2019 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Xu et al. accumulation in pp7l mutants (see Fig. 1B). Although photosynthetic activity is reduced in pp7l mutants (see Figs. 1A and 3A), no enrichment for genes related to photosynthesis or the chloroplast could be identified among those downregulated in pp7l-1 mutant seed- lings with the gene ontology (GO) analysis tool DAVID (Huang et al., 2009; Fig. 4A). Instead, under the categories “cellular component,” “biological process,” and “molecular function,” DAVID detected an enrichment of genes encoding proteins for the “RNA polymerase II transcription factor complex,” “response to sucrose starvation,” and “glutaredoxin activity,” respectively (Fig. 4A). These groups en- compass genes encoding bHLH transcription factors, or extracellular proteins like protease inhibitor/seed storage/lipid transfer family proteins, xyloglucan endotransglucosylase/hydrolase proteins, and per- oxidase superfamily proteins. However, among the elevated transcripts within the “cellular component” category, most component cate- gories were associated with the chloroplast, in partic- ular “nucleoid,” “plastid chromosome,” “chloroplast stroma,” “envelope,” and “thylakoid” (Fig. 4B). In the category “biological process,” the GOs “chloroplast rRNA processing” and “chlorophyll biosynthesis” were the most highly enriched (Fig. 4B). In particular, the rise in transcripts for proteins involved in chloro- plast RNA metabolism has already been noted in mu- tants and conditions that impair protein synthesis in the chloroplast (Leister and Kleine, 2016). Investigation of plastid-encoded transcripts in the lncRNA-Seq data showed that transcripts of only five genes were reduced by more than twofold. These en- code subunit 8 of the cytochrome b6 f complex, several PSII proteins (D1, PsbK, PsbI), and the PSI protein PsaJ (Supplemental Table S2). Conversely, expression levels of 44 plastid genes rose by more than twofold. Among these are genes encoding tRNAs, ribosomal proteins, subunits of RNA polymerase and ATP synthase, and Figure 3. Comparative analysis of pp7l-1 and sig (sig2 and sig6) mu- the PSII proteins CP47, D2, and PsbT. tants. A, Phenotypes of 4- to 7-d–old Col-0 and the different pp7l mutant Taken together, these transcriptome changes point to seedlings together with Imaging PAM pictures showing maximum a role for PP7L in posttranscriptional processes in the quantum yields of PSII (Fv/Fm) SDs are provided in Supplemental Table chloroplast. S1. B, Phenotypes and Imaging PAM images representing the Fv/Fm activity of 4- to 10-d–old Col-0, pp7l-1 and sig2 and sig6 mutant seedlings. C, Phenotypes of 3-week–old Col-0, pp7l-1, and sig mutant plants and Imaging PAM pictures showing maximum quantum yields of Lack of PP7L Impairs Chloroplast Translation PSII (Fv/Fm). Scale bars = 1 cm. D, Northern-blot analysis of SIG2 and and Development SIG6 transcripts in 4- and 5-d–old Col-0 and pp7l-1 plants. For the detection of SIG6 mRNA levels, the sig6 mutant was included as a To investigate whether the defect in photosyn- control, because several unspecific bands were detected with the SIG6 thetic activity found in the pp7l mutants was a conse- probe. Total RNAs were fractionated in a formaldehyde-containing quence of reduced accumulation of photosynthetic denaturing gel, transferred onto a nylon membrane, and probed with proteins, shot-gun proteomics was performed on total (a-32P)dCTP-labeled cDNA fragments specific for the different SIG protein extracts from 4-d–old wild-type and pp7l-1 seed- transcripts. The rRNA was visualized by staining the membrane with lings grown under LD conditions. In the pp7l-1 mutant, methylene blue (M.B.) and served as a loading control. E, Graph rep- resenting the changes (relative to Col-0 values) in SIG mRNA levels in pp7l-1, as determined by lncRNA-Seq analysis. Significant differences (DESeq2 analysis running with the fit type set to “parametric”; adjusted “parametric” and not “mean” fit settings in DESeq2. F, Venn diagrams P , 0.05) with respect to Col-0 are denoted by asterisks. The SD was not depicting the degree of overlap between the sets of genes whose ex- calculated, because differentially expressed genes were identified with pression levels were altered by at least twofold (up or down) in the pp7l‐ 1 mutant compared with the sig6 mutant. Plant Physiol. Vol. 180, 2019 327 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Role of PP7L in Chloroplast Development seen for the protein levels in the pp7l-1 mutant (Fig. 5, A and B; Supplemental Table S2), which was confirmed by RNA gel-blot analysis for representative instances (Fig. 5D). A scatter plot demonstrates that the relative changes in protein amounts are negatively correlated with those in the levels of the corresponding mRNAs (Fig. 5B). For example, psbD mRNA was nearly at wild- type levels (protein: ;10% of wild type) and rbcL transcripts were only slightly reduced if at all (RNA-Seq data: 100% wild type; RNA gel blot: 75% wild type) but amounts of the corresponding proteins were drastically reduced (;15% of wild type; Fig. 5, C and D). Thus, the reduced amounts of chloroplast-encoded proteins in the pp7l-1 mutant cannot be explained by a relative dearth of their transcripts. Closer inspection of Figs. 3D and 5D revealed, however, that amounts of chloroplast rRNAs were abnormally low in pp7l-1 seedlings, as indicated by methylene-blue staining of membranes. In the chloroplasts of land plants, the rRNAs are encoded in the rRNA operon, which contains the genes for tRNAs for Ile, Ala, and Arg, and for 16S, 23S, 4.5S, and 5S rRNAs (Supplemental Fig. S3A). This gene cluster is transcribed as a single molecule and processed via endonucleolytic cleavage and exonucleolytic trimming events (Strittmatter and Kössel, 1984). Indeed, RNA gel-blot analyses showed that in all pp7l alleles, levels of mature 23S rRNA were lower than in the wild type, whereas unprocessed precursors overaccumulated and 16S rRNA was unaffected (Supplemental Fig. S3B). This, together with the fact that amounts of chloroplast ribosomal proteins were reduced in pp7l-1 (Fig. 5A), suggests that pp7l mutants might experience a general reduction in chloroplast translation capacity—an in- ference that is compatible with the overall analysis of lncRNA-Seq data. To assess this possibility, the effects of the pp7l-1 mutation on the translational efficiency of chloroplast transcripts were further examined by ana- lyzing the association of psbD and rbcL mRNAs with polysomes. Seedling extracts were fractionated in Suc gradients under conditions that preserve polysome in- tegrity, and mRNAs were identified by hybridization with specific probes. In general, efficiently translated Figure 4. GO analysis of genes whose expression is reduced (A) or RNAs migrate deep into the gradient because almost all elevated (B) in pp7l-1 seedlings compared to Col-0 grown for 4 d in LD are associated with ribosomes. Methylene blue staining conditions. GO annotations for the cellular component, biological of rRNAs in wild type and pp7l-1 polysome gradients process, and molecular function categories were extracted from DAVID (Huang et al., 2009). GO terms with a .2-fold change and a Benjamini showed similar, though not identical distributions, in- corrected value of ,0.05 are shown. TF, transcription factor. dicating that there was no general difference in ribo- some distribution between mutant and wild-type plants. However, psbD and rbcL mRNA were shifted chloroplast-encoded subunits of PSI and PSII, the toward nonpolysomal fractions in pp7l-1 as compared chloroplast ATP synthase, the cytochrome b6 f complex, to the wild type (Fig. 5E), indicating that these mRNAs the large subunit of Rubisco, and also ribosomal pro- were less likely to be associated with polysomes in the teins accumulated to lower levels than that in wild type pp7l-1 mutant and that at least some of the mRNAs in (Fig. 5, A and B; Supplemental Table S3). Western-blot pp7l-1 chloroplasts are less engaged in translation. analyses confirmed this for representative subunits of The effects of the pp7l-1 mutation on the translation of the different photosynthesis complexes (Fig. 5C), and chloroplast transcripts were further investigated by also confirmed that protein levels were restored in oe- in vivo labeling, which allows one to quantify de novo PP7L pp7l-1 and -2 lines. In contrast, lncRNA-Seq in- protein synthesis. To this end, synthesis of plastid- dicated that most of the corresponding transcripts— encoded thylakoid membrane proteins was monitored except psbA—accumulated to higher levels than those by pulse labeling of wild-type, pp7l-1 mutant, and 328 Plant Physiol. Vol. 180, 2019 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Xu et al. oe-PP7L pp7l-1 seedlings in the presence of cyclohexi- mide, which inhibits the translation of nuclear-encoded proteins. After pulse labeling for 30 min, de novo syn- thesis of the D1 and D2 proteins was found to be strongly reduced in pp7l-1 relative to the wild type (Fig. 5F)—although psbD (encoding the D2 protein) transcript levels were notably high (Fig. 5D; Supplemental Table S2). To further investigate the role of PP7L in chloro- plast development, the ultrastructure of pp7l chlo- roplasts was investigated. Cotyledons from 4-d–old wild-type and pp7l-1 seedlings were fixed, embed- ded, sectioned, and examined for the morphology of chloroplasts by transmission electron microscopy (TEM). Thin sections of wild-type cotyledons revealed normally developed chloroplasts with their internal structures intact. Because the plant material was directly fixed after 8 h in the dark, wild-type chloroplasts did not show any starch granules and, moreover, the stroma and grana thylakoids were well structured (Fig. 6). The chloroplasts of the pp7l-1 mutant contained large starch granules and an ele- vated number of plastoglobuli. Moreover, the thy- lakoids in these chloroplasts were less well ordered, indicating a delay in chloroplast development in the pp7l-1 mutant. Together, these findings suggest that chloroplast transcription is not primarily affected in pp7l, but hint at a disturbance of subsequent processes in chloroplast gene expression that would account for the reduction in synthesis of chloroplast proteins and the delay in chloroplast development in the pp7l mutants. Photosynthesis Is Not Affected in Photomorphogenesis and Light Signaling Mutants In addition to driving photosynthesis, light is a major factor in controlling the expression of photosynthesis genes (Berry et al., 2013). The most prominent light- sensing proteins that mediate light-responsive gene Figure 5. Accumulation of chloroplast-encoded proteins is perturbed at steps downstream of chloroplast transcription in pp7l-1. A, Heat-map (compl.) seedlings, and aliquots (7 and 3.5 mg from Col-0; 7 mg from all showing relative protein (determined by shot-gun proteomics) and other lines) were fractionated in a formaldehyde-containing denaturing transcript levels (determined by lncRNA-Seq) in 4-d–old pp7l-1 seed- gel, transferred onto a nylon membrane, and probed with (a-32P)dCTP- lings compared to Col-0 grown in LD conditions. Only a selection of all labeled cDNA fragments specific for transcripts encoding individual data are shown. B, Scatter plot showing the anticorrelation between subunits of PSII (psbA, psbB, psbD) and rbcL. rRNA was visualized by changes in the levels of mRNAs and their protein products. The blue line staining the membrane with methylene blue (M.B.) and served as a indicates the theoretical 1:1 correlation. C, Immunoblot analysis of loading control. E, Chloroplast mRNAs accumulate in nonpolysomal representative thylakoid proteins. Total protein extracts from 4-d–old fractions in pp7l-1 plants. Whole-cell extracts from Col-0 and pp7l- wild-type (Col-0), pp7l-1, and oe-PP7L pp7l complemented (compl.) 1 plants were fractionated in linear 0.44–1.6 M (15% to 55%) Suc gra- seedlings were fractionated by SDS-PAGE, and blots were probed with dients by ultracentrifugation. Gradients were divided into 11 fractions, antibodies raised against individual subunits of photosynthetic com- and RNA was isolated from equal volumes. RNA blots were hybridized plexes. Decreasing levels of wild-type proteins were loaded in the lanes with [a-32P]dCTP-labeled cDNA fragments specific for psbD and rbcL. marked Col-0, 0.53 Col-0, and 0.253 Col-0 (13 Col = 5 mg). Actin and Ribosomal RNAs were stained with M.B. Arrows indicate the difference the Coomassie Brilliant Blue (C.B.B.)-stained membrane served as in 23S rRNA accumulation. F, In vivo pulse-labeling of thylakoid loading controls. Please note that RbcL protein levels are decreased in membrane proteins with [35S]Met in the presence of cycloheximide pp7l-1 seedlings. Thus they are also reduced on the C.B.B.-stained indicates that translation occurs at reduced rates in pp7l-1 chloroplasts. membrane. D, Steady-state levels of transcripts of photosynthetic genes. Proteins were resolved by SDS-PAGE after pulse-labeling for 30 min, Total RNA was isolated from 4-d–old Col-0, pp7l-1, and oe-PP7L pp7l and visualized by autoradiography. The C.B.B.-stained membrane served as a loading control. Plant Physiol. Vol. 180, 2019 329 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Role of PP7L in Chloroplast Development However, to our knowledge, these measurements have not been extensively performed on cry, phy, and cop1 mutants grown under normal light conditions, and Fv/Fm values can only be extracted from the controls used in stress studies of adult plants (Kleine et al., 2007; Rusaczonek et al., 2015). Thus, we deter- mined Fv/Fm values in 4-week–old wild-type, cop1-4, cry1-304 cry2-1, phyA-211 phyB-9, and—as a control— pp7l-1 mutant plants grown under LD conditions. As expected, Fv/Fm was reduced in the younger leaves of pp7l-1 plants, but none of the other mutants showed any decrease in the maximum quantum yield of PSII (Supplemental Fig. S4). Promoters of Genes Deregulated in pp7l-1 Contain PIF-Related Cis-Acting Elements To study the transcriptional networks activated by lack of PP7L, the lncRNA-Seq data were investigated in more detail. Cis-acting elements that are overrep- resented in promoters of differentially regulated genes in the pp7l-1 mutant were identified by scanning 1,000-basepair (bp) promoter sequences for 8-bp cis- elements with the expectation-maximization–based program Amadeus (Linhart et al., 2008). Because a very large set of genes is differentially regulated in the pp7l-1 mutant, we focused on genes with more than 4- fold up- or downregulation (363 down; 188 up). The putative cis-elements thus identified were used to search the JASPAR plant database (http://jaspar. genereg.net/; Khan et al., 2018) for Arabidopsis transcription factors known to bind to similar se- quences. The highest similarity between queried mo- tifs is defined by the lowest E-value. The motifs with Figure 6. Chloroplast development is disturbed in the pp7l-1 mutant. the lowest E-value found by JASPAR for both the TEM images of ultrathin sections of 4-d–old wild-type and pp7l-1 cot- promoters of up- and downregulated gene sets con- yledons. The wild type shows chloroplasts with a well-developed thy- tained the G-box motif CACGTG (Fig. 7B). The G-box lakoid system and grana stacks. Chloroplasts in the pp7l-1 mutant motif is highly conserved and is bound by bHLH and contain large starch grains, although the seedlings were stored in bZIP transcription factors in organisms ranging from darkness before the cotyledons were thin-sectioned. Moreover, pp7l- yeasts to humans. In plants, these two transcription 1 chloroplasts contain more plastoglobuli and the thylakoid system is less developed. Bar = 1 mm. C, chloroplast; G, grana; P, plastoglobule; factor families have expanded massively; e.g. in S, starch grain. Arabidopsis, .100 bHLH transcription factors have been described (Carretero-Paulet et al., 2010). Se- quences flanking the G-boxes define the specificity of expression are phys and cry1 and cry2. Crys mediate protein binding (Williams et al., 1992) and enable one blue-light induction of nuclear genes that encode chlo- to predict the in vitro binding of bZIP homodimers roplast photosynthesis proteins or components of the (Ezer et al., 2017). In addition, a number of studies chloroplast transcriptional apparatus (reviewed in Yu have experimentally determined the binding motifs of et al., 2010). Diverse developmental and physiological several transcription factors (Becker et al., 2017). In processes are controlled by phyA and phyB, including fact, among the downregulated gene set, binding sites Chl synthesis (McCormac and Terry, 2002). In fact, 4- for the phytochrome-interacting factors PIF4 and d–old pp7l seedlings contain less Chl than do wild- PIF5, as well as the bHLH transcription factor MYC3, type seedlings (Fig. 7A). Moreover, photoactivated were identified by JASPAR, whereas interaction sites crys (Yu et al., 2010) and phys (Li et al., 2011) repress for PIF1, PIF3, and the abscisic-acid response element the COP1-dependent degradation of transcription binding factor 1 were found in the upregulated gene factors that control the expression of light-responsive set (Fig. 7B). genes, and Chl content is reduced in cop1-4 seedlings To obtain more information on the putative function just as in pp7l-1 seedlings (Fig. 7A). Fv/Fm measure- of PP7L, the list of genes that are differentially ments are a simple read-out for PSII function. expressed in the pp7l-1 mutant was submitted to the 330 Plant Physiol. Vol. 180, 2019 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Xu et al. Figure 7. Chl content of 4-d–old pp7l-1 and cop1-4 seedlings, and identification of cis-elements in pp7l-1 deregulated genes and conditions that have a similar effect on the transcriptome to those seen in the pp7l-1 mutant. A, Determination of total Chl (Chl a + b) content. Note that Chl is significantly reduced in 4-d–old pp7l-1 seedlings, in contrast to its essentially wild-type content in 4-week–old pp7l-1 plants (see Supplemental Fig. S2B). Pigments were acetone-extracted, measured spectrophotometrically, and concentrations were determined as described in Jeffrey and Humphrey (1975). Data are shown as mean values 6 SD from three different plant pools. Each pool contained .20 plants. The significant difference (Tukey’s test; P , 0.05) between the mutants and Col-0 was identified by Tukey’s test, and is denoted by an asterisk. B, Sequence logos of the most signifi- cantly identified cis-elements of genes whose expression was reduced or elevated in the pp7l- 1 mutant compared to Col-0. The names of the putative transcription factors binding to the iden- tified cis-elements, together with their respective logos, are also shown. C, Identification of pertur- bations that induce similar changes in gene ex- pression to those identified in the pp7l-1 mutant. The gene expression changes of the 200 most strongly up- or downregulated genes in 4-d–old pp7l-1 seedlings were submitted to the Geneves- tigator Signature Tool (Hruz et al., 2008). Mutants or conditions that triggered the most similar gene expression profiles were identified by Pearson distance clustering. Note that the higher the relative similarity parameter, the higher the similarity to the pp7l-1 signature. D, Phenotypes of 4-week–old Col-0, pp7l-1, and the various csn and pif mutant plants. In the bottom row, Imaging PAM pictures depict the maximum quantum yields of PSII (Fv/Fm). Scale bars = 1 cm. Genevestigator database (Hruz et al., 2008; https:// mutant—displayed a growth phenotype similar to that www.genevestigator.com/gv/) to identify mutants of pp7l-1, none of them exhibited any change (relative to and conditions associated with similar transcriptome that in the wild type) in the Fv/Fm value (Fig. 7D). profiles. The most similar transcriptomic responses were found in the csn5a-2 csn5b mutant (Dohmann et al., 2008), in which the function of the CSN multi- phyB mRNA and Protein Levels Are Elevated in pp7l-1 protein complex is compromised, and the pif1 pif3 pif4 pif5 quadruple mutant (Leivar et al., 2008; Fig. 7C). Inspection of the lncRNA-Seq data revealed that These results prompted us to test whether the Fv/Fm transcripts coding for several phy/PIF signaling- was reduced in the emerging leaves of 4-week–old associated proteins like pTAC12 (Chen et al., 2010), csn5a-1, csn5a-2, csn5b-1, pif1 pif5, pif4 pif5, pif1 pif4 GOLDEN2-LIKE1 (Martín et al., 2016), and NUCLEO- pif5, and pif1 pif3 pif4 pif5 quadruple mutants, as it is in SIDE DIPHOSPHATE KINASE2 (Shen et al., 2005; the pp7l mutants. The csn5a csn5b double mutant was Supplemental Table S2) accumulate to higher levels in not included, because it ceases growing at earlier stages pp7l-1 mutants. Moreover, PHYB mRNA levels were of development (Dohmann et al., 2005). However, twofold-increased (Supplemental Table S2), a finding although all mutants—with the exception of the csn5b-1 that was confirmed for all three pp7l alleles by RT-qPCR Plant Physiol. Vol. 180, 2019 331 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Role of PP7L in Chloroplast Development (Supplemental Fig. S5A). To test whether the elevated PP7L Is Involved in the Tolerance to Salt and PHYB mRNA levels were reflected in corresponding High-Light Stress changes in phyB protein levels, immunoblot analysis Perturbations of organellar gene expression fre- using antiphyB antibodies was performed on total quently result in the alteration of acclimation and tol- protein extracts isolated from 4-d–old seedlings grown erance responses to adverse growth conditions (Leister under LD conditions. Indeed, in three independent et al., 2017), which include the tolerance to salt (Lee experiments, the phyB protein level was found to be et al., 2014) and high-light stress (Coll et al., 2009). To ;2.5-fold elevated in the pp7l-1 mutant, but was similar investigate whether PP7L might play a role in the re- to that in wild type in oe-PP7L pp7l-1 (Supplemental Fig. sponse to the aforementioned stresses, seeds of wild S5B). Therefore, we asked whether the pp7l Fv/Fm type, pp7l mutants, and PP7L overexpression lines (oe- phenotype might be a consequence of the over- PP7L; Supplemental Fig. S6) were germinated on Murashige and Skoog (MS) medium supplemented expression of phyB. For this purpose, a line was used in with 100 mM NaCl and on MS medium without sup- which phyB was overexpressed under the control of the plement as control (Fig. 8A). Additionally, seeds were 35S promoter in the No-0 background (designated germinated under high light (1,000 mmol photons m22 “ABO”; Wagner et al., 1991). However, the maximum s21; Fig. 8C), and germination rates were followed from quantum yield of PSII was not reduced in 3- and 4- d 2 to d 7. After 2 d on control MS medium under d–old ABO seedlings (Supplemental Fig. S5C). growth light, 60% of wild-type and oe-PP7L line seeds We therefore conclude that the reduced photosyn- germinated (Fig. 8A). Germination rates of pp7l mu- thetic capacity of the pp7l mutants is not caused (solely) tants (48%) were slightly delayed, but after 3 d all lines by their abnormally high levels of phyB. displayed an ;100% germination rate. The delay in Figure 8. PP7L affects salt and high-light toler- ance. A, Seed germination of wild type (Col-0), pp7l mutants and oe-PP7L lines on MS medium without NaCl (“MS”), and on MS supplemented with 100-mM NaCl. The data represent mean values 6 SD of three independent experiments, each performed with 100 seeds per treatment and genotype. B, Phenotypes of wild type (Col-0), pp7l mutants, and oe-PP7L lines grown for 2 weeks on MS supplemented with 100 mM NaCl. Scale bar = 1 cm. C, Germination of wild-type (Col-0), pp7l mutants, and oe-PP7L line seeds grown un- der high light (1,000 mmol m22 s21). The data represent mean values 6 SD of three independent experiments, each performed with 100 seeds per treatment and genotype. D, Graph displaying Fv/Fm values of 4-week–old (Col-0), pp7l mutants, and oe-PP7L lines treated for the indicated time with high light (1,000 mmol m22 s21). The data represent mean values 6 SD of three independent experiments. E, Imaging PAM pictures showing the maximum quantum yields of PSII (Fv/Fm) of 4- week–old plants subjected to 1 h of high light (1,000 mmol m22 s21). Scale bar = 1 cm. 332 Plant Physiol. Vol. 180, 2019 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Xu et al. germination of pp7l mutants was pronounced when seeds were challenged with 100 mM NaCl: Only 2% to 5% of pp7l mutants germinated after 2 d, whereas oe- PP7L lines and wild type displayed ;55% germination (Fig. 8A). Although after 4 d all lines displayed nearly 100% germination, the enhanced sensitivity of the pp7l mutants to NaCl was manifested after 2 weeks; the growth of pp7l seedlings was substantially delayed and cotyledons appeared bleached (Fig. 8B). When seeds were germinated under high light, the germination rates of pp7l mutants were lower than those of the wild-type and the oe-PP7L lines during the whole time course (Fig. 8C). The oe-PP7L-7 line germi- nated even slightly faster than the wild type. To Figure 9. Presence of PP7L in Brassicales. The phylogenetic tree for a investigate the high-light phenotype in adult plants, part of the Brassicales species is based on the information from 4-week–old plants were shifted to high light and the Wikström et al. (2001) and Edger et al. (2018). Arrows indicate esti- maximum quantum yield of PSII was measured after 1, mated dates of divergence (in MYA) at different nodes; open circle, PP7L 2, and 4 h (Fig. 8, D and E). In the wild type, high light absent; black circle, PP7L present. caused a drop of the Fv/Fm value from 0.81 to 0.56, 0.48, and 0.45 after 1, 2, and 4 h of treatment, respectively. The pp7l mutant displayed photoinhibition that was papaya, which belongs to the Caricaceae family. The slightly enhanced compared to that in the wild type. Brassicaceae and Caricaceae diverged from a common However, at 0.67, the Fv/Fm value of the pp7l mutants ancestor ;72–92 million years ago (MYA; Wikström was already significantly lower in normal growth con- et al., 2001; Edger et al., 2018). In light of the absence ditions (Fig. 8, D and E). Importantly, overexpression of of PP7L in C. papaya, and its presence in T. hassleriana, it PP7L rendered plants more high-light–tolerant with can be concluded that PP7L evolved after the diver- higher Fv/Fm values under high-light treatment com- gence of Caricaceae from the other Brassicales families, pared to that of the wild type. but before the separation of Cleomaceae from Brassi- These results show that PP7L plays a positive role in caceae, i.e. between 43 and 92 MYA. Therefore, regu- seed germination especially under adverse growth lation of chloroplast and plant development in conditions and that overexpression of PP7L renders Brassicales differs, at least with respect to the elements plants more high-light–tolerant. that are dependent on PP7L, from that in species descended from other plant orders. PP7L Evolved 43–92 Million Years Ago in the Lineage Leading to Brassicales DISCUSSION The type-7 subfamily of PPPs is specific to plants Lack of PP7L leads to pleiotropic phenotypic effects, (Farkas et al., 2007; Uhrig et al., 2013). The increasing which include a small rosette size, increased anthocy- number of plant genomes sequenced in recent years anin levels (Fig. 1), susceptibility to salt and high-light (see for example: http://plants.ensembl.org) allowed stress (Fig. 8) and perturbed chloroplast development us to refine the analysis of the evolution of the type-7 (Fig. 6), accompanied by a general reduction in the ex- subfamily. To this end, type-7 members were identified pression of proteins synthesized in the chloroplast by BLAST searches in the PLAZA database (https:// (Fig. 5). Using the maximum quantum yield of PSII (Fv/ bioinformatics.psb.ugent.be/plaza/; Van Bel et al., Fm) as a proxy for chloroplast development, we have 2018) and PP7 was found to be ubiquitously present shown that chloroplast development is disturbed in in the plant families included in the database. PP7L and cotyledons (Fig. 3), as well as in emerging true leaves long PP7, however, were only present in Arabidopsis (Fig. 1) in pp7l mutants, a finding that was further cor- lyrata, Arabidopsis, Brassica oleracea, Brassica rapa, Cap- roborated by ultrastructural analysis of cotyledons of 4- sella rubella, Schrenkiella parvula, and Tarenaya hassleriana d–old seedlings (Fig. 6). (Fig. 9). These plant species are members of the Bras- sicaceae or Cleomaceae (T. hassleriana) families, which PP7L Acts as a Positive Regulator of Chloroplast are grouped in the order of Brassicales. To expand this Development in Seedlings and in Emerging True Leaves analysis to more Brassicales species and to species be- yond the Brassicales, the NCBI Blastp suite (https:// The different cytological origins of chloroplasts in blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE=Proteins) was cotyledons and leaves (Waters and Langdale, 2009; used to identify further sequences homologous to the Jarvis and López-Juez, 2013; Van Dingenen et al., 2016), PP7L protein sequence. In fact, PP7L was also detected together with differences between the programs of gene in Camelina sativa and Aethionema arabicum, both of expression triggered by light in the cotyledons and the which belong to the Brassicaceae, but not in Carica shoot apex (López-Juez et al., 2008), suggest that Plant Physiol. Vol. 180, 2019 333 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Role of PP7L in Chloroplast Development chloroplasts in cotyledons develop through mecha- classified as being specific for chloroplast development nisms that are at least partly distinct from those in in cotyledons or true leaves are less specific than was leaves. Moreover, various mutants have been isolated once thought. More sensitive phenotyping methods in which the greening of cotyledons and leaves are like Imaging PAM analysis (this study) or the applica- differentially affected (Waters and Langdale, 2009). tion of diverse growth conditions (as in the case of sco2; Thus, chloroplast development mutants have been de- Zagari et al., 2017) will be instrumental in the future scribed that exhibit variegated phenotypes in true analysis of mutants that appear to be specifically af- leaves, but develop normal green cotyledons. Among fected at certain developmental stages. these, the best-characterized chloroplast biogenesis mutants are those designated as immutans mutants and variegated (var; Putarjunan et al., 2013). The genes The Role of PP7L in Chloroplast Development Is Not IMMUTANS and VAR code for the plastid terminal Mediated by the Principal Photomorphogenesis Pathways oxidase and the AtFtsH2 subunit of the thylakoid FtsH metalloprotease complex, respectively. The FtsH It has been noted before that light and chloroplast complex is mainly involved in turnover of the PSII signaling pathways converge at some point (Ruckle reaction-center D1 protein. Conversely, the three et al., 2007; Leister et al., 2014; Martín et al., 2016). In snowy cotyledon (sco) mutants, namely sco1 (Albrecht agreement with this, we have identified PIF-related cis- et al., 2006), sco2/cyo1 (Shimada et al., 2007; Albrecht elements in the promoters of pp7l-deregulated genes et al., 2008), and sco3 (Albrecht et al., 2010), have white (Fig. 7B), and the pp7l gene expression signature re- or pale green cotyledons but green leaves. The SCO sembles the changes in gene expression induced by csn class appears not to have a common specific function and pif mutations (Fig. 7C). Moreover, in pp7l, transcripts that causes the cotyledon-specific mutant phenotypes, coding for several phy/PIF signaling-associated proteins because its members include proteins involved in dif- (Supplemental Table S2) accumulate to a greater extent ferent processes, such as plastid protein translation than that in wild type, and phyB expression is itself (SCO1; Albrecht et al., 2006), folding of Cys-rich thy- elevated at the transcript (Supplemental Fig. S5A; lakoid proteins (SCO2/CYO1; Shimada et al., 2007), or Supplemental Table S2) and protein level (Supplemental association with microtubules and the peroxisome Fig. S5B). In seedlings, chloroplasts develop from etio- (SCO3; Albrecht et al., 2010). Seeds of many oilseed plasts that develop from proplastids/eoplasts in plants like Arabidopsis are green due to the presence of darkness (Pribil et al., 2014). Upon exposure to light, photosynthetically active chloroplasts during embryo- photoactivated phys inhibit PIFs and the proteasome, genesis, which dedifferentiate into nonphotosynthetic thus inducing transcriptional reprogramming, which eoplasts during the desiccation phase (Liebers et al., in turn promotes photomorphogenesis, including 2017). Immature green sco embryos dissected from si- greening (Xu et al., 2015). Consequently, Chl content liques are green (Ruppel and Hangarter, 2007; Albrecht is reduced in phyB and phyA phyB seedlings (Neff and et al., 2008, 2010), and precocious germination of sco2 Chory, 1998), and Chl content is also reduced in cop1- and sco3 mutants rescues their sco cotyledon phenotype 4 seedlings and pp7l-1 seedlings (Fig. 7A). Therefore, (Albrecht et al., 2008, 2010). Thus, it was suggested that we tested the possibility that PP7L acts through the the defects are specific to the development of chloro- phy/PIF or proteasome pathway. However, none of plasts from proplastids/eoplasts in cotyledons, and the tested mutants lacking the main PIFs, photore- SCO proteins were assumed to be cotyledon-specific ceptors, or the COP and CSN proteins associated with factors (Waters and Langdale, 2009). However, under photomorphogenesis show any change in Fv/Fm short-day conditions, smaller and pale-green true (Supplemental Fig. S4). In addition, overexpression leaves can be observed in sco2 mutants, and disruption of phyB does not result in a reduction of photosyn- of SCO2 in Lotus japonicus results not only in paler thesis (Supplemental Fig. S5). This indicates that (1) cotyledons but also in variegated true leaves (Zagari PP7L operates through a different pathway to regu- et al., 2017). late chloroplast development, and (2) that greening Two of the six nuclear-encoded SIGs that interact and photosystem biogenesis are controlled by sepa- with PEP have also been proposed to have a specific rate systems. function in the development of cotyledon chloroplasts Photosynthetic activity in pp7l mutants during (Privat et al., 2003; Ishizaki et al., 2005). Although the seedling development is most strongly affected in cotyledons of SIG2 antisense plants are Chl-deficient, young seedlings (Fig. 3A; Supplemental Table S1), re- the true leaves are green (Privat et al., 2003); similarly, covers gradually during subsequent seedling develop- sig6-1 null mutants exhibit delayed greening of cotyle- ment, and is essentially normal after 10 d (Fig. 3B). dons, but the true leaves were described as normal Recently, the linear developmental program of thyla- (Ishizaki et al., 2005). Under our growth conditions, the koid biogenesis in cotyledons was documented by true leaves of sig6-1 also appeared normal in color, but a TEM, RNA-Seq, and photosynthesis protein level more detailed analysis of photosynthesis by Imaging analysis (Liang et al., 2018). When seeds are germinated pulse amplitude modulation (PAM) revealed that under continuous light, thylakoid biogenesis begins chloroplast development was perturbed (Fig. 3C). In between 24 and 36 h after imbibition (HAI), and PSII consequence, some proteins that were previously complexes are first inserted into the thylakoid 334 Plant Physiol. Vol. 180, 2019 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Xu et al. membrane at 36 HAI, when polysomes attach to the mutation in FUG1, which encodes a chloroplast protein flattening thylakoids (Liang et al., 2018). Cotransla- with homology to cpIF2, is introduced into var2, the tional insertion of PSII by thylakoid-membrane-bound high-light sensitivity of var2 is suppressed—pointing to polysomes further enriches PSII-light-harvesting com- an importance of the balance between chloroplast plex II at 60 HAI (2.5 d; Liang et al., 2018). Protein protein synthesis and degradation in stress acclimation subunits of the PSI complex are detected only after 84 (Miura et al., 2007). HAI, suggesting that PSI assembly is initiated at ;84 HAI, corresponding to 3.5 d and thus matching our results in which full photosynthetic activity is reached How Does the Nucleocytosolic Compartment Influence between days 3 and 4. Importantly, the results of Liang Chloroplast Development? et al. (2018) suggest that thylakoid-bound polysomes mediate critical stages of chloroplast biogenesis. In- Chloroplast proteins are synthesized both in the deed, this is supported by the fact that a disturbance in nucleocytosolic compartment and the organelle, which chloroplast ribosome assembly in the rh3-4 mutant, in necessitates coordination of the expression of nuclear which the DEAD-box RNA helicase RH3 is defective and chloroplast genomes. Additionally, gene expres- (Asakura et al., 2012), causes a delay in thylakoid de- sion in the chloroplast itself is a complex process that velopment (Liang et al., 2018). In pp7l, fewer mature depends on the supply of a plethora of nuclear-encoded chloroplast 23S rRNA molecules (Figs. 3D and 5D; proteins. Primary transcripts are produced by PEP and Supplemental Fig. S3) and smaller amounts of ribo- nuclear-encoded phage-type RNA polymerase (Liere somal proteins accumulate (Fig. 5A; Supplemental et al., 2011). Subsequently, these RNAs undergo pro- Table S3). Thus, we hypothesize that the delay in thy- cessing, splicing, editing, turnover, and maturation lakoid development in the pp7l mutant (Fig. 6) is caused (Stern et al., 2010; Stoppel and Meurer, 2012) before by a perturbation of rRNA maturation and accumula- they are translated (Tiller and Bock, 2014). Lack of tion of chloroplast ribosomal proteins, which results in nuclear-encoded proteins at early stages of chloroplast a reduction in the numbers of functional ribosomes and development impacts on phenotypes in various de- a decrease in protein synthesis in the chloroplast (Fig. 5, grees, ranging from no visible phenotype to pale green E and F). cotyledons and rosette leaves to seedling- and embryo- lethality (Barkan and Small, 2014; Kleine and Leister, 2015). In screens for mutants with altered chloroplast PP7L Contributes to the Tolerance to Abiotic Stresses biology at this stage, most studies have identified chloroplast-localized proteins involved in chloroplast Plants as sessile organisms are constantly exposed to development and showing a photosynthesis pheno- environmental changes and stresses, and they conse- type. Exceptions include PALE GREEN1 (Hsieh et al., quently evolved mechanisms to cope with different 2017) and SCO3 (Albrecht et al., 2010), which are lo- stresses. Because (1) pp7l mutants are more susceptible calized to mitochondria and peroxisomes, respectively. to salt and high-light treatment and (2) overexpression Interestingly, the cytoskeleton is altered in sco3-1, and of PP7L enhances the capacity of plants to tolerate high microtubule inhibitors have similar effects on chloro- light intensities (see Fig. 8), PP7L is clearly involved in plast biogenesis to sco3-1 (Albrecht et al., 2010), sug- salt and high-light tolerance. The reasons for the mul- gesting a role for the cytoskeleton and peroxisomes in tiple stress susceptibility of pp7l mutants is not plainly chloroplast biogenesis. Thus, it is increasingly clear evident, but one plausible explanation is that it can be that, in addition to the already known pathways, the ascribed at least in part to the perturbations of chloro- development of chloroplasts requires a number of other plast gene expression observed in pp7l mutants (see processes. Fig. 5). In fact, it was already noted before that changes The import of photosynthetic preproteins is medi- in chloroplast gene expression can trigger alterations in ated by transclocon at the outer envelope membrane of tolerance responses, presumably via retrograde sig- chloroplasts (Toc)33 and Toc159 proteins. Accordingly, naling (Leister et al., 2017). Interestingly, an impairment the plastid protein import1 (ppi1; toc33) and ppi2 (toc159) in chloroplast gene expression results not automatically mutants are defective in chloroplast biogenesis (Bauer in stress susceptibility, but can render plants even more et al., 2000; Kubis et al., 2003). Moreover, not only the resistant to stresses. This is exemplified by the “happy process of protein uptake by the chloroplast, but pre- on norflurazon” mutants hon5 and hon23 affected in the import steps such as quality control by the cytosolic ClpR4 subunit of the chloroplast-localized Clp protease ubiquitin–proteasome system (Ling et al., 2012) and the complex, and a protein with homology to prokaryotic establishment of the appropriate balance between translation initiation factor2, cpIF2, respectively; they chloroplast and cytosolic translation (Wang et al., 2018), display a constitutive stress response that enhances affect chloroplast development. resistance against a combined low-temperature/high- Notably, alterations in chloroplast protein import in light stress (Saini et al., 2011). Moreover, lack of the the ppi2 and translocon at the inner chloroplast envelope 56- major chloroplast FtsH complex component FtsH2 in 1 mutants result in decreased accumulation of plastid the var2 mutant leads to a reduced capacity to acclimate ribosomal proteins (Köhler et al., 2016) as in the pp7l- to high-light stress (Miura et al., 2007). When a 1 mutant (Fig. 5A; Supplemental Table S3). However, Plant Physiol. Vol. 180, 2019 335 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Role of PP7L in Chloroplast Development reduced ribosomal protein accumulation is not a fea- Anthocyanin accumulation is also induced by abiotic ture common to all albino/pale green (apg) mutants, be- stresses like high light (Kleine et al., 2007), cold (Schulz cause ribosomal subunits accumulate to wild-type et al., 2015), or specifically by Suc (Solfanelli et al., 2006). levels in the apg1, apg2, and apg3 mutants, which are Photosynthesis-derived Suc is exported from chloro- defective in proteins that participate in diverse chloro- plasts to other cellular compartments or to non- plast functions (Motohashi et al., 2012). Because Toc159 photosynthetic sink tissues, and carbon supply can be is an import receptor for photosynthesis-related pro- sensed through sugar signaling (Wingler, 2018). De- teins, a straightforward explanation for the ppi2 pho- spite the obvious linkage of photosynthesis and energy tosynthesis phenotype is that it is caused by a defect in metabolism and signaling, little is known about the the import of the photosynthesis proteins. However, connections between cytoplasmic and nucleus- the mechanism must be more complex, because in the localized key regulators of cellular metabolism like ppi2 mutant many nuclear-encoded photosynthesis the SNF1-related protein kinase1 and organelles proteins are downregulated at the transcript level, (Wurzinger et al., 2018). Moreover, mRNA levels of which points to the activation of retrograde signaling many photosynthetic genes are dependent on SNF1- (Bischof et al., 2011). This is not the case—at least for the related protein kinase1 activity (Zhang et al., 2009). vast majority of mRNAs for photosynthesis proteins— One possible PP7L function would be that PP7L is at the in the pp7l-1 mutant (Supplemental Table S2) nor in the receiver site of sugar signaling in the nucleus. The sco2/cyo1 mutant (Shimada et al., 2007). Why mutants perturbed starch mobilization (Fig. 6) and the small with similar molecular chloroplast phenotypes do not growth phenotypes of the pp7l mutant (Fig. 1) would be trigger similar gene expression profiles in the nucleus in accordance with sugar-signaling regulators playing a remains elusive. central role in the regulation of starch metabolism (Sakr It is interesting to note that, although the mRNA et al., 2018; Wurzinger et al., 2018). With its predicted levels of PHYB, SIG2, SIG6, GLK1, and Chl biosynthesis function as an inactive Ser/Thr phosphatase, it is un- genes are upregulated in pp7l, greening in pp7l is de- likely that it is a transcription factor and future studies fective (Fig. 7A), questioning whether the elevated cy- will have to clarify the molecular function of PP7L. tosolic mRNA levels reflect functional protein levels. The Brassicaceae Arabidopsis is a member of the or- Among the elevated nuclear-encoded transcripts, the der Brassicales, which is known for the production of GO category “chloroplast rRNA processing” is, in ad- mustard oil (glucosinate) compounds (Cardinal- dition to “chlorophyll biosynthesis,” the most highly McTeague et al., 2016). This order contains, in addi- enriched term (Fig. 4). Processing of 23S rRNA is de- tion to several model plants, many species that are of fective in the pp7l mutant (Supplemental Fig. S3), which economical relevance, such as mustard (Brassica juncea), might be a secondary effect of reduced chloroplast broccoli (Brassica oleracea var. italica), cabbage (Brassica protein translation. Because chloroplast rRNA matu- oleracea var. capitata), kale (Brassica oleracea var. sabell- ration requires a plethora of nuclear-encoded proteins ica), rapeseed (Brassica napus), and caper (Capparis spi- (Stoppel and Meurer, 2012), this finding could also nosa). Because PP7L evolved specifically in the order support the concept of negative correlation of mRNA Brassicales (Fig. 9) and because of the prominent phe- and protein levels. Indeed, an increasing number of notype of the pp7l mutant, it appears that PP7L is spe- observations indicate that mRNA transcript levels and cific for the regulation of chloroplast and plant protein synthesis are often correlated only to a limited development in Brassicales. Other plant orders might degree (e.g. Oelze et al., 2014). Before a protein destined (1) have recruited a protein different from PP7L or (2) for the chloroplast reaches the quality-control and im- might differ in their chloroplast and plant development port stages, it must first undergo a series of complex and therefore not require PP7L. Thus, further investi- processes. The expression level of its mRNA is itself gation of PP7L should help to elucidate the aspects of determined by the relative rates of transcription and chloroplast and plant development specific to this or- degradation (Roy and von Arnim, 2013), and the der, which are of potential significance for the im- decoding of mRNA by the ribosome involves a plethora provement of agronomically valuable crops. of initiation, elongation, and termination factors (Moore et al., 2016). Once the protein is finally made, it may also require posttranslational modification or be degraded by the proteasome (Arsova et al., 2018). This MATERIALS AND METHODS shows that many steps from primary transcript to Plant Material and Growth Conditions protein translation and modification can account for a The pp7l:En-1 mutant (ZIGIA line V2-880) was identified in the mutant negative correlation between mRNA and protein a- collection of Wisman et al. (1998) based on alterations in the effective quantum bundance. The remaining key questions are which yield of PSII (FII). The mutants pp7l-1 (SALK_018295), pp7l-2 (SALK_033071), nucleo(cytosolic) process is exactly disturbed in the pp7l and pp7l-3 (SALK_022053) were identified in the SIGnAL database (Alonso mutant provoking the delayed chloroplast develop- et al., 2003). The sig2-2 (SALK_045706) mutant allele has been described pre- ment, and which transcript changes in pp7l are attrib- viously (Kanamaru et al., 2001), as have the mutants sig5-1 (SALK_049021; (Tsunoyama et al., 2004), sig6-1 (SAIL_893_C09; Woodson et al., 2013), cop1-4 utable to retrograde signaling activated by the delay in (McNellis et al., 1994), cry1-304 cry2-1 (Mockler et al., 1999), csn5a-1, csn5a-2, chloroplast development. Of note is the 8-fold-higher csn5b-1 (Dohmann et al., 2005), pif1pif5, pif4pif5, pif1pif4pif5, pif1pif3pif4pif5 anthocyanin content in the pp7l mutant (Fig. 1B). (Pfeiffer et al., 2014), phyB-9 and phyA-211 phyB-9 (Su et al., 2015), and ABO 336 Plant Physiol. Vol. 180, 2019 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Xu et al. (Shinomura et al., 1998). All mutants except of ABO are in the Col-0 Polysome Analysis background. Arabidopsis (Arabidopsis thaliana) plants were routinely grown on potting Polysome isolation was performed according to Barkan (1993). Seedlings soil (Stender) under controlled greenhouse conditions and on a 16-/8-h light/ (200 mg) were ground in liquid N using a mortar and pestle. Subsequently, the dark cycle (daylight was supplemented with illumination from HQI Powerstar microsomal membranes were extracted and solubilized with 0.5% (w/v) 400W/NDL, providing a total fluence of ;180 mmol photons m22 s21 on leaf sodium deoxycholate. The solubilized material was then layered onto 15/55% surfaces). Wuxal Super fertilizer (8% N, 8% P2O5, and 6% K2O; MANNA) was (w/v) Suc step-gradients (corresponding to 0.44/1.6 M) and centrifuged at used according to the manufacturer’s instructions. Where indicated, seedlings 250,000 g for 65 min at 4°C. The step gradient was fractionated, and the mRNA were grown on 0.8% (w/v) agar (Sigma-Aldrich) containing 1% (w/v) Suc at associated with polysomes was extracted with phenol/chloroform/isoamyl 22°C under 100 mmol photons m22 s21 provided by white fluorescent lamps. alcohol (25:24:1), precipitated at room temperature with 95% (v/v) ethanol, and subjected to RNA gel-blot analysis. Complementation of pp7l Mutants, Expression, and In Vivo Labeling of Thylakoid Proteins Intracellular Localization of eGFP Fusions Four-d–old seedlings were preincubated in 1 mL of labeling solution con- To complement the mutant phenotype, the sequence of PP7L (AT5G10900) taining 20 mg/mL cycloheximide, 10 mM ris(hydroxymethyl)aminomethane without the stop codon was amplified from genomic DNA with the primers (TRIS), 5 mM MgCl2, 20 mM KCl (pH 6.8), and 0.1% (v/v) TWEENÒ20 (a indicated in Supplemental Table S4. This PCR product was introduced into the polyethylene glycol sorbitan monolaurate) under illumination at a fluence rate donor vector pDONR207 by the BP clonase reaction (Invitrogen) as described in of 80 mmol photons m22$s21 for 30 min to block cytosolic translation. Then the Gateway manual. After sequence analysis of the recombined DNA se- labeling was performed in the same solution containing 1 mCi mL21 [35S]Met quence, the attL recombination clonase reaction (Invitrogen) was performed to under the same lighting conditions for 30 min. After labeling, the thylakoid introduce the fusion construct into the destination vector pB7FWG2 (35S pro- membranes were isolated and the proteins were fractionated on 10% sodium moter, eGFP), generating 35S:PP7L-GFP. Subsequently, the construct was in- dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and analyzed troduced into Col-0, pp7l-1, pp7l-2, and pp7l-3 backgrounds. Transformed by fluorography (Variable Mode Imager Typhoon TRIO; Amersham seedlings were selected with the herbicide Basta. For transformations into pp7l Biosciences). mutants, plants showing a wild-type phenotype were selected. Then expression of PP7L mRNA levels in different genotypes was analyzed by RT-PCR using an Protein Isolation and Western Blotting appropriate primer combination (Supplemental Table S4). Subcellular localization analysis of fused proteins was conducted in proto- Seedling samples (100 mg) were ground in liquid N and total proteins were plasts isolated from wild type. Protoplasts were imaged with a Fluorescence solubilized in protein extraction buffer (0.125 M TRIS, 1% SDS, 10% [v/v] Axio Imager microscope (Zeiss) after transformation as described in Dovzhenko glycerol, 0.05 M sodium metabisulfite) containing 1 mM phenylmethylsulfonyl et al. (2003). Fluorescence was excited with the X-Cite Series 120 fluorescence fluoride and protease inhibitor cocktail (Sigma-Aldrich). Cell debris was re- lamp (EXFO) and images were collected at 500–550 nm (eGFP fluorescence) and moved by centrifugation for 10 min and the supernatant was boiled with SDS 670–750 nm (Chl autofluorescence). loading buffer at 95°C for 10 min. Equal amounts of total proteins were frac- tionated on 10% SDS-PA gels and transferred topolyvinylidene fluoride membranes (Millipore) using a semidry method. The primary antibodies di- rected against PsbA (1:4,000), PsbD (1:5,000), RbcL (1:10,000), PsaD (1:1,000), Chl a Fluorescence Measurements AtpB (1:4,000), and cytb6 (1: 10,000) used in this study were obtained from Chl fluorescence parameters were measured using an imaging Chl fluo- Agrisera. The B6-B3 monoclonal antibody specific for phyB is described in rometer (Imaging PAM, M-Series; Walz). The method employed involved Hirschfeld et al. (1998). Signals were detected with the PierceECL Western preprogrammed treatments of 30-min dark periods to determine Fv/Fm, for Blotting Kit (Thermo Fisher Scientific) and quantified using the software ImageJ which F0 was measured at a low frequency of pulse-modulated measuring (National Institutes of Health). light, whereas Fm was obtained at saturation pulses of ;2,700 mmol photons m22 s21, for a duration of 0.8 s. The calculations and plotting of the parameters Nucleic Acid Extraction were performed using the ImagingWin software. Leaf tissue (100 mg) was homogenized in extraction buffer containing 200 mM TRIS/HCl (pH 7.5), 25 mM NaCl, 25 mM ethylenediaminetetraacetic acid, and 0.5% (w/v) SDS. After centrifugation, DNA was precipitated from the super- Chl Concentration Measurements natant by adding isopropyl alcohol. After washing with 70% (v/v) ethanol, the DNA was dissolved in distilled water. Chl content was measured according to Jeffrey and Humphrey (1975). For RNA isolation, frozen tissue was ground in liquid N. Total RNA was Briefly, 100 mg of fresh tissue was ground in 1 mL of cold 80% (v/v) acetone extracted by using TRIzol reagent (Invitrogen) according to the manufacturer’s and centrifuged at 12,000 g for 5 min at 4°C and the supernatant was saved. protocol. RNA samples were treated with DNase I (New England BioLabs) and Another 1 mL of 80% (v/v) acetone was added, and the extract was centrifuged quantified with a Nanodrop spectrophotometer (Thermo Fisher Scientific). again. The procedure was repeated until the green color of the pellet was no Isolated RNA was stored at 280°C until further use. longer discernible. Chl concentrations were calculated from spectroscopic ab- sorbance measurements at 663, 646, and 710 nm. RNA Gel-Blot Analysis For northern blotting, total RNA was extracted using Trizol reagent (Invi- Anthocyanin Measurement trogen). Aliquots (7 mg) of each sample were electrophoresed on a 1.5% (w/v) agarose gel containing formaldehyde, and transferred to a nylon membrane Measurement of anthocyanin content was conducted according to Neff and (Hybond-N+; Amersham Biosciences). The membrane was hybridized over- Chory (1998). Briefly, 100-mg samples of seedlings of each genotype that had night at 65°C with 32P-labeled cDNA fragments probes. Primers used to amplify been exposed to diverse light treatments were incubated with 300 mL of 1% (v/ the probes are listed in Supplemental Table S4. Signals were quantified with a v) HCl in methanol at 4°C overnight in the dark with shaking. Then 200 mL of phosphoimager (Typhoon; Amersham Biosciences) using the program Image- distilled water and 500 mL of chloroform were added, vortexed, and briefly Quant (GE Healthcare). centrifuged to separate anthocyanins from Chls. The total anthocyanin content was determined by measuring A530 and A657 of the aqueous phase using a spectrophotometer (Amersham Biosciences). The equation (A53020.25 3 cDNA Synthesis and RT-qPCR Analysis A657) 3 TV/(FW 3 1000) was used to quantify the relative amount of antho- cyanin, where TV = total volume of the extract (in milliliters) and FW = fresh Total RNA was extracted with the Direct-zol RNA Kit (Zymo Research) weight of seedlings (in grams). according to the manufacturer’s protocol, and 2 mg of the RNA was employed Plant Physiol. Vol. 180, 2019 337 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Role of PP7L in Chloroplast Development to synthesize cDNA using the iScript cDNA Synthesis Kit (Bio-Rad). RT-qPCR lncRNA-Seq and Data Analysis analysis was performed on a Bio-Rad iQ5 real-time PCR instrument with the iQ SYBR Green Supermix (Bio-Rad). The gene-specific primers used for this assay Total RNA from plants was isolated using Trizol (Invitrogen) and purified are listed in Supplemental Table S4. Each sample was quantified in triplicate using Direct-zol RNA MiniPrep Plus columns (Zymo Research) according to the and normalized using AT4G36800, which codes for an RUB1-conjugating en- manufacturer’s instructions. RNA integrity and quality was assessed by a 2100 zyme (RCE1) as an internal control. Bioanalyzer (Agilent). rRNA depletion, generation of RNA-Seq libraries, and 150-bp paired-end sequencing on an HiSeq 2500 system (Illumina) were con- ducted at Novogene Biotech with standard Illumina protocols. Three inde- Sample Preparation for Quantitative Proteomics pendent biological replicates were used per genotype. RNA-Seq reads were analyzed on the Galaxy platform (https://usegalaxy. Total proteins were extracted from 4-d–old wild-type and pp7l-1 seedlings org/). After grooming FASTQ files, adaptors were removed with Trimmomatic (four biological replicates for each genotype) by grinding 200 mg of frozen plant (Bolger et al., 2014), and sequencing quality was accessed with FastQC (http:// material into a fine powder in liquid N. After suspending the powder in 1 mL of www.bioinformatics.babraham.ac.uk/projects/fastqc/). Reads were mapped extraction buffer (100 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic to The Arabidopsis Information Resource genome annotation 10 with the acid, pH 7.5, 150 mM NaCl, 10 mM dithiothreitol, 1% [w/v] SDS, 13 Roche gapped-read mapper TopHat 2.1.1 (Kim et al., 2013) set for Forward Read cOmplete Protease Inhibitor Cocktail), cell debris was removed by centrifuga- unstranded libraries and adjusting the maximum intron length to 5,000 bp. tion at 10,000 g for 30 min. Proteins were then precipitated with chloroform– Reads were counted with featureCounts (Liao et al., 2014) with the help of the methanol (Wessel and Flügge, 1984), pellets were solubilized in 6 M guanidine gene annotation in Araport11 (www.araport.org/data/araport11). Differen- hydrochloride and protein concentration was determined by bicinchoninic tially expressed genes were obtained with DESeq2 (Love et al., 2014) running assay (Pierce BCA Protein Assay Kit; Thermo Fisher Scientific). Proteome ali- with the fit type set to “parametric,” and applying a twofold change cutoff and quots of 100 mg were reduced with 10 mM dithiothreitol (30 min, 37°C), alky- an adjusted P value , 0.05. lated with 50 mM iodoacetamide (30 min, room temperature, in the dark), and Sequencing data have been deposited to Gene Expression Omnibus (Edgar recovered by chloroform-methanol precipitation before digestion with trypsin et al., 2002) under the accession number GSE122495. (SERVA) at a proteome/enzyme ratio 100:1 (w/w). After overnight incubation at 37°C, stable-isotope dimethyl labeling was performed in solution according to Boersema et al. (2009). Briefly, the samples were labeled separately with TEM 20 mM NaCH3CN and either 40 mM CH2O (wild type, “light formaldehyde”) or 40 mM 13CD2O (pp7l-1, “heavy formaldehyde”) for 4 h at 37°C. Excess reagent The plant material was directly fixed after 4-d–old seedlings had been kept was quenched by adding 100 mM of TRIS at pH 6.8 (1 h, 37°C). After acidifi- for 8 h in darkness. Cotyledons were fixed using 75-mM cacodylate buffer cation with 2% (v/v) formic acid, equal amounts of light and heavy labeled containing 2.5% (v/v) glutaraldehyde. Postfixation was carried out using 1% samples were combined to generate four duplex samples. Before liquid (w/v) osmium tetroxide in water for 1 h followed by en-bloc staining with 1% chromatography-tandem mass spectrometry (LC-MS/MS) analysis, all sam- (w/v) uranyl acetate. After dehydration in a graded acetone series, the sample ples were desalted with home-made C18 stage tips (Rappsilber et al., 2007), material was gradually infiltrated with Spurr’s Resin. After polymerization, the eluates were vacuum-dried to near-dryness and stored at 280°C before LC- embedded samples were ultrathin-sectioned and poststained with lead citrate. MS/MS analysis. TEM was carried out using a model no. EM912 (Zeiss) with an integrated OMEGA-filter in the zero-loss mode and with an acceleration voltage of 80 kV. For image acquisition, we used a Tröndle 2k 3 2k slow-scan charge-coupled LC-MS/MS device camera together with the respective software package (Tröndle Restlichtverstärkersysteme). LC-MS/MS was performed on a nano-LC-system (Ultimate 3000 RSLC; Thermo Fisher Scientific) coupled to an Impact II high resolution quadrupole time of flight (Bruker) using a CaptiveSpray nano electrospray ionization Accession Numbers source (Bruker Daltonics). The nano-LC system was equipped with an Acclaim Pepmap nano-trap column (C18, 100 Å, 700 mm 3 2 cm; Thermo Fisher Sci- Sequence data from this article can be found in the GenBank/EMBL data entific) and an Acclaim Pepmap RSLC analytical column (C18, 100 Å, 75 mm 3 libraries under the accession numbers provided in Supplemental Table S4. 50 cm; Thermo Fisher Scientific). The peptide mixture was fractionated by ap- plying a linear gradient of 5% to 45% acetonitrile at a flow rate of 250 nL/min over a period of 150 min. MS1 spectra were acquired at 3 Hz with a mass range Supplemental Data from m/z 200–2000, with the Top-18 most intense peaks selected for MS/MS analysis using an intensity-dependent spectra acquisition time between 4 and The following supplemental materials are available. 16. The mass spectrometry proteomics data have been deposited to the Pro- Supplemental Figure S1. Identification of pp7l T-DNA insertion mutants. teomeXchange Consortium via the PRIDE (Vizcaíno et al., 2016) partner re- pository with the dataset identifier PXD012105. Supplemental Figure S2. Chl content and complementation of pp7l T-DNA insertion mutants. Supplemental Figure S3. rRNA maturation in 4-d–old Col-0 and pp7l mu- MS Data Analysis tant seedlings. MS raw files were analyzed with MaxQuant software (version 1.6.0.13; Supplemental Figure S4. Phenotypes of 4-week–old Col-0, pp7l-1, cop1-4, https://maxquant.org/) and peak lists were searched against the Arabidopsis and the different photoreceptor (cry1-304 cry2-1 and phyA-211 phyB-9) UniProt database (version February 2017) using the Andromeda search engine mutant plants. with default settings. Cys carbamidomethylation was set as a fixed modifica- tion, and acetylation of protein N-termini and Met oxidation were set as vari- Supplemental Figure S5. PhyB levels are elevated in pp7l mutants, but able modifications. Lys and N-terminal demethylation (light formaldehyde overexpression of phyB does not result in a pp7l phenotype. +28.03 D; heavy formaldehyde +34.06 D) were set as labels, and “requantified” Supplemental Figure S6. Overexpression of PP7L in Col-0. was enabled. Supplemental Table S1. SDS of the Fv/Fm values measured in the different pp7l mutant seedlings presented in Figure 3A. Statistical Analysis Supplemental Table S2. Genes whose transcript levels differed by more All statistical and bioinformatics analysis were performed using the soft- than twofold relative to Col-0 in 4-d–old LD-grown pp7l-1 seedlings. wares Perseus (version 1.6.0.2) and Microsoft Office Excel. Two-tailed Student’s Supplemental Table S3. Relative protein accumulation in the pp7l-1 mu- t‐test (P , 0.05) was used to define which protein groups are significantly tant compared to Col-0 grown for 4 d in LD conditions as determined by changed and a Box-plot analysis of the ratio of the frequency of occurrence of shotgun proteomics. each peptide in pp7l-1 mutant versus wild-type samples was used to identify significant (P # 0.05) outliers from the mean ratio of all peptides. Supplemental Table S4. Primers used in this study. 338 Plant Physiol. Vol. 180, 2019 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Xu et al. ACKNOWLEDGMENTS Chan KX, Phua SY, Crisp P, McQuinn R, Pogson BJ (2016) Learning the languages of the chloroplast: Retrograde signaling and beyond. Annu We thank Claus Schwechheimer, Ute Höcker, Christian Schmitz-Linne- Rev Plant Biol 67: 25–53 weber, Ferenc Nagy, and Eva Adam for providing the csn, cop1, sig5, and Chen D, Xu G, Tang W, Jing Y, Ji Q, Fei Z, Lin R (2013) Antagonistic basic P35S:PHYB-YFP phyB-9 seeds, respectively. We thank Paul Hardy for critical helix–loop–helix/bZIP transcription factors form transcriptional mod- reading of the manuscript and Elisabeth Gerick for technical assistance. ules that integrate light and reactive oxygen species signaling in Ara- Received January 17, 2019; accepted February 5, 2019; published February 13, bidopsis. Plant Cell 25: 1657–1673 2019. Chen M, Galvão RM, Li M, Burger B, Bugea J, Bolado J, Chory J (2010) Arabidopsis HEMERA/pTAC12 initiates photomorphogenesis by phy- tochromes. Cell 141: 1230–1240 LITERATURE CITED Chi W, He B, Mao J, Jiang J, Zhang L (2015) Plastid Sigma factors: Their individual functions and regulation in transcription. Biochim Biophys Abdallah F, Salamini F, Leister D (2000) A prediction of the size and Acta 1847: 770–778 evolutionary origin of the proteome of chloroplasts of Arabidopsis. Coll NS, Danon A, Meurer J, Cho WK, Apel K (2009) Characterization of Trends Plant Sci 5: 141–142 soldat8, a suppressor of singlet oxygen-induced cell death in Arabidopsis Albrecht V, Ingenfeld A, Apel K (2006) Characterization of the snowy seedlings. Plant Cell Physiol 50: 707–718 cotyledon 1 mutant of Arabidopsis thaliana: The impact of chloroplast Dohmann EM, Kuhnle C, Schwechheimer C (2005) Loss of the CONSTI- elongation factor G on chloroplast development and plant vitality. Plant TUTIVE PHOTOMORPHOGENIC9 signalosome subunit 5 is sufficient Mol Biol 60: 507–518 to cause the cop/det/fus mutant phenotype in Arabidopsis. Plant Cell Albrecht V, Ingenfeld A, Apel K (2008) Snowy cotyledon 2: The identifi- 17: 1967–1978 cation of a zinc finger domain protein essential for chloroplast devel- Dohmann EM, Levesque MP, Isono E, Schmid M, Schwechheimer C opment in cotyledons but not in true leaves. Plant Mol Biol 66: 599–608 (2008) Auxin responses in mutants of the Arabidopsis CONSTITUTIVE Albrecht V, Simková K, Carrie C, Delannoy E, Giraud E, Whelan J, Small PHOTOMORPHOGENIC9 signalosome. Plant Physiol 147: 1369–1379 ID, Apel K, Badger MR, Pogson BJ (2010) The cytoskeleton and the Dovzhenko A, Dal Bosco C, Meurer J, Koop HU (2003) Efficient regen- peroxisomal-targeted snowy cotyledon3 protein are required for chlo- eration from cotyledon protoplasts in Arabidopsis thaliana. Protoplasma roplast development in Arabidopsis. Plant Cell 22: 3423–3438 222: 107–111 Alonso JM, Stepanova AN, Leisse TJ, Kim CJ, Chen H, Shinn P, Edgar R, Domrachev M, Lash AE (2002) Gene Expression Omnibus: NCBI Stevenson DK, Zimmerman J, Barajas P, Cheuk R, et al (2003) gene expression and hybridization array data repository. Nucleic Acids Genome-wide insertional mutagenesis of Arabidopsis thaliana. Science Res 30: 207–210 301: 653–657 Edger PP, Hall JC, Harkess A, Tang M, Coombs J, Mohammadin S, Arsova B, Watt M, Usadel B (2018) Monitoring of plant protein post- Schranz ME, Xiong Z, Leebens-Mack J, Meyers BC, et al (2018) Bras- translational modifications using targeted proteomics. Front Plant Sci sicales phylogeny inferred from 72 plastid genes: A reanalysis of the 9: 1168 phylogenetic localization of two paleopolyploid events and origin of Asakura Y, Galarneau E, Watkins KP, Barkan A, van Wijk KJ (2012) novel chemical defenses. Am J Bot 105: 463–469 Chloroplast RH3 DEAD box RNA helicases in maize and Arabidopsis Ezer D, Shepherd SJK, Brestovitsky A, Dickinson P, Cortijo S, function in splicing of specific group II introns and affect chloroplast Charoensawan V, Box MS, Biswas S, Jaeger KE, Wigge PA (2017) The ribosome biogenesis. Plant Physiol 159: 961–974 G-box transcriptional regulatory code in Arabidopsis. Plant Physiol 175: Barkan A (1993) Nuclear mutants of maize with defects in chloroplast 628–640 polysome assembly have altered chloroplast RNA metabolism. Plant Farkas I, Dombrádi V, Miskei M, Szabados L, Koncz C (2007) Arabidopsis Cell 5: 389–402 PPP family of serine/threonine phosphatases. Trends Plant Sci 12: Barkan A, Small I (2014) Pentatricopeptide repeat proteins in plants. Annu 169–176 Rev Plant Biol 65: 415–442 Genoud T, Santa Cruz MT, Kulisic T, Sparla F, Fankhauser C, Métraux JP Bauer J, Chen K, Hiltbunner A, Wehrli E, Eugster M, Schnell D, Kessler F (2008) The protein phosphatase 7 regulates phytochrome signaling in (2000) The major protein import receptor of plastids is essential for Arabidopsis. PLoS One 3: e2699 chloroplast biogenesis. Nature 403: 203–207 Gommers CMM, Monte E (2018) Seedling establishment: A dimmer Baulcombe DC, Saunders GR, Bevan MW, Mayo MA, Harrison BD (1986) switch-regulated process between dark and light signaling. Plant Expression of biologically-active viral satellite RNA from the nuclear Physiol 176: 1061–1074 genome of transformed plants. Nature 321: 446–449 Hirschfeld M, Tepperman JM, Clack T, Quail PH, Sharrock RA (1998) Becker MG, Walker PL, Pulgar-Vidal NC, Belmonte MF (2017) SeqEnrich: Coordination of phytochrome levels in phyB mutants of Arabidopsis as A tool to predict transcription factor networks from co-expressed Ara- revealed by apoprotein-specific monoclonal antibodies. Genetics 149: bidopsis and Brassica napus gene sets. PLoS One 12: e0178256 523–535 Berry JO, Yerramsetty P, Zielinski AM, Mure CM (2013) Photosynthetic Hooper CM, Castleden IR, Tanz SK, Aryamanesh N, Millar AH (2017) gene expression in higher plants. Photosynth Res 117: 91–120 SUBA4: The interactive data analysis centre for Arabidopsis subcellular Bischof S, Baerenfaller K, Wildhaber T, Troesch R, Vidi PA, Roschitzki protein locations. Nucleic Acids Res 45(D1): D1064–D1074 B, Hirsch-Hoffmann M, Hennig L, Kessler F, Gruissem W, et al (2011) Hruz T, Laule O, Szabo G, Wessendorp F, Bleuler S, Oertle L, Widmayer Plastid proteome assembly without Toc159: Photosynthetic protein im- P, Gruissem W, Zimmermann P (2008) Genevestigator v3: A reference port and accumulation of N-acetylated plastid precursor proteins. Plant expression database for the meta-analysis of transcriptomes. Adv Bio- Cell 23: 3911–3928 informa 2008: 420747 Bobik K, Burch-Smith TM (2015) Chloroplast signaling within, between Hsieh WY, Liao JC, Wang HT, Hung TH, Tseng CC, Chung TY, Hsieh and beyond cells. Front Plant Sci 6: 781 MH (2017) The Arabidopsis thiamin-deficient mutant pale green1 lacks Boersema PJ, Raijmakers R, Lemeer S, Mohammed S, Heck AJR (2009) thiamin monophosphate phosphatase of the vitamin B1 biosynthesis Multiplex peptide stable isotope dimethyl labeling for quantitative pathway. Plant J 91: 145–157 proteomics. Nat Protoc 4: 484–494 Huang W, Sherman BT, Lempicki RA (2009) Bioinformatics enrichment Bolger AM, Lohse M, Usadel B (2014) Trimmomatic: A flexible trimmer for tools: Paths toward the comprehensive functional analysis of large gene Illumina sequence data. Bioinformatics 30: 2114–2120 lists. Nucleic Acids Res 37: 1–13 Cardinal-McTeague WM, Sytsma KJ, Hall JC (2016) Biogeography and Ishizaki Y, Tsunoyama Y, Hatano K, Ando K, Kato K, Shinmyo A, Kobori diversification of Brassicales: A 103-million-year tale. Mol Phylogenet M, Takeba G, Nakahira Y, Shiina T (2005) A nuclear-encoded sigma Evol 99: 204–224 factor, Arabidopsis SIG6, recognizes Sigma-70 type chloroplast pro- Carretero-Paulet L, Galstyan A, Roig-Villanova I, Martínez-García JF, moters and regulates early chloroplast development in cotyledons. Plant Bilbao-Castro JR, Robertson DL (2010) Genome-wide classification and J 42: 133–144 evolutionary analysis of the bHLH family of transcription factors in Jarvis P, López-Juez E (2013) Biogenesis and homeostasis of chloroplasts Arabidopsis, poplar, rice, moss, and algae. Plant Physiol 153: 1398–1412 and other plastids. Nat Rev Mol Cell Biol 14: 787–802 Plant Physiol. Vol. 180, 2019 339 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Role of PP7L in Chloroplast Development Jeffrey SW, Humphrey GF (1975) New spectrophotometric equations for Liebers M, Grübler B, Chevalier F, Lerbs-Mache S, Merendino L, determining chlorophylls a, b, c1 and c2 in higher plants, algae and Blanvillain R, Pfannschmidt T (2017) Regulatory shifts in plastid natural phytoplankton. Biochem Physiol Pflanz 167: 191–194 transcription play a key role in morphological conversions of plastids Jiao Y, Lau OS, Deng XW (2007) Light-regulated transcriptional networks during Plant development. Front Plant Sci 8: 23 in higher plants. Nat Rev Genet 8: 217–230 Liere K, Weihe A, Börner T (2011) The transcription machineries of plant Kanamaru K, Nagashima A, Fujiwara M, Shimada H, Shirano Y, mitochondria and chloroplasts: Composition, function, and regulation. Nakabayashi K, Shibata D, Tanaka K, Takahashi H (2001) An Arabi- J Plant Physiol 168: 1345–1360 dopsis sigma factor (SIG2)-dependent expression of plastid-encoded Lillo C, Kataya AR, Heidari B, Creighton MT, Nemie-Feyissa D, Ginbot tRNAs in chloroplasts. Plant Cell Physiol 42: 1034–1043 Z, Jonassen EM (2014) Protein phosphatases PP2A, PP4 and PP6: Me- Keeling PJ (2010) The endosymbiotic origin, diversification and fate of diators and regulators in development and responses to environmental plastids. Philos Trans R Soc Lond B Biol Sci 365: 729–748 cues. Plant Cell Environ 37: 2631–2648 Khan A, Fornes O, Stigliani A, Gheorghe M, Castro-Mondragon JA, van Ling Q, Huang W, Baldwin A, Jarvis P (2012) Chloroplast biogenesis is der Lee R, Bessy A, Chèneby J, Kulkarni SR, Tan G, et al (2018) regulated by direct action of the ubiquitin-proteasome system. Science JASPAR 2018: Update of the open-access database of transcription factor 338: 655–659 binding profiles and its web framework. Nucleic Acids Res 46(D1): Linhart C, Halperin Y, Shamir R (2008) Transcription factor and micro- D260–D266 RNA motif discovery: The Amadeus platform and a compendium of Kim DH, Kang JG, Yang SS, Chung KS, Song PS, Park CM (2002) A metazoan target sets. Genome Res 18: 1180–1189 phytochrome-associated protein phosphatase 2A modulates light sig- López-Juez E, Dillon E, Magyar Z, Khan S, Hazeldine S, de Jager SM, nals in flowering time control in Arabidopsis. Plant Cell 14: 3043–3056 Murray JA, Beemster GT, Bögre L, Shanahan H (2008) Distinct light- Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R, Salzberg SL (2013) initiated gene expression and cell cycle programs in the shoot apex and TopHat2: Accurate alignment of transcriptomes in the presence of in- cotyledons of Arabidopsis. Plant Cell 20: 947–968 sertions, deletions and gene fusions. Genome Biol 14: R36 Love MI, Huber W, Anders S (2014) Moderated estimation of fold change Kleine T, Leister D (2015) Emerging functions of mammalian and plant and dispersion for RNA-seq data with DESeq2. Genome Biol 15: 550 mTERFs. Biochim Biophys Acta 1847: 786–797 Martín G, Leivar P, Ludevid D, Tepperman JM, Quail PH, Monte E (2016) Kleine T, Leister D (2016) Retrograde signaling: Organelles go networking. Phytochrome and retrograde signalling pathways converge to antago- Biochim Biophys Acta 1857: 1313–1325 nistically regulate a light-induced transcriptional network. Nat Com- Kleine T, Kindgren P, Benedict C, Hendrickson L, Strand A (2007) mun 7: 11431 Genome-wide gene expression analysis reveals a critical role for McCormac AC, Terry MJ (2002) Light-signalling pathways leading to the CRYPTOCHROME1 in the response of Arabidopsis to high irradiance. co-ordinated expression of HEMA1 and Lhcb during chloroplast de- Plant Physiol 144: 1391–1406 velopment in Arabidopsis thaliana. Plant J 32: 549–559 Kleine T, Maier UG, Leister D (2009) DNA transfer from organelles to the McNellis TW, von Arnim AG, Araki T, Komeda Y, Miséra S, Deng XW nucleus: The idiosyncratic genetics of endosymbiosis. Annu Rev Plant (1994) Genetic and molecular analysis of an allelic series of cop1 mutants Biol 60: 115–138 suggests functional roles for the multiple protein domains. Plant Cell 6: Köhler D, Helm S, Agne B, Baginsky S (2016) Importance of translocon 487–500 subunit Tic56 for rRNA processing and chloroplast ribosome assembly. Miura E, Kato Y, Matsushima R, Albrecht V, Laalami S, Sakamoto W Plant Physiol 172: 2429–2444 (2007) The balance between protein synthesis and degradation in chlo- Kosugi S, Hasebe M, Tomita M, Yanagawa H (2009) Systematic identifi- roplasts determines leaf variegation in Arabidopsis yellow variegated cation of cell cycle-dependent yeast nucleocytoplasmic shuttling pro- mutants. Plant Cell 19: 1313–1328 teins by prediction of composite motifs. Proc Natl Acad Sci USA 106: Mockler TC, Guo H, Yang H, Duong H, Lin C (1999) Antagonistic actions 10171–10176 of Arabidopsis cryptochromes and phytochrome B in the regulation of Kubis S, Baldwin A, Patel R, Razzaq A, Dupree P, Lilley K, Kurth J, floral induction. Development 126: 2073–2082 Leister D, Jarvis P (2003) The Arabidopsis ppi1 mutant is specifically Møller SG, Kim YS, Kunkel T, Chua NH (2003) PP7 is a positive regulator defective in the expression, chloroplast import, and accumulation of of blue light signaling in Arabidopsis. Plant Cell 15: 1111–1119 photosynthetic proteins. Plant Cell 15: 1859–1871 Moore M, Gossmann N, Dietz KJ (2016) Redox regulation of cytosolic Lee K, Lee HJ, Kim DH, Jeon Y, Pai HS, Kang H (2014) A nuclear-encoded translation in plants. Trends Plant Sci 21: 388–397 chloroplast protein harboring a single CRM domain plays an important Motohashi R, Rödiger A, Agne B, Baerenfaller K, Baginsky S (2012) role in the Arabidopsis growth and stress response. BMC Plant Biol 14: Common and specific protein accumulation patterns in different albino/ 98 pale-green mutants reveals regulon organization at the proteome level. Leister D, Kleine T (2016) Definition of a core module for the nuclear Plant Physiol 160: 2189–2201 retrograde response to altered organellar gene expression identifies GLK Nagashima A, Hanaoka M, Shikanai T, Fujiwara M, Kanamaru K, overexpressors as gun mutants. Physiol Plant 157: 297–309 Takahashi H, Tanaka K (2004) The multiple-stress responsive plastid Leister D, Romani I, Mittermayr L, Paieri F, Fenino E, Kleine T (2014) sigma factor, SIG5, directs activation of the psbD blue light-responsive Identification of target genes and transcription factors implicated in promoter (BLRP) in Arabidopsis thaliana. Plant Cell Physiol 45: 357–368 translation-dependent retrograde signaling in Arabidopsis. Mol Plant 7: Neff MM, Chory J (1998) Genetic interactions between phytochrome A, 1228–1247 phytochrome B, and cryptochrome 1 during Arabidopsis development. Leister D, Wang L, Kleine T (2017) Organellar gene expression and accli- Plant Physiol 118: 27–35 mation of plants to environmental stress. Front Plant Sci 8: 387 Oelze ML, Muthuramalingam M, Vogel MO, Dietz KJ (2014) The link Leivar P, Monte E (2014) PIFs: systems integrators in plant development. between transcript regulation and de novo protein synthesis in the ret- Plant Cell 26: 56–78 rograde high light acclimation response of Arabidopsis thaliana. BMC Leivar P, Monte E, Oka Y, Liu T, Carle C, Castillon A, Huq E, Quail PH Genomics 15: 320 (2008) Multiple phytochrome-interacting bHLH transcription factors Osterlund MT, Hardtke CS, Wei N, Deng XW (2000) Targeted destabili- repress premature seedling photomorphogenesis in darkness. Curr Biol zation of HY5 during light-regulated development of Arabidopsis. Na- 18: 1815–1823 ture 405: 462–466 Li J, Li G, Wang H, Wang Deng X (2011) Phytochrome signaling mecha- Pfannschmidt T, Blanvillain R, Merendino L, Courtois F, Chevalier F, nisms. Arabidopsis Book 9: e0148 Liebers M, Grübler B, Hommel E, Lerbs-Mache S (2015) Plastid RNA Liang Z, Zhu N, Mai KK, Liu Z, Tzeng D, Osteryoung KW, Zhong S, polymerases: Orchestration of enzymes with different evolutionary or- Staehelin LA, Kang BH (2018) Thylakoid-bound polysomes and a igins controls chloroplast biogenesis during the plant life cycle. J Exp Bot dynamin-related protein, FZL, mediate critical stages of the linear 66: 6957–6973 chloroplast biogenesis program in greening Arabidopsis cotyledons. Pfeiffer A, Shi H, Tepperman JM, Zhang Y, Quail PH (2014) Combina- Plant Cell 30: 1476–1495 torial complexity in a transcriptionally centered signaling hub in Ara- Liao Y, Smyth GK, Shi W (2014) featureCounts: An efficient general pur- bidopsis. Mol Plant 7: 1598–1618 pose program for assigning sequence reads to genomic features. Bio- Pogson BJ, Albrecht V (2011) Genetic dissection of chloroplast biogenesis informatics 30: 923–930 and development: An overview. Plant Physiol 155: 1545–1551 340 Plant Physiol. Vol. 180, 2019 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved. Xu et al. Pribil M, Labs M, Leister D (2014) Structure and dynamics of thylakoids in Uhrig RG, Labandera AM, Moorhead GB (2013) Arabidopsis PPP family land plants. J Exp Bot 65: 1955–1972 of serine/threonine protein phosphatases: Many targets but few en- Privat I, Hakimi MA, Buhot L, Favory JJ, Mache-Lerbs S (2003) Charac- gines. Trends Plant Sci 18: 505–513 terization of Arabidopsis plastid sigma-like transcription factors SIG1, Ulker B, Peiter E, Dixon DP, Moffat C, Capper R, Bouché N, Edwards R, SIG2 and SIG3. Plant Mol Biol 51: 385–399 Sanders D, Knight H, Knight MR (2008) Getting the most out of pub- Putarjunan A, Liu X, Nolan T, Yu F, Rodermel S (2013) Understanding licly available T-DNA insertion lines. Plant J 56: 665–677 chloroplast biogenesis using second-site suppressors of immutans and Van Bel M, Diels T, Vancaester E, Kreft L, Botzki A, Van de Peer Y, var2. Photosynth Res 116: 437–453 Coppens F, Vandepoele K (2018) PLAZA 4.0: An integrative resource Rappsilber J, Mann M, Ishihama Y (2007) Protocol for micro-purification, for functional, evolutionary and comparative plant genomics. Nucleic enrichment, pre-fractionation and storage of peptides for proteomics Acids Res 46(D1): D1190–D1196 using StageTips. Nat Protoc 2: 1896–1906 Van Dingenen J, Blomme J, Gonzalez N, Inzé D (2016) Plants grow with a Roy B, von Arnim AG (2013) Translational regulation of cytoplasmic little help from their organelle friends. J Exp Bot 67: 6267–6281 mRNAs. Arabidopsis Book 11: e0165 Varotto C, Pesaresi P, Maiwald D, Kurth J, Salamini F, Leister D (2000) Ruckle ME, DeMarco SM, Larkin RM (2007) Plastid signals remodel light Identification of photosynthetic mutants of Arabidopsis by automatic signaling networks and are essential for efficient chloroplast biogenesis screening for altered effective quantum yield of photosystem 2. Photo- in Arabidopsis. Plant Cell 19: 3944–3960 synthetica 38: 497–504 Ruppel NJ, Hangarter RP (2007) Mutations in a plastid-localized elonga- Vizcaíno JA, Csordas A, Del-Toro N, Dianes JA, Griss J, Lavidas I, Mayer tion factor G alter early stages of plastid development in Arabidopsis G, Perez-Riverol Y, Reisinger F, Ternent T, et al (2016) 2016 update of thaliana. BMC Plant Biol 7: 37  the PRIDE database and its related tools. Nucleic Acids Res 44: 11033 Rusaczonek A, Czarnocka W, Kacprzak S, Wito n D, Slesak I, Szechy nska- Wagner D, Tepperman JM, Quail PH (1991) Overexpression of phyto- Hebda M, Gawro nski P, Karpi nski S (2015) Role of phytochromes A chrome B induces a short hypocotyl phenotype in transgenic Arabi- and B in the regulation of cell death and acclimatory responses to UV stress in Arabidopsis thaliana. J Exp Bot 66: 6679–6695 dopsis. Plant Cell 3: 1275–1288 Saini G, Meskauskiene R, Pijacka W, Roszak P, Sjögren LL, Clarke AK, Wang R, Zhao J, Jia M, Xu N, Liang S, Shao J, Qi Y, Liu X, An L, Yu F Straus M, Apel K (2011) ‘happy on norflurazon’ (hon) mutations implicate (2018) Balance between cytosolic and chloroplast translation affects leaf perturbance of plastid homeostasis with activating stress acclimatiza- variegation. Plant Physiol 176: 804–818 tion and changing nuclear gene expression in norflurazon-treated Wang X, Li W, Piqueras R, Cao K, Deng XW, Wei N (2009) Regulation of seedlings. Plant J 65: 690–702 COP1 nuclear localization by the COP9 signalosome via direct interac- Sakr S, Wang M, Dédaldéchamp F, Perez-Garcia MD, Ogé L, Hamama L, tion with CSN1. Plant J 58: 655–667 Atanassova R (2018) The sugar-signaling hub: Overview of regulators Waters MT, Langdale JA (2009) The making of a chloroplast. EMBO J 28: and interaction with the hormonal and metabolic network. Int J Mol Sci 2861–2873 19: 2506 Wessel D, Flügge UI (1984) A method for the quantitative recovery of Schmitz-Linneweber C, Lampe MK, Sultan LD, Ostersetzer-Biran O protein in dilute solution in the presence of detergents and lipids. Anal (2015) Organellar maturases: A window into the evolution of the Biochem 138: 141–143 spliceosome. Biochim Biophys Acta 1847: 798–808 Wikström N, Savolainen V, Chase MW (2001) Evolution of the angio- Schulz E, Tohge T, Zuther E, Fernie AR, Hincha DK (2015) Natural var- sperms: Calibrating the family tree. Proc Biol Sci 268: 2211–2220 iation in flavonol and anthocyanin metabolism during cold acclimation Williams ME, Foster R, Chua NH (1992) Sequences flanking the hexameric in Arabidopsis thaliana accessions. Plant Cell Environ 38: 1658–1672 G-box core CACGTG affect the specificity of protein binding. Plant Cell Shen Y, Kim JI, Song PS (2005) NDPK2 as a signal transducer in the 4: 485–496 phytochrome-mediated light signaling. J Biol Chem 280: 5740–5749 Wingler A (2018) Transitioning to the next phase: The role of sugar sig- Shikanai T (2015) RNA editing in plants: Machinery and flexibility of site naling throughout the plant life cycle. Plant Physiol 176: 1075–1084 recognition. Biochim Biophys Acta 1847: 779–785 Wisman E, Hartmann U, Sagasser M, Baumann E, Palme K, Hahlbrock K, Shimada H, Mochizuki M, Ogura K, Froehlich JE, Osteryoung KW, Saedler H, Weisshaar B (1998) Knock-out mutants from an En-1 muta- Shirano Y, Shibata D, Masuda S, Mori K, Takamiya K (2007) Arabi- genized Arabidopsis thaliana population generate phenylpropanoid bio- dopsis cotyledon-specific chloroplast biogenesis factor CYO1 is a protein synthesis phenotypes. Proc Natl Acad Sci USA 95: 12432–12437 disulfide isomerase. Plant Cell 19: 3157–3169 Woodson JD, Chory J (2008) Coordination of gene expression between Shinomura T, Hanzawa H, Schäfer E, Furuya M (1998) Mode of phyto- organellar and nuclear genomes. Nat Rev Genet 9: 383–395 chrome B action in the photoregulation of seed germination in Arabi- Woodson JD, Perez-Ruiz JM, Schmitz RJ, Ecker JR, Chory J (2013) Sigma dopsis thaliana. Plant J 13: 583–590 factor-mediated plastid retrograde signals control nuclear gene expres- Solfanelli C, Poggi A, Loreti E, Alpi A, Perata P (2006) Sucrose-specific sion. Plant J 73: 1–13 induction of the anthocyanin biosynthetic pathway in Arabidopsis. Wurzinger B, Nukarinen E, Nägele T, Weckwerth W, Teige M (2018) The Plant Physiol 140: 637–646 SnRK1 kinase as central mediator of energy signaling between different Stern DB, Goldschmidt-Clermont M, Hanson MR (2010) Chloroplast RNA organelles. Plant Physiol 176: 1085–1094 metabolism. Annu Rev Plant Biol 61: 125–155 Xu X, Paik I, Zhu L, Huq E (2015) Illuminating progress in phytochrome- Stoppel R, Meurer J (2012) The cutting crew—Ribonucleases are key mediated light signaling pathways. Trends Plant Sci 20: 641–650 players in the control of plastid gene expression. J Exp Bot 63: 1663–1673 Yi C, Deng XW (2005) COP1—From plant photomorphogenesis to mam- Strittmatter G, Kössel H (1984) Cotranscription and processing of 23S, 4.5S malian tumorigenesis. Trends Cell Biol 15: 618–625 and 5S rRNA in chloroplasts from Zea mays. Nucleic Acids Res 12: Yu QB, Li G, Wang G, Sun JC, Wang PC, Wang C, Mi HL, Ma WM, Cui J, 7633–7647 Cui YL, et al (2008) Construction of a chloroplast protein interaction Su L, Hou P, Song M, Zheng X, Guo L, Xiao Y, Yan L, Li W, Yang J (2015) Synergistic and antagonistic action of phytochrome (Phy) A and PhyB network and functional mining of photosynthetic proteins in Arabidopsis during seedling de-etiolation in Arabidopsis thaliana. Int J Mol Sci 16: thaliana. Cell Res 18: 1007–1019 12199–12212 Yu X, Liu H, Klejnot J, Lin C (2010) The cryptochrome blue light receptors. Tiller N, Bock R (2014) The translational apparatus of plastids and its role Arabidopsis Book 8: e0135 in plant development. Mol Plant 7: 1105–1120 Zagari N, Sandoval-Ibañez O, Sandal N, Su J, Rodriguez-Concepcion M, Tsunoyama Y, Ishizaki Y, Morikawa K, Kobori M, Nakahira Y, Takeba G, Stougaard J, Pribil M, Leister D, Pulido P (2017) SNOWY COTYLE- Toyoshima Y, Shiina T (2004) Blue light-induced transcription of DON 2 promotes chloroplast development and has a role in leaf vari- plastid-encoded psbD gene is mediated by a nuclear-encoded tran- egation in both Lotus japonicus and Arabidopsis thaliana. Mol Plant 10: scription initiation factor, AtSig5. Proc Natl Acad Sci USA 101: 721–734 3304–3309 Zhang Y, Primavesi LF, Jhurreea D, Andralojc PJ, Mitchell RA, Powers Ühlken C, Horvath B, Stadler R, Sauer N, Weingartner M (2014) MAIN- SJ, Schluepmann H, Delatte T, Wingler A, Paul MJ (2009) Inhibition of LIKE1 is a crucial factor for correct cell division and differentiation in SNF1-related protein kinase1 activity and regulation of metabolic Arabidopsis thaliana. Plant J 78: 107–120 pathways by trehalose-6-phosphate. Plant Physiol 149: 1860–1871 Plant Physiol. Vol. 180, 2019 341 Downloaded from on July 18, 2019 - Published by www.plantphysiol.org Copyright © 2019 American Society of Plant Biologists. All rights reserved.