Intra-chloroplast proteases: A holistic network view of chloroplast proteolysis.

Different proteases and peptidases are present within chloroplasts and non-photosynthetic plastids to process precursor proteins and to degrade cleaved chloroplast transit peptides and damaged, misfolded, or otherwise unwanted proteins. Collectively, these proteases and peptidases form a proteolysis network, with complementary activities and hierarchies, and build-in redundancies. Furthermore, this network is distributed across the different intra-chloroplast compartments (lumen, thylakoid, stroma, envelope). The challenge is to determine the contributions of each peptidase (system) to this network in chloroplasts and non-photosynthetic plastids. This will require an understanding of substrate recognition mechanisms, degrons, substrate and product size limitations, as well as the capacity and degradation kinetics of each protease. Multiple extra-plastidial degradation pathways complement these intra-chloroplast proteases. This review summarizes our current understanding of these intra-chloroplast proteases in Arabidopsis and crop plants with an emphasis on considerations for building a qualitative and quantitative network view.

Detection and editing of the updated Arabidopsis plastid- and mitochondrial-encoded proteomes through PeptideAtlas.

Arabidopsis (Arabidopsis thaliana) ecotype Col-0 has plastid and mitochondrial genomes encoding over 100 proteins. Public databases (e.g. Araport11) have redundancy and discrepancies in gene identifiers for these organelle-encoded proteins. RNA editing results in changes to specific amino acid residues or creation of start and stop codons for many of these proteins, but the impact of RNA editing at the protein level is largely unexplored due to the complexities of detection. Here, we assembled the nonredundant set of identifiers, their correct protein sequences, and 452 predicted nonsynonymous editing sites of which 56 are edited at lower frequency. We then determined accumulation of edited and/or unedited proteoforms by searching ∼259 million raw tandem MS spectra from ProteomeXchange, which is part of PeptideAtlas (www.peptideatlas.org/builds/arabidopsis/). We identified all mitochondrial proteins and all except 3 plastid-encoded proteins (NdhG/Ndh6, PsbM, and Rps16), but no proteins predicted from the 4 ORFs were identified. We suggest that Rps16 and 3 of the ORFs are pseudogenes. Detection frequencies for each edit site and type of edit (e.g. S to L/F) were determined at the protein level, cross-referenced against the metadata (e.g. tissue), and evaluated for technical detection challenges. We detected 167 predicted edit sites at the proteome level. Minor frequency sites were edited at low frequency at the protein level except for cytochrome C biogenesis 382 at residue 124 (Ccb382-124). Major frequency sites (>50% editing of RNA) only accumulated in edited form (>98% to 100% edited) at the protein level, with the exception of Rpl5-22. We conclude that RNA editing for major editing sites is required for stable protein accumulation.

Detection of the Arabidopsis Proteome and Its Post-translational Modifications and the Nature of the Unobserved (Dark) Proteome in PeptideAtlas.

This study describes a new release of the Arabidopsis thaliana PeptideAtlas proteomics resource (build 2023-10) providing protein sequence coverage, matched mass spectrometry (MS) spectra, selected post-translational modifications (PTMs), and metadata. 70 million MS/MS spectra were matched to the Araport11 annotation, identifying ∼0.6 million unique peptides and 18,267 proteins at the highest confidence level and 3396 lower confidence proteins, together representing 78.6% of the predicted proteome. Additional identified proteins not predicted in Araport11 should be considered for the next Arabidopsis genome annotation. This release identified 5198 phosphorylated proteins, 668 ubiquitinated proteins, 3050 N-terminally acetylated proteins, and 864 lysine-acetylated proteins and mapped their PTM sites. MS support was lacking for 21.4% (5896 proteins) of the predicted Araport11 proteome: the "dark" proteome. This dark proteome is highly enriched for E3 ligases, transcription factors, and for certain (e.g., CLE, IDA, PSY) but not other (e.g., THIONIN, CAP) signaling peptides families. A machine learning model trained on RNA expression data and protein properties predicts the probability that proteins will be detected. The model aids in discovery of proteins with short half-life (e.g., SIG1,3 and ERF-VII TFs) and for developing strategies to identify the missing proteins. PeptideAtlas is linked to TAIR, tracks in JBrowse, and several other community proteomics resources.

The Arabidopsis PeptideAtlas: Harnessing worldwide proteomics data to create a comprehensive community proteomics resource.

We developed a resource, the Arabidopsis PeptideAtlas (www.peptideatlas.org/builds/arabidopsis/), to solve central questions about the Arabidopsis thaliana proteome, such as the significance of protein splice forms and post-translational modifications (PTMs), or simply to obtain reliable information about specific proteins. PeptideAtlas is based on published mass spectrometry (MS) data collected through ProteomeXchange and reanalyzed through a uniform processing and metadata annotation pipeline. All matched MS-derived peptide data are linked to spectral, technical, and biological metadata. Nearly 40 million out of ∼143 million MS/MS (tandem MS) spectra were matched to the reference genome Araport11, identifying ∼0.5 million unique peptides and 17,858 uniquely identified proteins (only isoform per gene) at the highest confidence level (false discovery rate 0.0004; 2 non-nested peptides ≥9 amino acid each), assigned canonical proteins, and 3,543 lower-confidence proteins. Physicochemical protein properties were evaluated for targeted identification of unobserved proteins. Additional proteins and isoforms currently not in Araport11 were identified that were generated from pseudogenes, alternative start, stops, and/or splice variants, and small Open Reading Frames; these features should be considered when updating the Arabidopsis genome. Phosphorylation can be inspected through a sophisticated PTM viewer. PeptideAtlas is integrated with community resources including TAIR, tracks in JBrowse, PPDB, and UniProtKB. Subsequent PeptideAtlas builds will incorporate millions more MS/MS data.

Discovery of AAA+ Protease Substrates through Trapping Approaches.

Proteases play essential roles in cellular proteostasis. Mechanisms through which proteases recognize their substrates are often hard to predict and therefore require experimentation. In vivo trapping allows systematic identification of potential substrates of proteases, their adaptors, and chaperones. This combines in vivo genetic modifications of proteolytic systems, stabilized protease-substrate interactions, affinity enrichments of trapped substrates, and mass spectrometry (MS)-based identification. In vitro approaches, in which immobilized protease components are incubated with isolated cellular proteome, complement this in vivo approach. Both approaches can provide information about substrate recognition signals, degrons, and conditional effects. This review summarizes published trapping studies and their biological outcomes, and provides recommendations for substrate trapping of the processive AAA+ Clp, Lon, and FtsH chaperone proteolytic systems.

Does the Ubiquitination Degradation Pathway Really Reach inside of the Chloroplast? A Re-Evaluation of Mass Spectrometry-Based Assignments of Ubiquitination.

A recent paper in Science Advances by Sun et al. claims that intra-chloroplast proteins in the model plant Arabidopsis can be polyubiquitinated and then extracted into the cytosol for subsequent degradation by the proteasome. Most of this conclusion hinges on several sets of mass spectrometry (MS) data. If the proposed results and conclusion are true, this would be a major change in the proteolysis/proteostasis field, breaking the long-standing dogma that there are no polyubiquitination mechanisms within chloroplast organelles (nor in mitochondria). Given its importance, we reanalyzed their raw MS data using both open and closed sequence database searches and encountered many issues not only with the results but also discrepancies between stated methods (e.g., use of alkylating agent iodoacetamide (IAA)) and observed mass modifications. Although there is likely enrichment of ubiquitination signatures in a subset of the data (probably from ubiquitination in the cytosol), we show that runaway alkylation with IAA caused extensive artifactual modifications of N termini and lysines to the point that a large fraction of the desired ubiquitination signatures is indistinguishable from artifactual acetamide signatures, and thus, no intra-chloroplast polyubiquitination conclusions can be drawn from these data. We provide recommendations on how to avoid such perils in future work.

Proteomics, phylogenetics, and coexpression analyses indicate novel interactions in the plastid CLP chaperone-protease system.

The chloroplast chaperone CLPC1 unfolds and delivers substrates to the stromal CLPPRT protease complex for degradation. We previously used an in vivo trapping approach to identify interactors with CLPC1 in Arabidopsis thaliana by expressing a STREPII-tagged copy of CLPC1 mutated in its Walker B domains (CLPC1-TRAP) followed by affinity purification and mass spectrometry. To create a larger pool of candidate substrates, adaptors, or regulators, we carried out a far more sensitive and comprehensive in vivo protein trapping analysis. We identified 59 highly enriched CLPC1 protein interactors, in particular proteins belonging to families of unknown functions (DUF760, DUF179, DUF3143, UVR-DUF151, HugZ/DUF2470), as well as the UVR domain proteins EXE1 and EXE2 implicated in singlet oxygen damage and signaling. Phylogenetic and functional domain analyses identified other members of these families that appear to localize (nearly) exclusively to plastids. In addition, several of these DUF proteins are of very low abundance as determined through the Arabidopsis PeptideAtlas http://www.peptideatlas.org/builds/arabidopsis/ showing that enrichment in the CLPC1-TRAP was extremely selective. Evolutionary rate covariation indicated that the HugZ/DUF2470 family coevolved with the plastid CLP machinery suggesting functional and/or physical interactions. Finally, mRNA-based coexpression networks showed that all 12 CLP protease subunits tightly coexpressed as a single cluster with deep connections to DUF760-3. Coexpression modules for other trapped proteins suggested specific functions in biological processes, e.g., UVR2 and UVR3 were associated with extraplastidic degradation, whereas DUF760-6 is likely involved in senescence. This study provides a strong foundation for discovery of substrate selection by the chloroplast CLP protease system.

Structure, function, and substrates of Clp AAA+ protease systems in cyanobacteria, plastids, and apicoplasts: A comparative analysis.

ATPases Associated with diverse cellular Activities (AAA+) are a superfamily of proteins that typically assemble into hexameric rings. These proteins contain AAA+ domains with two canonical motifs (Walker A and B) that bind and hydrolyze ATP, allowing them to perform a wide variety of different functions. For example, AAA+ proteins play a prominent role in cellular proteostasis by controlling biogenesis, folding, trafficking, and degradation of proteins present within the cell. Several central proteolytic systems (e.g., Clp, Deg, FtsH, Lon, 26S proteasome) use AAA+ domains or AAA+ proteins to unfold protein substrates (using energy from ATP hydrolysis) to make them accessible for degradation. This allows AAA+ protease systems to degrade aggregates and large proteins, as well as smaller proteins, and feed them as linearized molecules into a protease chamber. This review provides an up-to-date and a comparative overview of the essential Clp AAA+ protease systems in Cyanobacteria (e.g., Synechocystis spp), plastids of photosynthetic eukaryotes (e.g., Arabidopsis, Chlamydomonas), and apicoplasts in the nonphotosynthetic apicomplexan pathogen Plasmodium falciparum. Recent progress and breakthroughs in identifying Clp protease structures, substrates, substrate adaptors (e.g., NblA/B, ClpS, ClpF), and degrons are highlighted. We comment on the physiological importance of Clp activity, including plastid biogenesis, proteostasis, the chloroplast Protein Unfolding Response, and metabolism, across these diverse lineages. Outstanding questions as well as research opportunities and priorities to better understand the essential role of Clp systems in cellular proteostasis are discussed.

The CLP and PREP protease systems coordinate maturation and degradation of the chloroplast proteome in Arabidopsis thaliana.

A network of peptidases governs proteostasis in plant chloroplasts and mitochondria. This study reveals strong genetic and functional interactions in Arabidopsis between the chloroplast stromal CLP chaperone-protease system and the PREP1,2 peptidases, which are dually localized to chloroplast stroma and the mitochondrial matrix. Higher order mutants defective in CLP or PREP proteins were generated and analyzed by quantitative proteomics and N-terminal proteomics (terminal amine isotopic labeling of substrates (TAILS)). Strong synergistic interactions were observed between the CLP protease system (clpr1-2, clpr2-1, clpc1-1, clpt1, clpt2) and both PREP homologs (prep1, prep2) resulting in embryo lethality or growth and developmental phenotypes. Synergistic interactions were observed even when only one of the PREP proteins was lacking, suggesting that PREP1 and PREP2 have divergent substrates. Proteome phenotypes were driven by the loss of CLP protease capacity, with little impact from the PREP peptidases. Chloroplast N-terminal proteomes showed that many nuclear encoded chloroplast proteins have alternatively processed N-termini in prep1prep2, clpt1clpt2 and prep1prep2clpt1clpt2. Loss of chloroplast protease capacity interferes with stromal processing peptidase (SPP) activity due to folding stress and low levels of accumulated cleaved cTP fragments. PREP1,2 proteolysis of cleaved cTPs is complemented by unknown proteases. A model for CLP and PREP activity within a hierarchical chloroplast proteolysis network is proposed.

Exploring the proteome associated with the mRNA encoding the D1 reaction center protein of Photosystem II in plant chloroplasts.

Synthesis of the D1 reaction center protein of Photosystem II is dynamically regulated in response to environmental and developmental cues. In chloroplasts, much of this regulation occurs at the post-transcriptional level, but the proteins responsible are largely unknown. To discover proteins that impact psbA expression, we identified proteins that associate with maize psbA mRNA by: (i) formaldehyde cross-linking of leaf tissue followed by antisense oligonucleotide affinity capture of psbA mRNA; and (ii) co-immunoprecipitation with HCF173, a psbA translational activator that is known to bind psbA mRNA. The S1 domain protein SRRP1 and two RNA Recognition Motif (RRM) domain proteins, CP33C and CP33B, were enriched with both approaches. Orthologous proteins were also among the enriched protein set in a previous study in Arabidopsis that employed a designer RNA-binding protein as a psbA RNA affinity tag. We show here that CP33B is bound to psbA mRNA in vivo, as was shown previously for CP33C and SRRP1. Immunoblot, pulse labeling, and ribosome profiling analyses of mutants lacking CP33B and/or CP33C detected some decreases in D1 protein levels under some conditions, but no change in psbA RNA abundance or translation. However, analogous experiments showed that SRRP1 represses psbA ribosome association in the dark, represses ycf1 ribosome association, and promotes accumulation of ndhC mRNA. As SRRP1 is known to harbor RNA chaperone activity, we postulate that SRRP1 mediates these effects by modulating RNA structures. The uncharacterized proteins that emerged from our analyses provide a resource for the discovery of proteins that impact the expression of psbA and other chloroplast genes.

GFS9 Affects Piecemeal Autophagy of Plastids in Young Seedlings of Arabidopsis thaliana.

Chloroplasts, and plastids in general, contain abundant protein pools that can be major sources of carbon and nitrogen for recycling. We have previously shown that chloroplasts are partially and sequentially degraded by piecemeal autophagy via the Rubisco-containing body. This degradation occurs during plant development and in response to the environment; however, little is known about the fundamental underlying mechanisms. To discover the mechanisms of piecemeal autophagy of chloroplasts/plastids, we conducted a forward-genetics screen following ethyl-methanesulfonate mutagenesis of an Arabidopsis (Arabidopsis thaliana) transgenic line expressing chloroplast-targeted green fluorescent protein (CT-GFP). This screen allowed us to isolate a mutant, gfs9-5, which hyperaccumulated cytoplasmic bodies labeled with CT-GFP of up to 1.0 µm in diameter in the young seedlings. We termed these structures plastid bodies (PBs). The mutant was defective in a membrane-trafficking factor, green fluorescent seed 9 (GFS9), and PB accumulation in gfs9-5 was promoted by darkness and nutrient deficiency. Transmission electron microscopy indicated that gfs9-5 hyperaccumulated structures corresponding to autophagosomes and PBs. gfs9-5 hyperaccumulated membrane-bound endogenous ATG8 proteins, transgenic yellow fluorescent protein (YFP)-ATG8e proteins and autophagosome-like structures labeled with YFP-ATG8e. The YFP-ATG8e signal was associated with the surface of plastids and their protrusions in gfs9-5. Double mutants of gfs9 and autophagy-defective 5 did not accumulate PBs. In gfs9-5, the YFP-ATG8e proteins and PBs could be delivered to the vacuole and autophagic flux was increased. We discuss a possible connection between GFS9 and autophagy and propose a potential use of gfs9-5 as a new tool to study piecemeal plastid autophagy.

Autocatalytic Processing and Substrate Specificity of Arabidopsis Chloroplast Glutamyl Peptidase.

Chloroplast proteostasis is governed by a network of peptidases. As a part of this network, we show that Arabidopsis (Arabidopsis thaliana) chloroplast glutamyl peptidase (CGEP) is a homo-oligomeric stromal Ser-type (S9D) peptidase with both exo- and endo-peptidase activity. Arabidopsis CGEP null mutant alleles (cgep) had no visible phenotype but showed strong genetic interactions with stromal CLP protease system mutants, resulting in reduced growth. Loss of CGEP upregulated the chloroplast protein chaperone machinery and 70S ribosomal proteins, but other parts of the proteostasis network were unaffected. Both comparative proteomics and mRNA-based coexpression analyses strongly suggested that the function of CGEP is at least partly involved in starch metabolism regulation. Recombinant CGEP degraded peptides and proteins smaller than ∼25 kD. CGEP specifically cleaved substrates on the C-terminal side of Glu irrespective of neighboring residues, as shown using peptide libraries incubated with recombinant CGEP and mass spectrometry. CGEP was shown to undergo autocatalytic C-terminal cleavage at E946, removing 15 residues, both in vitro and in vivo. A conserved motif (A[S/T]GGG[N/G]PE946) immediately upstream of E946 was identified in dicotyledons, but not monocotyledons. Structural modeling suggested that C-terminal processing increases the upper substrate size limit by improving catalytic cavity access. In vivo complementation with catalytically inactive CGEP-S781R or a CGEP variant with an unprocessed C-terminus in a cgep clpr2-1 background was used to demonstrate the physiological importance of both CGEP peptidase activity and its autocatalytic processing. CGEP homologs of photosynthetic and nonphotosynthetic bacteria lack the C-terminal prosequence, suggesting it is a recent functional adaptation in plants.

Tissue-type specific accumulation of the plastoglobular proteome, transcriptional networks, and plastoglobular functions.

Plastoglobules are dynamic protein-lipid microcompartments in plastids enriched for isoprenoid-derived metabolites. Chloroplast plastoglobules support formation, remodeling, and controlled dismantling of thylakoids during developmental transitions and environmental responses. However, the specific molecular functions of most plastoglobule proteins are still poorly understood. This review harnesses recent co-mRNA expression data from combined microarray and RNA-seq information in ATTED-II on an updated inventory of 34 PG proteins, as well as proteomics data across 30 Arabidopsis tissue types from ATHENA. Hierarchical clustering based on relative abundance for the plastoglobule proteins across non-photosynthetic and photosynthetic tissue types showed their coordinated protein accumulation across Arabidopsis parts, tissue types, development, and senescence. Evaluation of mRNA-based forced networks at different coefficient thresholds identified a central hub with seven plastoglobule proteins and four peripheral modules. Enrichment of specific nuclear transcription factors (e.g. Golden2-like) and support for crosstalk between plastoglobules and the plastid gene expression was observed, and specific ABC1 kinases appear part of a light signaling network. Examples of other specific findings are that FBN7b is involved with upstream steps of tetrapyrrole biosynthesis and that ABC1K9 is involved in starch metabolism. This review provides new insights into the functions of plastoglobule proteins and an improved framework for experimental studies.

Extreme variation in rates of evolution in the plastid Clp protease complex.

Eukaryotic cells represent an intricate collaboration between multiple genomes, even down to the level of multi-subunit complexes in mitochondria and plastids. One such complex in plants is the caseinolytic protease (Clp), which plays an essential role in plastid protein turnover. The proteolytic core of Clp comprises subunits from one plastid-encoded gene (clpP1) and multiple nuclear genes. TheclpP1 gene is highly conserved across most green plants, but it is by far the fastest evolving plastid-encoded gene in some angiosperms. To better understand these extreme and mysterious patterns of divergence, we investigated the history ofclpP1 molecular evolution across green plants by extracting sequences from 988 published plastid genomes. We find thatclpP1 has undergone remarkably frequent bouts of accelerated sequence evolution and architectural changes (e.g. a loss of introns andRNA-editing sites) within seed plants. AlthoughclpP1 is often assumed to be a pseudogene in such cases, multiple lines of evidence suggest that this is rarely true. We applied comparative native gel electrophoresis of chloroplast protein complexes followed by protein mass spectrometry in two species within the angiosperm genusSilene, which has highly elevated and heterogeneous rates ofclpP1 evolution. We confirmed thatclpP1 is expressed as a stable protein and forms oligomeric complexes with the nuclear-encoded Clp subunits, even in one of the most divergentSilene species. Additionally, there is a tight correlation between amino acid substitution rates inclpP1 and the nuclear-encoded Clp subunits across a broad sampling of angiosperms, suggesting continuing selection on interactions within this complex.

The Plastid and Mitochondrial Peptidase Network in Arabidopsis thaliana: A Foundation for Testing Genetic Interactions and Functions in Organellar Proteostasis.

Plant plastids and mitochondria have dynamic proteomes. Protein homeostasis in these organelles is maintained by a proteostasis network containing protein chaperones, peptidases, and their substrate recognition factors. However, many peptidases, as well as their functional connections and substrates, are poorly characterized. This review provides a systematic insight into the organellar peptidase network in Arabidopsis thaliana We present a compendium of known and putative Arabidopsis peptidases and inhibitors, and compare the distribution of plastid and mitochondrial peptidases to the total peptidase complement. This comparison shows striking biases, such as the (near) absence of cysteine and aspartic peptidases and peptidase inhibitors, whereas other peptidase families were exclusively organellar; reasons for such biases are discussed. A genome-wide mRNA-based coexpression data set was generated based on quality controlled and normalized public data, and used to infer additional plastid peptidases and to generate a coexpression network for 97 organellar peptidase baits (1742 genes, making 2544 edges). The graphical network includes 10 modules with specialized/enriched functions, such as mitochondrial protein maturation, thermotolerance, senescence, or enriched subcellular locations such as the thylakoid lumen or chloroplast envelope. The peptidase compendium, including the autophagy and proteosomal systems, and the annotation based on the MEROPS nomenclature of peptidase clans and families, is incorporated into the Plant Proteome Database.

In Vivo Trapping of Proteins Interacting with the Chloroplast CLPC1 Chaperone: Potential Substrates and Adaptors.

The chloroplast stromal CLP protease system is essential for growth and development. It consists of a proteolytic CLP core complex that likely dynamically interacts with oligomeric rings of CLPC1, CLPC2, or CLPD AAA+ chaperones. These ATP-dependent chaperones are predicted to bind and unfold CLP protease substrates, frequently aided by adaptors (recognins), and feed them into the proteolytic CLP core for degradation. To identify new substrates and possibly also new adaptors for the chloroplast CLP protease system, we generated an in vivo CLPC1 substrate trap with a C-terminal STREPII affinity tag in Arabidopsis thaliana by mutating critical glutamate residues (E374A and E718A) in the two Walker B domains of CLPC1 required for the hydrolysis of ATP (CLPC1-TRAP). On the basis of homology to nonplant CLPB/C chaperones, it is predicted that interacting substrates are unable to be released; that is, they are trapped. When expressed in the wild type, this CLPC1-TRAP induced a dominant visible phenotype, whereas no viable mutants that express CLPC1-TRAP in the clpc1-1 null mutant could be recovered. Affinity purification of the CLPC1-TRAP resulted in a dozen proteins highly enriched compared with affinity-purified CLPC1 with a C-terminal STREPII affinity tag (CLPC1-WT). These enriched proteins likely represent CLP protease substrates or new adaptors. Several of these trapped proteins overaccumulated in clp mutants or were found as interactors for the adaptor CLPS1, supporting their functional relationship to CLP function. Importantly, the affinity purification of this CLPC1-TRAP also showed high enrichment of all CLPP, CLPR, and CLPT subunits, indicating the stabilization of the CLPC to CLP core interaction and providing direct support for their physical and functional interaction.

The Plastoglobule-Localized Metallopeptidase PGM48 Is a Positive Regulator of Senescence in Arabidopsis thaliana.

Plastoglobuli (PG) are thylakoid-associated monolayer lipid particles with a specific proteome of ∼30 PG core proteins and isoprenoid and neutral lipids. During senescence, PGs increase in size, reflecting their role in dismantling thylakoid membranes. Here, we show that the only PG-localized peptidase PGM48 positively regulates leaf senescence. We discovered that PGM48 is a member of the M48 peptidase family with PGM48 homologs, forming a clade (M48D) only found in photosynthetic organisms. Unlike the M48A, B, and C clades, members of M48D have no transmembrane domains, consistent with their unique subcellular location in the PG. In vitro assays showed Zn-dependent proteolytic activity and substrate cleavage upstream of hydrophobic residues. Overexpression of PGM48 accelerated natural leaf senescence, whereas suppression delayed senescence. Quantitative proteomics of PG from senescing rosettes of PGM48 overexpression lines showed a dramatically reduced level of CAROTENOID CLEAVAGE ENZYME4 (CCD4) and significantly increased levels of the senescence-induced ABC1 KINASE7 (ABC1K7) and PHYTYL ESTER SYNTHASE1 (PES1). Yeast two-hybrid experiments identified PG core proteins ABC1K3, PES1, and CCD4 as PGM48 interactors, whereas several other PG-localized proteins and chlorophyll degradation enzymes did not interact. We discuss mechanisms through which PGM48 could possibly accelerate the senescence process.