TY - JOUR
AU - Mackenzie, Sally A.
AB - Abstract Plastids comprise a complex set of organelles in plants that can undergo distinctive patterns of differentiation and redifferentiation during their lifespan. Plastids localized to the epidermis and vascular parenchyma are distinctive in size, structural features, and functions. These plastids are termed “sensory” plastids, and here we show their proteome to be distinct from chloroplasts, with specialized stress-associated features. The distinctive sensory plastid proteome in Arabidopsis (Arabidopsis thaliana) derives from spatiotemporal regulation of nuclear genes encoding plastid-targeted proteins. Perturbation caused by depletion of the sensory plastid-specific protein MutS HOMOLOG1 conditioned local, programmed changes in gene networks controlling chromatin, stress-related phytohormone, and circadian clock behavior and producing a global, systemic stress response in the plant. We posit that the sensory plastid participates in sensing environmental stress, integrating this sensory function with epigenetic and gene expression circuitry to condition heritable stress memory. Plastids comprise a defining feature of the plant cell, present-day outcomes of an endosymbiotic event that distinguishes the plant kingdom. Arising through fission rather than de novo biogenesis during the development of a plant, plastids undergo specialized differentiations. The transition from proplastid to chloroplast defines tissues that will be photosynthetic within a plant, with amyloplasts occurring in nonphotosynthetic tissues (Enami et al., 2011), chromoplasts within reproductive and senescing organs (Barsan et al., 2012), lipid-containing plastids during flower and seed development (Rottet et al., 2015), and so on. These transitions involve proteome and metabolic changes that, in most cases, are not yet fully defined. Integration of mutational, metabolic, and infrastructural data for plastid development shows plastid biogenesis to be surprisingly complex (Charuvi et al., 2012; Pogson et al., 2015; van Wijk and Kessler, 2017). Evidence has accumulated for a class of plastids that is specialized for stress sensing and acclimation. For example, plastids within the epidermis that are approximately 30% the size of a mesophyll chloroplast have been associated with osmotic stress response (Veley et al., 2012) and coordination of auxin induction for stem growth under shade conditions (Procko et al., 2016). Similarly sized plastids within the vascular parenchyma have been associated with ABA biosynthesis and high-light and drought responses (Endo et al., 2008; Galvez-Valdivieso et al., 2009). These two groups of plastids appear to share at least a portion of their proteomes distinct from mesophyll chloroplasts (Virdi et al., 2016), suggesting that more detailed proteome investigations would provide insight into these specializations. Studies of vascular and epidermal plastids identify MSH1, a plant-specific mitochondrial protein (Abdelnoor et al., 2003; Xu et al., 2011) that is also associated with this specialized plastid type. Spatiotemporal regulation ensures that MSH1 functions in epidermis, vascular parenchyma, and reproductive cells; the GFP-tagged protein localizes to small plastids within these tissues (Virdi et al., 2016). Steady-state MSH1 transcript levels decline in plants under abiotic stress (Shedge et al., 2010; Xu et al., 2011), and RNA interference (RNAi) suppression of MSH1 triggers constitutive and wide-ranging stress and developmental pathways in a plant grown under normal conditions (Xu et al., 2012; Virdi et al., 2016). Detailed analysis of msh1 T-DNA mutants shows dramatically altered expression of genes controlling abiotic and biotic stress, oxidative and redox, phytohormone response, and circadian rhythm pathways (Shao et al., 2017). The MSH1-RNAi or msh1 mutant phenotype is also characterized by a range in phenotype intensity, plant variegation, variable growth rate, delayed maturity and flowering, altered circadian rhythm, altered nonphotoactive redox behavior, and abiotic stress tolerance (Shedge et al., 2010; Xu et al., 2012). Earlier hemicomplementation experiments showed that this integrated and pleiotropic response to MSH1 suppression emanates from the plastid (Xu et al., 2011). These observations imply that the “sensory” plastid, in which MSH1 resides, differentiates in a spatially and temporally defined manner and influences a wide range of integrated stress-response and developmental networks. The global stress phenotype that derives from MSH1 RNAi knockdown is characterized by genome-wide DNA methylome repatterning (Virdi et al., 2015), implying an epigenomic component to the phenotype. Transgene-null progeny from these lines can retain the msh1 phenotype, more uniform and slightly attenuated, as heritable “memory” lines (Xu et al., 2012) that display altered methylation (X. Yang, R. Sanchez, H. Kundariya, Y. Wamboldt, J. Barreras, and S. Mackenzie, unpublished data). These data provide nascent evidence of epigenetic changes, induced by this unusual plastid type, that are programmed and presumably adaptive for the plant. Based on observations by our group and others, we postulate that epidermal cells, in direct contact with the external environment, and vascular parenchyma cells, which are able to transmit signals systemically, harbor a plastid type that is specialized for environmental sensing. Here, we investigate this putative “sensory plastid” in Arabidopsis (Arabidopsis thaliana), its proteome and stress-responsive behavior relative to neighboring mesophyll chloroplasts, and show the sensory plastid and chloroplast proteomes to be overlapping but distinct. Many stress-related proteins earlier assigned to the chloroplast appear, in fact, to be localized to sensory plastids. Investigation of the translatome within sensory plastid-containing cells, following MSH1 disruption, provides evidence of calcium signaling, redox, chromatin remodeling, and circadian clock network changes that, together, may account for the strikingly pleiotropic developmental phenotypes and the plant’s capacity for transgenerational stress memory. RESULTS Sensory Plastids and Chloroplasts Can Be Differentially Sorted from Arabidopsis Floral Stem Tissues Previous work shows that MSH1 expression within the epidermis and vascular parenchyma results in MSH1 localization to the nucleoid and thylakoid membrane of the sensory plastid (Virdi et al., 2016). PPD3, a putative stress protein related to PsbP, also localizes to sensory plastids and may physically interact with MSH1 (Virdi et al., 2016), providing evidence of distinctive sensory plastid proteome features. We predicted that differential accumulation of proteins in sensory plastid versus chloroplast populations would reveal distinct functional information if the organelles could be physically separated. Taking advantage of functional complementation of the msh1 mutant with a native promoter::MSH1::GFP construction (Virdi et al., 2016), we adapted fluorescence-activated cell sorting (FACS)-based procedures to separate Arabidopsis sensory plastids from mesophyll chloroplasts (Fig. 1A). Figure 1. Open in new tabDownload slide Sorting of sensory plastids and chloroplasts. A, A chart of the experimental procedure. Sensory plastids were tagged with MSH1::GFP expressed under the control of the MSH1 native promoter. FACS was used to separate GFP-tagged sensory plastids from mesophyll chloroplasts. Mass spectrometry allowed proteome analysis of the individual fractions. B, Plants expressing an MSH1::GFP fusion in mutant msh1 background (DUAL) were used to extract plastids. C and D, Cross sections of floral stems showing GFP-associated sensory plastids confined to vascular parenchyma and epidermal cells (C; scale bar, 50 µm) and a nontransgenic plant for comparison (D; scale bar, 50 µm). E and F, Confocal laser-scanning image of sorted sensory plastids (E; scale bar, 5 µm) and chloroplasts (F; scale bar, 5 µm). Electron microscopy image of sorted sensory plastid (G; scale bar, 2 µm) and chloroplast (H; scale bar, 2 µm). I, Planar area of sensory plastids and chloroplasts. Values are means ± sd (n = 4). J, Plastoglobule count in sensory plastids and chloroplasts. Values are means ± sd (n = 4). Figure 1. Open in new tabDownload slide Sorting of sensory plastids and chloroplasts. A, A chart of the experimental procedure. Sensory plastids were tagged with MSH1::GFP expressed under the control of the MSH1 native promoter. FACS was used to separate GFP-tagged sensory plastids from mesophyll chloroplasts. Mass spectrometry allowed proteome analysis of the individual fractions. B, Plants expressing an MSH1::GFP fusion in mutant msh1 background (DUAL) were used to extract plastids. C and D, Cross sections of floral stems showing GFP-associated sensory plastids confined to vascular parenchyma and epidermal cells (C; scale bar, 50 µm) and a nontransgenic plant for comparison (D; scale bar, 50 µm). E and F, Confocal laser-scanning image of sorted sensory plastids (E; scale bar, 5 µm) and chloroplasts (F; scale bar, 5 µm). Electron microscopy image of sorted sensory plastid (G; scale bar, 2 µm) and chloroplast (H; scale bar, 2 µm). I, Planar area of sensory plastids and chloroplasts. Values are means ± sd (n = 4). J, Plastoglobule count in sensory plastids and chloroplasts. Values are means ± sd (n = 4). Complemented lines stably expressing the MSH1::GFP fusion (Fig. 1B) were used to extract total plastids from floral stems (Fig. 1, C and D), where MSH1-expressing tissues are enriched (Virdi et al., 2016). Sensory plastids and mesophyll chloroplasts were fractionated by FACS, discriminating based on positive GFP signal in sensory plastids and separable levels of autofluorescence in the two fractions (Supplemental Figs. S1, A and B). Multiple sorting replicates recovered sufficient amounts of the two populations for downstream experiments (Supplemental Fig. S2C). Postsorting evaluation of organelle populations indicated a sufficiently high level of purity (∼90%) for separation of sensory plastids (Supplemental Fig. S2D). Sorted plastids were inspected for integrity using laser-scanning confocal microscopy for differential GFP fluorescence. Punctate GFP signal could be seen within nucleoids of sensory plastids (Fig. 1E) and was not detectable in mesophyll chloroplasts (Fig. 1F). Observed organelle size differences were consistent with previous in planta observations (Virdi et al., 2016). Electron microscopy imaging (Fig. 1, G and H) showed differences in internal structural features. Sorted sensory plastids had a planar area of 3.98 ± 0.72 µm2 (Fig. 1I), relative to chloroplasts at 11.76 ± 0.67 µm2, with granal stacking appearing less complex in sensory plastids and plastoglobuli counts lower (Fig. 1J). These data served as confirmation for the successful isolation of a distinct plastid type. Quantitative measurement of chlorophyll fluorescence per equal counts in each plastid population (Fig. 2A) indicated that, following normalization for plastid size, the chlorophyll level of sensory plastids was approximately 66% that of chloroplasts (Fig. 2B). Figure 2. Open in new tabDownload slide Features distinguishing sensory plastids and chloroplasts. A, Chlorophyll mean fluorescence intensity (MFI) in sensory plastids and chloroplasts. B, Normalized chlorophyll MFI per planar area (chloroplast MFI as 100%). C, Distribution of proteins identified in sensory plastids and chloroplasts. D, GO enrichment analysis of combined data sets obtained in this study (“sensory + chloroplast”) in comparison to the plastid data set curated in PPDB. E, GO terms (% of counts) in the total proteome of chloroplasts. F, GO terms (% of counts) in the total proteome of sensory plastids. Fuchsia color in E and F denotes the ranking for the category “photosynthesis” in each plastid population. Figure 2. Open in new tabDownload slide Features distinguishing sensory plastids and chloroplasts. A, Chlorophyll mean fluorescence intensity (MFI) in sensory plastids and chloroplasts. B, Normalized chlorophyll MFI per planar area (chloroplast MFI as 100%). C, Distribution of proteins identified in sensory plastids and chloroplasts. D, GO enrichment analysis of combined data sets obtained in this study (“sensory + chloroplast”) in comparison to the plastid data set curated in PPDB. E, GO terms (% of counts) in the total proteome of chloroplasts. F, GO terms (% of counts) in the total proteome of sensory plastids. Fuchsia color in E and F denotes the ranking for the category “photosynthesis” in each plastid population. The Chloroplast Proteome Is Predominantly Organized for Photosynthesis, Whereas the Sensory Plastid Proteome May Be Optimized for Stress Response Sorted plastid populations were subjected to mass spectrometry-based proteomics using nano liquid chromatography-tandem mass spectrometry (nanoLC-MS/MS). Since our objective was to elucidate possible differences between sensory plastid and chloroplast samples and because protein numbers in proteomic experiments can vary significantly from study to study, we filtered and retained only those hits predicted to be chloroplast localized or present in either the PPDB (Sun et al., 2009) or the AT_CHLORO (Bruley et al., 2012) databases to enhance stringency (Supplemental Fig. S2, A and B). Protein presence/absence was assessed if identification could be established at greater than 99% probability for at least two identified peptides (see “Materials and Methods”). Surprisingly, sensory plastids exhibited significantly more hits overall (833) than mesophyll chloroplasts (304), while 246 hits were shared by both populations (Fig. 2C; Supplemental Data Set 1). Given these observations, we tested whether we could reconstitute, by combining sensory plastid and chloroplast data sets, the functional (Gene Ontology [GO], biological function) ranking for the list of plastid proteins within the PPDB database. As shown in Figure 2D, a combined sensory plastid and chloroplast assembly shows compelling similarity to the available chloroplast proteome database for GO categories, adding confidence in our proteome results. Some differences between the reconstituted sample and the publicly available proteome data were anticipated because of discrepancies between databases (PPDB versus AT_CHLORO); these factors were taken into account within the reference list to filter the proteome data. Although our study is foundational for the model plant Arabidopsis, for maize (Zea mays), the PPDB database also reports 173 fewer hits for mesophyll chloroplast than for the bundle sheath only. While it is difficult to assess the purity of plastid preparations used in all previous studies, it is possible that Arabidopsis chloroplast databases contain, to varying degrees, a mixture of hits coming from heterogeneous plastid populations. Another way to validate our results involved intersection of the sensory plastid proteome data set with single-gene studies that investigated tissue-specific expression of plastid-localized proteins within the vascular tissue, as in the case of reticulata mutants. For example, PHOSPHOENOLPYRUVATE/PHOSPHATE TRANSLOCATOR1 (PPT1 or CUE1), NAD(P)H-THIOREDOXIN REDUCTASE, CARBAMOYL PHOSPHATE SYNTHETASE α, CARBAMOYL PHOSPHATE SYNTHETASE β, and At3g08640, encoding an alpha-tandem protein, are all vascular parenchyma specific or vascular enriched (Lundquist et al., 2014) and were captured in our sensory plastid proteome data set but not in the chloroplast data set (Supplemental Table S1). We interrogated the functional categorization of the derived sensory plastid and chloroplast proteomes to find considerable differences between the two data sets. Not surprisingly, the main functional category in mesophyll chloroplasts was “photosynthesis” (Fig. 2E). However, functional categories for “oxidation-reduction process,” “response to cytokinin,” and “response to cadmium ion” ranked above “photosynthesis” in sensory plastids (Fig. 2F). Proteins shared by both sensory and chloroplast populations included proteins for the photosynthetic machinery, including proteins related to PSI and PSII assembly and components of electron transport important for photosynthesis (Supplemental Table S2). These results imply that the sensory plastid may be photosynthetically competent to some extent. Moreover, a previous study showed that MSH1 associates and copurifies with electron transport components in the sensory plastid (Virdi et al., 2016). To better understand the proteome differences found, we performed GO enrichment analysis for protein hits that were present exclusively in each plastid population. Some of the components that were present within the chloroplast but seemingly reduced or absent from sensory plastid fell into categories of “transport” or “photosynthesis” (Supplemental Fig. S3A; Supplemental Data Set 1), but these may contribute to optimization rather than essential factors. In contrast, categories within the sensory plastid-specific data set were mainly stress-related, including “oxidation-reduction process,” “response to cadmium ion,” “cytokinin,” “cold,” “salt,” “oxidative stresses,” and “defense response to bacterium” (Supplemental Fig. S3B). Additionally, biosynthetic pathways for fatty acid and amino acid synthesis were enriched in the sensory plastid data set, emphasizing biosynthetic capacity of these organelles. In fact, nearly the entire shikimate pathway (from PEP to chorismate) was localized to the sensory plastids. Notably, nearly all the “reticulate” mutants for plastid-localized proteins in the vasculature or bundle sheath cells are affected in plastid primary metabolism (Lundquist et al., 2014). These data, collectively, indicate that mesophyll chloroplasts and “sensory” plastids represent functionally distinct plastid populations, with sensory plastids displaying features of specialization for enhanced stress response and metabolic functions. Sensory Plastid-Containing Cells Undergo Precise Changes in Gene Expression with Depletion of MSH1 GFP tagging of the MSH1 sensory plastid-specific protein permitted in planta investigation of its spatiotemporal expression patterns. We explored the pattern of expression of the Arabidopsis MSH1 gene under its native promoter in transgenic tobacco (Nicotiana tabacum). In tobacco epidermis, MSH1 localized to small plastids in the trichome stalk but was not detected in the trichome head (Fig. 3, A and B). These small plastids showed evidence of autofluorescence, in contrast to those in Arabidopsis, where trichome plastids show little or none (Fig. 3, C and D). The trichome head in tobacco harbored larger autofluorescent chloroplasts, which have been shown by others to undergo active photosynthesis and express a trichome head-specific Rubisco (Laterre et al., 2017). From our observations, the small-sized, MSH1-expressing plastids in the stalk are presumed to be sensory plastids. Noteworthy is the observation in tobacco, as in Arabidopsis, that plastid differentiation is accompanied by differential expression of MSH1 (Fig. 3E); these observations in tobacco emphasize the precise “on-off” promoter behavior required for sensory plastid differentiation across contiguous tissues (Fig. 3F). Figure 3. Open in new tabDownload slide Sensory plastids in tobacco. A, Trichome of transgenic tobacco line expressing MSH1::GFP in sensory plastids. Inset is an enlargement of sensory plastids in the trichome stalk taken from an independent experiment in the region indicated by the white arrow. B, Trichome of nontransgenic tobacco. C, Trichome of stable transgenic Arabidopsis line expressing MSH1::GFP in sensory plastids. Inset is a zoom image of sensory plastids from the region indicated by the white arrow. D, Trichome of nontransgenic Arabidopsis. Inset is a zoom image of sensory plastids. E, Epidermal sensory plastids are positive for MSH1::GFP. F, On and off modulation of MSH1 in different tissues of tobacco and Arabidopsis. VP, Vascular parenchyma; M, mesophyll; E, epidermis; T, trichome. All scale bars are 25 µm. Figure 3. Open in new tabDownload slide Sensory plastids in tobacco. A, Trichome of transgenic tobacco line expressing MSH1::GFP in sensory plastids. Inset is an enlargement of sensory plastids in the trichome stalk taken from an independent experiment in the region indicated by the white arrow. B, Trichome of nontransgenic tobacco. C, Trichome of stable transgenic Arabidopsis line expressing MSH1::GFP in sensory plastids. Inset is a zoom image of sensory plastids from the region indicated by the white arrow. D, Trichome of nontransgenic Arabidopsis. Inset is a zoom image of sensory plastids. E, Epidermal sensory plastids are positive for MSH1::GFP. F, On and off modulation of MSH1 in different tissues of tobacco and Arabidopsis. VP, Vascular parenchyma; M, mesophyll; E, epidermis; T, trichome. All scale bars are 25 µm. Continuity of sensory plastid behavior between Arabidopsis and tobacco is reiterated by the observation that RNAi suppression of MSH1 elicits markedly similar programmed changes in plant growth pattern and epigenetic memory in six different plant species (Xu et al., 2012). These observations prompted us to investigate the local cellular changes that occur in response to sensory plastid perturbation. Since MSH1 is down-regulated by environmental stresses and its depletion elicits a strong transcriptomic stress response, we used the msh1 phenomenon as a first proxy to understand sensory plastid functions. Capitalizing on the specificity of the MSH1 promoter to drive expression of a FLAG-tagged ribosome protein, together with affinity-purification and RNA-sequencing (RNA-seq; translatome), we investigated sensory plastid cell-specific responses in plants depleted of MSH1 by the translating ribosome affinity purification (TRAP)-seq method (Supplemental Fig. S4A; Reynoso et al., 2015). The experiments involved introduction of large-subunit ribosomal protein L18 fusion with N-terminal FLAG epitope tag under the control of the MSH1 promoter, introduced to both the msh1 mutant and Col-0 wild type (Fig. 4A; Supplemental Fig. S4B). Figure 4. Open in new tabDownload slide Cell-specific translatome profiling. A, Representative plants expressing a FLAG::RPL-18 fusion in both the wild type (WT; Col-0 RPL-18) and the msh1 mutant (msh1 RPL-18) backgrounds. B, Principal component analysis plot for translatome (IP) and transcriptome (total) reads. C, Prominent GO categories for biological function in the translatome and the transcriptome (this study) and total transcriptome (T; total plant data set from Shao et al., 2017). D, Network enrichment analysis of KEGG-derived pathways for the sensory plastid-associated translatome output. Figure 4. Open in new tabDownload slide Cell-specific translatome profiling. A, Representative plants expressing a FLAG::RPL-18 fusion in both the wild type (WT; Col-0 RPL-18) and the msh1 mutant (msh1 RPL-18) backgrounds. B, Principal component analysis plot for translatome (IP) and transcriptome (total) reads. C, Prominent GO categories for biological function in the translatome and the transcriptome (this study) and total transcriptome (T; total plant data set from Shao et al., 2017). D, Network enrichment analysis of KEGG-derived pathways for the sensory plastid-associated translatome output. Principal component analysis of derived translatome data clustered the mutant samples distinctly from wild type, providing evidence of a differential gene expression response following MSH1 depletion (Fig. 4B). Cell-specific translatome profiles captured evidence of differential translation for 3,124 genes in MSH1-depleted cells, whereas total transcriptome profiling from the same tissue was able to distinguish only 300 genes with differential expression. Similar differential enrichment effects have been reported for other systems (Mustroph et al., 2009; Endo et al., 2014), perhaps due to ribosome enrichment prior to RNA extraction in translatome procedures. Thus, this study with pooled floral stem samples and previous studies with selected phenotypes and rosettes tissues (Shao et al., 2017) were unable to capture the entire translatome gene response via transcriptome analysis (Supplemental Fig. S5C; Supplemental Data Set 2). Differentially expressed gene number in the msh1 mutant can vary markedly depending on the generation and phenotype intensity (Shao et al., 2017). Yet, GO ontology analysis of the transcriptome in this study (Supplemental Data Set 3) confirms similar global stress response to earlier reports (Shao et al., 2017), with categories altered for response to cold, wounding, oxidative stress, response to hormones (jasmonic acid, auxin, gibberellin, and ABA) and circadian rhythm (Fig. 4C). For translatome analysis, we used the full list of differentially translated genes to perform GO enrichment (Fig. 4C), network enrichment analysis test and network-based enrichment analysis (Fig. 4D). KEGG networks that were identified included MAPK signaling, circadian rhythm, Arg and Pro metabolism, metabolic pathways, biosynthesis of secondary metabolites, SNARE interactions in vascular transport, autophagy, phagosome, plant-pathogen interaction, and plant hormone signal transduction. GO enrichment analysis identified several related categories (Supplemental Data Set 3), the most prominent involving redox, stress response and signaling, chromatin remodeling, and cellular growth behavior. Within the “response to oxidative stress” category were genes for peroxidases, glutathione transferases, and thioredoxins, including ASCORBATE PEROXIDASE1, scavenging hydrogen peroxide in plant cells, and GLUTATHIONE PEROXIDASE8, for suppression of oxidative damage in the nucleus and cytosol. In cell redox homeostasis, thioredoxins were highly enriched, including THIOREDOXIN F-TYPE1 in the short-term activation of carbon metabolism necessary for growth under short-day conditions (Naranjo et al., 2016). These observations are indicative of ROS changes in the cell. Direct-measure oxido-reduction changes were earlier correlated with sensory plastid perturbation in floral stems of msh1 mutants (Virdi et al., 2016) and in msh1 hemicomplementation tests (Xu et al., 2012; Virdi et al., 2016). Oxidative stress can elicit plastid signaling events accompanied by calcium activation of calmodulins to trigger signaling cascades and influence nuclear stress response (Stael et al., 2012; Guo et al., 2016). We detected sensory cell-specific gene expression changes in calcium-mediated signaling pathways, including numerous calmodulins and rapid alkalization factor-like signal elements. Rapid alkalization factor elements play a role in cellular signaling under stress in Arabidopsis (Atkinson et al., 2013). MAPK phosphatase2 (AT3G06110), a positive regulator of the cellular response to oxidant challenge (Jin and Ellis, 2007), and WRKY DNA-binding protein 25 (AT2G30250), involved in abiotic stress response (Jiang and Deyholos, 2009), were also altered in the translatome. These data support the participation of ROS- and calcium-mediated stress signaling in sensory plastid-containing cells. MSH1 depletion was associated with altered expression of members of the circadian clock mechanism (LHY, GI, FT, SPA1, PRR9, CCA1, TCP11, and PRR5), several of which were similarly prominent in msh1-associated epigenetic memory response (X. Yang, R. Sanchez, H. Kundariya, Y. Wamboldt, J. Barreras, and S. Mackenzie, unpublished data). This intersection with earlier independent studies of the msh1 phenomenon suggests that MSH1 depletion primes a programmed signaling response pattern within sensory plastid-containing cells, where it is precisely localized. The identified networks provide a model for the highly pleiotropic msh1 phenotypes that emerge. The most significant category identified by GO enrichment analysis was “nucleosome assembly” and its related categories of “chromatin silencing,” “production of siRNA,” and “mRNA splicing, via spliceosome” (Fig. 5, A–C and E). These functional categories may link to the stress-related, genome-wide methylation changes that have been reported in msh1 mutants (Virdi et al., 2015) and with subsequent transgenerational memory (X. Yang, R. Sanchez, H. Kundariya, Y. Wamboldt, J. Barreras, and S. Mackenzie, unpublished data). Numerous histone genes were up-regulated in response to MSH1 depletion, including histones HTB1, HTB11, HTB4, HTB4, HTB9, HTB2, HIS4, a “winged-helix DNA-binding transcription factor family protein” (H1.1), and a high-mobility group A (HMGA) protein. H1.1 is a linker histone important for chromatin dynamics (Kotliński et al., 2017). There were also three HISTONE3-RELATED (HTR) genes, HTR4 (AT4g40030), HTR5, and HTR8, that encode H3.3, required for deposition or maintenance of DNA methylation over gene bodies (Wollmann et al., 2017). Specifically, H3.3 knockdown results in loss of methylation over gene bodies, and its enrichment near gene transcription start sites is positively associated with transcription (Stroud et al., 2012; Wollmann et al., 2012). SUVR5, which binds DNA to repress the expression of a subset of stimulus-response genes (Caro et al., 2012), was reduced in expression in msh1, perhaps factoring into the observed unleashing of several stimuli responders. ARGONAUTE4, involved in siRNA-mediated gene silencing (Zilberman et al., 2004), was also downregulated (Fig. 5, C and D). The expression levels of other members of the methylation and siRNA production machinery were altered following MSH1 depletion, including MET1, DME, and RDR2 (Fig. 5D). These data provide a unique profile of epigenomic reprogramming, and its cell-specific induction, when aligned with the marked changes in genome-wide methylation that are documented in msh1 (Virdi et al., 2015) and its heritable memory (X. Yang, R. Sanchez, H. Kundariya, Y. Wamboldt, J. Barreras, and S. Mackenzie, unpublished data). Figure 5. Open in new tabDownload slide Cell-specific influence on chromatin remodeling, DNA methylation, and calcium signaling gene expression in msh1 T-DNA mutants based on translatome data. Panels show gene expression changes in nucleosome assembly (A), chromatin silencing (B), production of siRNA (C), methylation machinery (D), mRNA splicing (E), and calcium-mediated signaling (F). Gene expression changes are measured by log2 fold-change (log2FC). Figure 5. Open in new tabDownload slide Cell-specific influence on chromatin remodeling, DNA methylation, and calcium signaling gene expression in msh1 T-DNA mutants based on translatome data. Panels show gene expression changes in nucleosome assembly (A), chromatin silencing (B), production of siRNA (C), methylation machinery (D), mRNA splicing (E), and calcium-mediated signaling (F). Gene expression changes are measured by log2 fold-change (log2FC). To better understand how sensory plastid-containing cells respond to induced stress, we divided the differentially translated genes into up-regulated and down-regulated subgroups, revealing a much larger number of transcription factors to be activated (120) than down-regulated (38), with basic helix-loop-helix, C2H2, AP2-EREBP, and MYB families most prominent (Fig. 6A). Globally, nonredundant GO category analysis showed that MSH1-depleted cells up-regulate pathways for chromatin remodeling and stress responses, including nucleosome assembly, chromatin assembly or disassembly, heterochromatin organization, response to hormones, and response to oxidative stress and other stresses like cold, salt, drought, and biotic factors. Meanwhile, down-regulated networks involved predominantly the regulation of cell shape, cell wall organization, and secondary metabolism (Supplemental Fig. S5, A and B), consistent with the reduced plant growth phenotype of msh1 that counterbalances enhanced stress response (Xu et al., 2011, 2012). The observed changes in gene expression networks within sensory plastid-containing cells were accompanied by changes in stress-related small molecules. Visualization of ROS species (H2O2 and O2 -) in msh1 mutants showed a higher concentration proximal to vascular tissues (Fig. 6, B and C), which likely accounts for generally higher measurements of ROS in MSH1-RNAi lines (Xu et al., 2012), and with oxidative stress responses observed in gene expression analyses. Figure 6. Open in new tabDownload slide Phenotypes mediated in msh1 by sensory plastid perturbation. A, Graph of transcription factor changes. B, Differential accumulation of H2O2 in floral stem transversal cuts from msh1 and wild-type (WT) control by nitro-tetrazolium blue chloride (NBT) staining. Arrows indicate areas with differential staining relative to the wild type. C, Accumulation of O2 − in transversal cuts of floral stems from msh1 and wild-type control by 3,3′-diaminobenzidine (DAB) staining. Arrows indicate areas with differential staining relative to the wild type. Figure 6. Open in new tabDownload slide Phenotypes mediated in msh1 by sensory plastid perturbation. A, Graph of transcription factor changes. B, Differential accumulation of H2O2 in floral stem transversal cuts from msh1 and wild-type (WT) control by nitro-tetrazolium blue chloride (NBT) staining. Arrows indicate areas with differential staining relative to the wild type. C, Accumulation of O2 − in transversal cuts of floral stems from msh1 and wild-type control by 3,3′-diaminobenzidine (DAB) staining. Arrows indicate areas with differential staining relative to the wild type. DISCUSSION The sensory plastid is distinguished from mesophyll chloroplast not only by size and morphology but proteome features. We found significant overlap for proteins related to PSI and PSII assembly, as well as components of the electron transport chain important for photosynthesis, but also markedly distinctive features for stress-related and metabolic pathway components. Therefore, sensory plastids may be uniquely equipped to trigger stress and developmental response by the plant. Studies of the reticulata mutant phenotypes, characterized by changes in vascular plastid behavior (Kinsman and Pyke, 1998), show that mesophyll chloroplast functions are influenced by, and responsive to, vascular plastid signals (Lundquist et al., 2014). The reticulata phenotype is recognized by leaf patterning with distinctive venation due to altered signaling between vascular plastids and mesophyll chloroplasts. The phenotype can be conditioned by several distinct loci, revealing vascular plastid-specific proteins that serve to validate our own data for sensory plastid-specific proteome components. The reticulata phenomenon invites an intriguing model for systemic stress signal transmission in plant systems, initiating with vascular tissue-localized signals that emanate to alter chloroplast behavior (Lundquist et al., 2014). A component of this signaling appears to involve phosphoenolpyruvate import to sensory plastids, which can be complemented by exogenous application of aromatic amino acids (Staehr et al., 2014). While both “sensory” plastids and mesophyll chloroplast data sets contain the category for response to oxidative stress (Fig. 2, E and F), our study lends support to localized plastid-derived stress responses. It is not clear whether the epidermal and vascular organelles are a singular population in proteome composition or comprise distinct organelle types. Analysis of the tobacco trichome showed that differentiation of sensory plastid relative to chloroplast is precise and not defined by meristematic origin. The epidermal plastid contains thylakoid grana that are already visible in L1 undifferentiated meristem cells (Charuvi et al., 2012). This is the case for plastids within the L3 layer that derives the vascular tissues as well. Earlier studies have suggested that epidermal plastids are nonphotosynthetic (Charuvi et al., 2012), implying that assembly and granal stacking of the thylakoid membrane reflects a functional electron transport chain that may lack other components of the photosynthetic apparatus. However, as shown in this study, purified Arabidopsis epidermal and vascular plastid fractions autofluoresce and contain chlorophyll. Our analysis of sensory plastid proteome data did not reveal key components of photosynthesis to be lacking, suggesting that sensory plastids may contain a full complement of the photosynthetic apparatus. Consequently, it is possible that sensory plastids are photosynthetically competent but are additionally equipped with stress and metabolic signaling mechanisms that specialize their functions. Our study showed the chloroplast to be less complex in proteome than previously estimated (http://ppdb.tc.cornell.edu/). This observation implies that some plastid-associated stress response features may have been misattributed in earlier studies to the chloroplast due to difficulties in fractionation of the two organelle types. It is possible that retrograde signaling circuits may differ across tissue types and that crop engineering efforts could be refined to target the correct plastid type, therefore avoiding undesired effects. Observations detailed here and in earlier studies are consistent with the hypothesis that the MSH1 system, presumably one of several that appear to emanate from the sensory plastid, plays a role in plant adaptation. Localization within the epidermis, vascular parenchyma, and bundle sheath cells likely serves to accommodate environmental sensing and signaling. MSH1 disruption is sufficient to trigger global stress response and developmental changes in plants, an effect that requires two generations to fully amplify (Shao et al., 2017). This global stress genomic posture, once achieved, includes extreme methylation repatterning that, in subsequent generations, evokes heritable nongenetic memory (Xu et al., 2012; Virdi et al., 2015; X. Yang, R. Sanchez, H. Kundariya, Y. Wamboldt, J. Barreras, and S. Mackenzie, unpublished data). To understand the msh1 global stress behavior and derived memory, we characterized translatome features within cells that contain sensory plastids under wild-type and msh1 conditions. Pathways identified within the translatome are presumed to represent at least some of the initial responses to MSH1 depletion. Networks for calcium signaling, oxidative stress, nucleosome and chromatin remodeling, and circadian clock were prominent sensory plastid-proximal responses. These outcomes appear strikingly resonant with studies of msh1 memory, the state that persists with MSH1 restoration following RNAi knockdown (Xu et al., 2012; X. Yang, R. Sanchez, H. Kundariya, Y. Wamboldt, J. Barreras, and S. Mackenzie, unpublished data). The memory phenotype involves DNA methylome repatterning, with altered expression of circadian clock, ABA, and oxidative stress responses (X. Yang, R. Sanchez, H. Kundariya, Y. Wamboldt, J. Barreras, and S. Mackenzie, unpublished data). The vascular circadian network is known to exert robust regulatory dominance over neighboring mesophyll cells, with transcriptional and physiological consequences (Endo et al., 2014). The msh1 T-DNA-null mutant produces a significant range in phenotype intensity and contains a much broader, more variable, and more extreme effect on gene expression networks (Shao et al., 2017) than we see in the memory phenotype. We postulate that cell-specific sensory plastid effects serve to trigger a system-wide plant response, with the intensity and range of phenotype displayed by each msh1 mutant plant conditioned by downstream factors, and with a central core of the phenomenon retained in heritable memory. Aspects of the MSH1 system appear to intersect with findings by other groups. Perturbation of plastids residing within epidermal and vascular parenchyma cells alters ABA- (Endo et al., 2008; Gorecka et al., 2014) and auxin-related (Procko et al., 2016) plant stress behaviors. Salt (Veley et al., 2012) and drought (Endo et al., 2008) stress pathways are triggered within these plastids, as is high-light response (Galvez-Valdivieso et al., 2009). Selection for second-site suppressors of the REPRESSOR OF SILENCING1 (ROS1) locus, which participates in cytosine demethylation (Shen et al., 2009), identifies the CHLOROPHYLL A/B BINDING PROTEIN UNDEREXPRESSED1 (CUE1) gene, which encodes a phosphoenolpyruvate translocator on the envelope membrane of vascular sensory plastids. This observation appears to link vascular plastid functions and epigenetic changes within the nucleus, influenced by availability of aromatic amino acids (Shen et al., 2009). The reports by other groups provide evidence that perturbation of sensory plastids results in ABA-mediated stress responses in the plant, as well as nuclear epigenetic changes. The msh1 mutant displays enhanced tolerance not only to drought (Virdi et al., 2016), but to heat (Shedge et al., 2010) and high light (Xu et al., 2011) as well. MSH1 suppression alters expression of the ABA pathway and methylation repatterning of genes involved in the regulation of ABA (Virdi et al., 2015; X. Yang, R. Sanchez, H. Kundariya, Y. Wamboldt, J. Barreras, and S. Mackenzie, unpublished data). Collectively, these studies support a model of sensory plastids as distinctly differentiated organelles that function in environmental sensing (Fig. 7). Translatome features in sensory plastid cells under perturbation suggest the involvement of ROS, calcium, and MAPK in this process, components of plastid stress signaling (Stael et al., 2012; Guo et al., 2016). Under natural conditions, MSH1 undergoes transcriptional suppression in response to stress and, through its depletion, elicits strongly pleiotropic plant responses (Xu et al., 2011, 2012). It will be interesting to learn whether this type of stress responsiveness is observed for other sensory plastid-specific protein genes. The MSH1 protein is presumed to have been mitochondrially localized early in its evolution (Abdelnoor et al., 2006), subsequently expanded in targeting to the plastid (Xu et al., 2011) and acquiring gene promoter features supporting spatiotemporal and stress-responsive regulation (Virdi et al., 2016). These components of the gene’s evolution appear fundamental to its neofunctionalization for triggering programmed plant stress and heritable memory behaviors, and this evolutionary scheme may prove instructive in modeling the origins of other sensory plastid-specific proteins. Figure 7. Open in new tabDownload slide Proposed model for plastid-specific modulation of chromatin remodeling and stress response pathways mediated by MSH1. We postulate that experimental or stress-associated depletion of MSH1 from the thylakoid/nucleoid site induces ROS/redox changes and release of calcium from the plastid, binding calmodulins (CaMs) to activate calmodulin-binding protein kinases (CBKs) and protein kinase C to activate MAPKs via phosphorylation. Chromatin remodeling events may unleash a localized stress response, involving changes in translation of stress-related transcription factors, ABA regulators, and circadian clock components. A subset of methylome changes that accompany this process display transgenerational heritability for the msh1 memory effect. Emboldened gene classes reflect observed gene expression changes in the sensory plastid translatome data set. Figure 7. Open in new tabDownload slide Proposed model for plastid-specific modulation of chromatin remodeling and stress response pathways mediated by MSH1. We postulate that experimental or stress-associated depletion of MSH1 from the thylakoid/nucleoid site induces ROS/redox changes and release of calcium from the plastid, binding calmodulins (CaMs) to activate calmodulin-binding protein kinases (CBKs) and protein kinase C to activate MAPKs via phosphorylation. Chromatin remodeling events may unleash a localized stress response, involving changes in translation of stress-related transcription factors, ABA regulators, and circadian clock components. A subset of methylome changes that accompany this process display transgenerational heritability for the msh1 memory effect. Emboldened gene classes reflect observed gene expression changes in the sensory plastid translatome data set. MATERIALS AND METHODS Plant Materials Arabidopsis (Arabidopsis thaliana) plants were grown in a controlled-environment walk-in chamber at 12 h day length and 22°C. The Col-0 line used in this study for transformation and control has been DNA sequenced and confirmed to represent the reference genome (Shao et al., 2016). Col-0 msh1 line complemented with MSH1P::MSH1::GFP was derived previously (Xu et al., 2011). The msh1 T-DNA mutant (SAIL-877-F01) was obtained from The Arabidopsis Stock Center (http://www.arabidopsis.org/). Col-0 transgenic plants expressing MSH1P::FLAG::RPL18 were generated in both msh1 homozygous mutant (SAIL-877-F01) and wild-type backgrounds with the cloned MSH1 sequence described by Xu et al. (2011) and with transformation by the floral-dip method (Clough and Bent, 1998). Positive transgenic lines were screened for FLAG tag by immunoblotting methods. Stable tobacco transgenic lines were generated by introducing the MSH1P::SSU-MSH1::GFP fusion construct (Xu et al., 2011) to leaf discs and regenerating transgenic plantlets as described (Virdi et al., 2016). Tobacco plants were grown under standard greenhouse conditions, and tissues were excised for confocal microscopy, with nontransgenic isolines as control. Plastid Extraction Crude plastid preparations from floral stems were carried out in extraction buffer (20 mm Tricine-NaOH, pH 8.4, 300 mm sorbitol, 10 mm KCl, 10 mm EDTA, 0.25% [w/v] bovine serum albumin, 4.5 mm Na-ascorbate, and 5 mm l-Cys). Samples were centrifuged at 2,000g for 5 min at 4°C and supernatants recovered and centrifuged at 4,000g for 10 min at 4°C. Plastid pellets were resuspended in wash buffer (20 mm HEPES, 300 mm sorbitol, 10 mm KCl, 2.5 mm EDTA, and 5 mm MgCl2), recovered by centrifugation at 5,000g for 10 min at 4°C and resuspended again in 5 mL of wash buffer for sorting. FACS Plastids were sorted on a BD FACS Aria II system. GFP and chlorophyll autofluorescence were detected with a 488-nm blue laser. The GFP signal was detected with a 505LP and 530/30BP filter combination and chlorophyll autofluorescence detected with a 670LP and 685/35BP filter combination. The sample flow rate was between 9,000 and 12,000 events/s in buffer containing 20 mm HEPES, 300 mm sorbitol, 10 mm KCl, 2.5 mm EDTA, and 5 mm MgCl2. The experiment was performed twice. For each experiment, four independent sortings were combined. Each independent sorting required the floral stems of 36 plants. Protein Extraction and Mass Spectrometry Plastids were lysed by addition of RapiGest, a cleavable, mass spectrometry-friendly detergent (Waters). The samples were then reduced with 5 mm DTT at 80°C, cooled, and alkylated with 3× molar excess of iodoacetamide. The iodoacetamide was quenched with additional DTT before digestion with 500 ng trypsin overnight at 37°C. The detergent was cleaved by the addition of trifluoroacetic acid and incubation at 37°C. Insoluble components of the detergent were carefully removed; samples were partially dried in a Savant SpeedVac concentrator SVC100H (SpectraLab Scientific) and then dissolved into mass spectrometry sample loading buffer (2.5% [v/v] acetonitrile and 0.1% [v/v] formic acid) according to yield. Samples were spun at 21,000g for 10 min before equal amounts were analyzed by nanoLC-MS/MS using a 2-h gradient on a 0.075 mm × 250 mm C18 Waters CSH column feeding into a Q-Exactive HF mass spectrometer. Proteomics All MS/MS samples were analyzed with Mascot (Matrix Science; version 2.5.1). Mascot was set to search NCBInr_20160821database (Arabidopsis thaliana, 94,081 entries), and cRAP_20150130 database (117 entries) assuming the digestion enzyme trypsin. Mascot was searched with a fragment ion mass tolerance of 0.060 D and a parent ion tolerance of 10.0 PPM. Deamidation of Asn and Gln, oxidation of Met, and carbamidomethyl of Cys were specified in Mascot as variable modifications. Scaffold (version Scaffold_4.7.2; Proteome Software) was used to validate MS/MS-based peptide and protein identifications. Peptide identifications were accepted if they could be established at greater than 80% probability by the Peptide Prophet algorithm (Keller et al., 2002) with Scaffold delta-mass correction. Protein identifications were accepted if they could be established at greater than 99% probability and contained at least two identified peptides. Protein probabilities were assigned by the Protein Prophet algorithm (Nesvizhskii et al., 2003). Proteins that contained similar peptides and could not be differentiated based on MS/MS analysis alone were grouped to satisfy the principles of parsimony. Proteins sharing significant peptide evidence were grouped into clusters. Peptide hits were annotated using UniProt (http://www.uniprot.org), and TAIR ID numbers were retrieved and retained only if present in two independent plastid sample replicates. To increase stringency, hits were filtered and retained only if predicted to be chloroplast-localized or present in either the PPDB (Sun et al., 2009) and/or the At_CHLORO (Bruley et al., 2012) databases. For functional categorization, we used unique representative ID numbers rather than gene families. Confocal Microscopy Confocal laser-scanning microcopy of sorted plastids and plant tissues was performed on a Nikon A1 confocal laser-scanning microscope mounted on the Nikon Eclipse 90 upright compound microscope. Images were acquired using a Nikon NIS Elements version 4.20. GFP (excitation, 488 nm; emission, 500–550 nm) and chlorophyll autofluorescence (excitation, 640.6 nm; emission, 663–738 nm). Alternatively, a Zeiss LSM 510 laser-scanning microscope was used with similar excitation/emission parameters with the corresponding Zen software. Transmission Electron Microscopy FACS-sorted plastids were fixed in 2.5% (v/v) glutaraldehyde in 0.05 m sodium cacodylate (pH 7.4), and transmission electron microscopy was carried out with a Hitachi H7500-I microscope and standard procedures at the University of NE Center for Biotechnology Microscopy Facility. Area measurements of plastid photographs were made with the ImageJ2 program (Rueden et al., 2017). Western-Blot Analysis Plant tissues were ground in liquid nitrogen and total proteins extracted in lysis buffer (50 mm sodium phosphate buffer, pH 7.0, 10 mm EDTA, 1% [v/v] Triton, 0.1% [w/v] SDS, 1× protease inhibitor, and β-mercaptoethanol) for 75 min at 4°C. Lysates were centrifuged at 20,000g for 1 h and protein samples were separated using the NuPAGE system (Invitrogen). Proteins were transferred to nitrocellulose membranes (Optitran; Whatman) for incubation in 1× PBS blocking buffer (137 mm NaCl, 2.7 mm KCl, 8 mm Na2HPO4, and 2 mm KH2PO4), 3% (w/v) bovine serum albumin, and 1% (v/v) Tween 20 for 1 h at room temperature, then for 1 h at room temperature with anti-FLAG antibody. After washing, membranes were incubated with the respective secondary antibody, and immunodetection performed with the Alkaline Phosphatase system (Roche Diagnostics). TRAP Polysomal mRNA from MSH1-specific cells was obtained by ribosome purification using the TRAP method essentially as reported by Reynoso et al. (2015). In brief, floral stem tissues were ground in liquid nitrogen and stored at −80°C. One hundred milliliters lysis buffer (0.2 m Tris, pH 9.0, 0.2 m KCl, 0.025 m EGTA, 0.035 m MgCl2, 1% [v/v] detergent mix [20% (w/v) polyoxyethylene (23) lauryl ether, 20% (v/v) Triton X-100, 20% (v/v) octylphenyl-polyethylene glycol, 20% (v/v) polyoxyethylene sorbitan monolaurate 20], 1% (w/v) DOC, 1% (v/v) PTE, 5 mm DTT, 1 mm PMSF, 50 µg/mL cycloheximide, 50 µg/mL chloramphenicol) were added to 50 mL of ground tissue. Then, 1 to 2 mL of anti-FLAG agarose beads (EZview Red ANTI-FLAG M2 Affinity Gel) was added per sample and incubated overnight at 4°C with gentle rotation. Samples were washed three times with wash buffer (0.2 m Tris, pH 9.0, 0.2 m KCl, 0.025 m EGTA, 0.035 m MgCl2, 5 mm DTT, 1 mm PMSF, 50 μg/mL cycloheximide, and 50 μg/mL chloramphenicol). Elution of samples was performed with wash buffer containing 24 µL FLAG peptide, 2 µL RNase inhibitor, and 2 µL RNase A inhibitor. Final RNA extraction was performed with the standard Trizol method (Invitrogen) by adding 1 µL of glycogen. RNA-Seq and Analysis Transcriptome and translatome for msh1 samples were sequenced by BGI-Tech on the HiSeq 4000 analyzer (Illumina) with paired-end 100-bp option for read length. Alignments were performed using RUM 2.0.4 (−min-length 70 parameters; Grant et al., 2011), retaining only uniquely mapped reads. The read count data were generated from the SAM files with QoRTs software package at −minMAPQ 25 flag (chloroplast and mitochondria reads were dropped; Hartley and Mullikin, 2015). DESeq2 (Love et al., 2014) was used for gene count normalization and to identify differentially expressed genes (FDR < 0.05, |log2FC| > 0.5). GO enrichment analyses were performed with the Database for Annotation, Visualization and Integrated Discovery (DAVID) v6.8, applying the Benjamini-Hochberg method for multiple testing adjustment of P values. Custom R scripts were used to generate heat maps. EnrichmentBrowser R package was applied for Network Based Enrichment Analysis, and the R package “neat” 705 version 1.1.1 was used for Network Enrichment Analysis Test. Accession Numbers Sequence data from this article have been submitted to the GenBank data libraries under accession number GSE114199. Supplemental Data The following supplemental materials are available. Supplemental Figure S1. Assessment of the plastid sorting technique. Supplemental Figure S2. Comparative proteomes of sensory plastids and chloroplasts relative to public database information. Supplemental Figure S3. Proteome distinctions between the two plastid types. Supplemental Figure S4. Translatome profiling. Supplemental Figure S5. Most significant GO biological processes associated with the cell-specific translatome response in msh1. Supplemental Table S1. List of reticulata-associated genes (Lundquist et al., 2014) found in the “sensory” plastid proteome and absent in chloroplast proteome. Supplemental Data Set 1. List of proteome outputs for chloroplast and sensory plastid. Supplemental Data Set 2. msh1 translatome and msh1 transcriptome. Supplemental Data Set 3. GO categories for translatome and transcriptome. Dive Curated Terms The following phenotypic, genotypic, and functional terms are of significance to the work described in this paper: CARBAMOYL CHEBI: CHEBI:33100 CCA1 Gramene: AT2G46830 CCA1 Araport: AT2G46830 dme Gramene: AT5G04560 dme Araport: AT5G04560 GLUTATHIONE CHEBI: CHEBI:16856 HEPES CHEBI: CHEBI:46756 HIS4 Gramene: AT2G28740 HIS4 Araport: AT2G28740 HMGA Gramene: AT1G14900 HMGA Araport: AT1G14900 HTB1 Gramene: AT1G07790 HTB1 Araport: AT1G07790 HTB11 Gramene: AT3G46030 HTB11 Araport: AT3G46030 HTB2 Gramene: AT5G22880 HTB2 Araport: AT5G22880 HTB4 Gramene: AT5G59910 HTB4 Araport: AT5G59910 HTB9 Gramene: AT3G45980 HTB9 Araport: AT3G45980 LHY Gramene: AT1G01060 LHY Araport: AT1G01060 MET1 Gramene: AT5G49160 MET1 Araport: AT5G49160 MSH1 Gramene: AT3G24320 MSH1 Araport: AT3G24320 RDR2 Gramene: AT4G11130 RDR2 Araport: AT4G11130 SPA1 Gramene: AT2G46340 SPA1 Araport: AT2G46340 THIOREDOXIN CHEBI: CHEBI:15033 electron CHEBI: CHEBI:10545 oxidative stress Planteome: TO:0002657 peptide CHEBI: CHEBI:16670 protein CHEBI: CHEBI:36080 proteins CHEBI: CHEBI:36080 salt CHEBI: CHEBI:24866 ACKNOWLEDGMENTS We thank Jose R. Barreras for bioinformatics assistance, Dirk Anderson for assistance in flow cytometry, Michael Naldrett from the proteomics and metabolomics facility of the Center for Biotechnology at University of Nebraska-Lincoln, and Drs. Xiaodong Yang and Vikas Shedge for helpful discussions. We also thank Sarah Pfaff for help with PCR screening. LITERATURE CITED Abdelnoor RV , Yule R, Elo A, Christensen AC, Meyer-Gauen G, Mackenzie SA ( 2003 ) Substoichiometric shifting in the plant mitochondrial genome is influenced by a gene homologous to MutS . Proc Natl Acad Sci USA 100 : 5968 – 5973 Google Scholar Crossref Search ADS PubMed WorldCat Abdelnoor RV , Christensen AC, Mohammed S, Munoz-Castillo B, Moriyama H, Mackenzie SA ( 2006 ) Mitochondrial genome dynamics in plants and animals: convergent gene fusions of a MutS homologue . J Mol Evol 63 : 165 – 173 Google Scholar Crossref Search ADS PubMed WorldCat Atkinson NJ , Lilley CJ, Urwin PE ( 2013 ) Identification of genes involved in the response of Arabidopsis to simultaneous biotic and abiotic stresses . Plant Physiol 162 : 2028 – 2041 Google Scholar Crossref Search ADS PubMed WorldCat Barsan C , Zouine M, Maza E, Bian W, Egea I, Rossignol M, Bouyssie D, Pichereaux C, Purgatto E, Bouzayen M, et al. ( 2012 ) Proteomic analysis of chloroplast-to-chromoplast transition in tomato reveals metabolic shifts coupled with disrupted thylakoid biogenesis machinery and elevated energy-production components . Plant Physiol 160 : 708 – 725 Google Scholar Crossref Search ADS PubMed WorldCat Bruley C , Dupierris V, Salvi D, Rolland N, Ferro M ( 2012 ) AT_CHLORO: A Chloroplast Protein Database Dedicated to Sub-Plastidial Localization . Front Plant Sci 3 : 205 Google Scholar Crossref Search ADS PubMed WorldCat Caro E , Stroud H, Greenberg MVC, Bernatavichute YV, Feng S, Groth M, Vashisht AA, Wohlschlegel J, Jacobsen SE ( 2012 ) The SET-domain protein SUVR5 mediates H3K9me2 deposition and silencing at stimulus response genes in a DNA methylation-independent manner . PLoS Genet 8 : e1002995 Google Scholar Crossref Search ADS PubMed WorldCat Charuvi D , Kiss V, Nevo R, Shimoni E, Adam Z, Reich Z ( 2012 ) Gain and loss of photosynthetic membranes during plastid differentiation in the shoot apex of Arabidopsis . Plant Cell 24 : 1143 – 1157 Google Scholar Crossref Search ADS PubMed WorldCat Clough SJ , Bent AF ( 1998 ) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J 16 : 735 – 743 Google Scholar Crossref Search ADS PubMed WorldCat Enami K , Ozawa T, Motohashi N, Nakamura M, Tanaka K, Hanaoka M ( 2011 ) Plastid-to-nucleus retrograde signals are essential for the expression of nuclear starch biosynthesis genes during amyloplast differentiation in tobacco BY-2 cultured cells . Plant Physiol 157 : 518 – 530 Google Scholar Crossref Search ADS PubMed WorldCat Endo A , Sawada Y, Takahashi H, Okamoto M, Ikegami K, Koiwai H, Seo M, Toyomasu T, Mitsuhashi W, Shinozaki K, et al. ( 2008 ) Drought induction of Arabidopsis 9-cis-epoxycarotenoid dioxygenase occurs in vascular parenchyma cells . Plant Physiol 147 : 1984 – 1993 Google Scholar Crossref Search ADS PubMed WorldCat Endo M , Shimizu H, Nohales MA, Araki T, Kay SA ( 2014 ) Tissue-specific clocks in Arabidopsis show asymmetric coupling . Nature 515 : 419 – 422 Google Scholar Crossref Search ADS PubMed WorldCat Galvez-Valdivieso G , Fryer MJ, Lawson T, Slattery K, Truman W, Smirnoff N, Asami T, Davies WJ, Jones AM, Baker NR ( 2009 ) The high light response in Arabidopsis involves ABA signaling between vascular and bundle sheath cells . Plant Cell 21 : 2143 – 2162 Google Scholar Crossref Search ADS PubMed WorldCat Gorecka M , Alvarez-Fernandez R, Slattery K, McAusland L, Davey PA, Karpinski S, Lawson T, Mullineaux PM ( 2014 ) Abscisic acid signalling determines susceptibility of bundle sheath cells to photoinhibition in high light-exposed Arabidopsis leaves . Philos Trans R Soc Lond B Biol Sci 369 : 20130234 Google Scholar Crossref Search ADS PubMed WorldCat Grant GR , Farkas MH, Pizarro AD, Lahens NF, Schug J, Brunk BP, Stoeckert CJ, Hogenesch JB, Pierce EA ( 2011 ) Comparative analysis of RNA-Seq alignment algorithms and the RNA-Seq unified mapper (RUM) . Bioinformatics 27 : 2518 – 2528 Google Scholar Crossref Search ADS PubMed WorldCat Guo H , Feng P, Chi W, Sun X, Xu X, Li Y, Ren D, Lu C, David Rochaix J, Leister D, et al. ( 2016 ) Plastid-nucleus communication involves calcium-modulated MAPK signalling . Nat Commun 7 : 12173 Google Scholar Crossref Search ADS PubMed WorldCat Hartley SW , Mullikin JC ( 2015 ) QoRTs: a comprehensive toolset for quality control and data processing of RNA-seq experiments . BMC Bioinformatics 16 : 224 Google Scholar Crossref Search ADS PubMed WorldCat Jiang Y , Deyholos MK ( 2009 ) Functional characterization of Arabidopsis NaCl-inducible WRKY25 and WRKY33 transcription factors in abiotic stresses . Plant Mol Biol 69 : 91 – 105 Google Scholar Crossref Search ADS PubMed WorldCat Jin SL , Ellis BE ( 2007 ) Arabidopsis MAPK phosphatase 2 (MKP2) positively regulates oxidative stress tolerance and inactivates the MPK3 and MPK6 MAPKs . J Biol Chem 282 : 25020 – 25029 Keller A , Nesvizhskii AI, Kolker E, Aebersold R ( 2002 ) Empirical statistical model to estimate the accuracy of peptide identifications made by MS/MS and database search . Anal Chem 74 : 5383 – 5392 Google Scholar Crossref Search ADS PubMed WorldCat Kinsman EA , Pyke KA ( 1998 ) Bundle sheath cells and cell-specific plastid development in Arabidopsis leaves . Development 125 : 1815 – 1822 Google Scholar PubMed OpenURL Placeholder Text WorldCat Kotliński M , Knizewski L, Muszewska A, Rutowicz K, Lirski M, Schmidt A, Baroux C, Ginalski K, Jerzmanowski A ( 2017 ) Phylogeny-based systematization of Arabidopsis proteins with histone H1 globular domain . Plant Physiol 174 : 27 – 34 Google Scholar Crossref Search ADS PubMed WorldCat Laterre R , Pottier M, Remacle C, Boutry M ( 2017 ) Photosynthetic trichomes contain a specific Rubisco with a modified pH-dependent activity . Plant Physiol 173 : 2110 – 2120 Google Scholar Crossref Search ADS PubMed WorldCat Love MI , Huber W, Anders S ( 2014 ) Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2 . Genome Biol 15 : 550 Google Scholar Crossref Search ADS PubMed WorldCat Lundquist PK , Rosar C, Bräutigam A, Weber APM ( 2014 ) Plastid signals and the bundle sheath: mesophyll development in reticulate mutants . Mol Plant 7 : 14 – 29 Google Scholar Crossref Search ADS PubMed WorldCat Mustroph A , Zanetti ME, Jang CJH, Holtan HE, Repetti PP, Galbraith DW, Girke T, Bailey-Serres J ( 2009 ) Profiling translatomes of discrete cell populations resolves altered cellular priorities during hypoxia in Arabidopsis . Proc Natl Acad Sci USA 106 : 18843 – 18848 Google Scholar Crossref Search ADS PubMed WorldCat Naranjo B , Diaz-Espejo A, Lindahl M, Cejudo FJ ( 2016 ) Type-f thioredoxins have a role in the short-term activation of carbon metabolism and their loss affects growth under short-day conditions in Arabidopsis thaliana . J Exp Bot 67 : 1951 – 1964 Google Scholar Crossref Search ADS PubMed WorldCat Nesvizhskii AI , Keller A, Kolker E, Aebersold R ( 2003 ) A statistical model for identifying proteins by tandem mass spectrometry . Anal Chem 75 : 4646 – 4658 Google Scholar Crossref Search ADS PubMed WorldCat Pogson BJ , Ganguly D, Albrecht-Borth V ( 2015 ) Insights into chloroplast biogenesis and development . Biochim Biophys Acta 1847 : 1017 – 1024 Procko C , Burko Y, Jaillais Y, Ljung K, Long JA, Chory J ( 2016 ) The epidermis coordinates auxin-induced stem growth in response to shade . Genes Dev 30 : 1529 – 1541 Google Scholar Crossref Search ADS PubMed WorldCat Reynoso MA , Juntawong P, Lancia M, Blanco FA, Bailey-Serres J, Zanetti ME ( 2015 ) Translating ribosome affinity purification (TRAP) followed by RNA sequencing technology (TRAP-SEQ) for quantitative assessment of plant rranslatomes . In JM Alonso and AN Stepanova , eds, Plant Functional Genomics: Methods and Protocols Springer , NY , pp 185 – 207 Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rottet S , Besagni C, Kessler F ( 2015 ) The role of plastoglobules in thylakoid lipid remodeling during plant development . Biochim Biophys Acta 1847 : 889 – 899 Google Scholar Crossref Search ADS PubMed WorldCat Rueden CT , Schindelin J, Hiner MC, DeZonia BE, Walter AE, Arena ET, Eliceiri KW ( 2017 ) ImageJ2: ImageJ for the next generation of scientific image data . BMC Bioinformatics 18 : 529 Google Scholar Crossref Search ADS PubMed WorldCat Shao M-R , Shedge V, Kundariya H, Lehle FR, Mackenzie SA ( 2016 ) Ws-2 introgression in a proportion of Arabidopsis thaliana Col-0 stock seed produces specific phenotypes and highlights the importance of routine genetic verification . Plant Cell 28 : 603 – 605 Google Scholar Crossref Search ADS PubMed WorldCat Shao MR , Kumar Kenchanmane Raju S, Laurie JD, Sanchez R, Mackenzie SA ( 2017 ) Stress-responsive pathways and small RNA changes distinguish variable developmental phenotypes caused by MSH1 loss . BMC Plant Biol 17 : 47 Google Scholar Crossref Search ADS PubMed WorldCat Shedge V , Davila J, Arrieta-Montiel MP, Mohammed S, Mackenzie SA ( 2010 ) Extensive rearrangement of the Arabidopsis mitochondrial genome elicits cellular conditions for thermotolerance . Plant Physiol 152 : 1960 – 1970 Google Scholar Crossref Search ADS PubMed WorldCat Shen J , Ren X, Cao R, Liu J, Gong Z ( 2009 ) Transcriptional gene silencing mediated by a plastid inner envelope phosphoenolpyruvate/phosphate translocator CUE1 in Arabidopsis . Plant Physiol 150 : 1990 – 1996 Google Scholar Crossref Search ADS PubMed WorldCat Staehr P , Löttgert T, Christmann A, Krueger S, Rosar C, Rolčík J, Novák O, Strnad M, Bell K, Weber APM, et al. ( 2014 ) Reticulate leaves and stunted roots are independent phenotypes pointing at opposite roles of the phosphoenolpyruvate/phosphate translocator defective in cue1 in the plastids of both organs . Front Plant Sci 5 : 126 Google Scholar Crossref Search ADS PubMed WorldCat Stael S , Wurzinger B, Mair A, Mehlmer N, Vothknecht UC, Teige M ( 2012 ) Plant organellar calcium signalling: an emerging field . J Exp Bot 63 : 1525 – 1542 Google Scholar Crossref Search ADS PubMed WorldCat Stroud H , Otero S, Desvoyes B, Ramírez-Parra E, Jacobsen SE, Gutierrez C ( 2012 ) Genome-wide analysis of histone H3.1 and H3.3 variants in Arabidopsis thaliana . Proc Natl Acad Sci USA 109 : 5370 – 5375 Google Scholar Crossref Search ADS PubMed WorldCat Sun Q , Zybailov B, Majeran W, Friso G, Olinares PDB, van Wijk KJ ( 2009 ) PPDB, the Plant Proteomics Database at Cornell . Nucleic Acids Res 37 : D969 – D974 Google Scholar Crossref Search ADS PubMed WorldCat van Wijk KJ , Kessler F ( 2017 ) Plastoglobuli: plastid microcompartments with integrated functions in metabolism, plastid developmental transitions, and environmental adaptation . Annu Rev Plant Biol 68 : 253 – 289 Google Scholar Crossref Search ADS PubMed WorldCat Veley KM , Marshburn S, Clure CE, Haswell ES ( 2012 ) Mechanosensitive channels protect plastids from hypoosmotic stress during normal plant growth . Curr Biol 22 : 408 – 413 Google Scholar Crossref Search ADS PubMed WorldCat Virdi KS , Laurie JD, Xu YZ, Yu J, Shao MR, Sanchez R, Kundariya H, Wang D, Riethoven JJM, Wamboldt Y, et al. ( 2015 ) Arabidopsis MSH1 mutation alters the epigenome and produces heritable changes in plant growth . Nat Commun 6 : 6386 Google Scholar Crossref Search ADS PubMed WorldCat Virdi KSS , Wamboldt Y, Kundariya H, Laurie JDD, Keren I, Kumar KRS, Block A, Basset G, Luebker S, Elowsky C, et al. ( 2016 ) MSH1 is a plant organellar DNA binding and thylakoid protein under precise spatial regulation to alter development . Mol Plant 9 : 245 – 260 Google Scholar Crossref Search ADS PubMed WorldCat Wollmann H , Holec S, Alden K, Clarke ND, Jacques PÉ, Berger F ( 2012 ) Dynamic deposition of histone variant H3.3 accompanies developmental remodeling of the Arabidopsis transcriptome . PLoS Genet 8 : e1002658 Google Scholar Crossref Search ADS PubMed WorldCat Wollmann H , Stroud H, Yelagandula R, Tarutani Y, Jiang D, Jing L, Jamge B, Takeuchi H, Holec S, Nie X, et al. ( 2017 ) The histone H3 variant H3.3 regulates gene body DNA methylation in Arabidopsis thaliana . Genome Biol 18 : 94 Google Scholar Crossref Search ADS PubMed WorldCat Xu Y-Z , Arrieta-Montiel MP, Virdi KS, de Paula WBM, Widhalm JR, Basset GJ, Davila JI, Elthon TE, Elowsky CG, Sato SJ, et al. ( 2011 ) MutS HOMOLOG1 is a nucleoid protein that alters mitochondrial and plastid properties and plant response to high light . Plant Cell 23 : 3428 – 3441 Google Scholar Crossref Search ADS PubMed WorldCat Xu Y-Z , Santamaria RdeL, Virdi KS, Arrieta-Montiel MP, Razvi F, Li S, Ren G, Yu B, Alexander D, Guo L, et al. ( 2012 ) The chloroplast triggers developmental reprogramming when mutS HOMOLOG1 is suppressed in plants . Plant Physiol 159 : 710 – 720 Google Scholar Crossref Search ADS PubMed WorldCat Zilberman D , Cao X, Johansen LK, Xie Z, Carrington JC, Jacobsen SE ( 2004 ) Role of Arabidopsis ARGONAUTE4 in RNA-directed DNA methylation triggered by inverted repeats . Curr Biol 14 : 1214 – 1220 Google Scholar Crossref Search ADS PubMed WorldCat Author notes 1 This work was supported by the Bill and Melinda Gates Foundation (OPP1088661). 2 Author for contact: sam795@psu.edu. 3 Senior author. The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Sally A. Mackenzie (sam795@psu.edu). S.A.M. conceptualized the research; S.A.M, K.S.V., and J.B. designed the experiments; J.B., Y.W., R.S., E.W.L., H.K., K.S.V., and C.E. conducted the experiments; S.A.M. and J.B. wrote the manuscript. [OPEN] Articles can be viewed without a subscription. www.plantphysiol.org/cgi/doi/10.1104/pp.18.00804 © 2018 American Society of Plant Biologists. All rights reserved. © The Author(s) 2018. Published by Oxford University Press on behalf of American Society of Plant Biologists. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.
TI - Specialized Plastids Trigger Tissue-Specific Signaling for Systemic Stress Response in Plants
JF - Plant Physiology
DO - 10.1104/pp.18.00804
DA - 2018-10-05
UR - https://www.deepdyve.com/lp/oxford-university-press/specialized-plastids-trigger-tissue-specific-signaling-for-systemic-v7KtSAsEpA
SP - 672
EP - 683
VL - 178
IS - 2
DP - DeepDyve
ER -