Skip to main content
NCBI home page
The Plant Cell logoLink to The Plant Cell
. 2023 May 13;35(8):3092–3108. doi: 10.1093/plcell/koad128

FTSH PROTEASE 3 facilitates Complex I degradation through a direct interaction with the Complex I subunit PSST

Abi S Ghifari 1, Aneta Ivanova 2, Oliver Berkowitz 3, James Whelan 4, Monika W Murcha 5,✉,b,c
PMCID: PMC10396367  PMID: 37177987

FTSH PROTEASE 3 directly interacts with the Complex I (CI) (NADH dehydrogenase) subunit PSST, to facilitate the disassembly of the CI matrix arm domain for turnover in the face of oxidative damage.

Abstract

Complex I (CI) (NADH dehydrogenase), the largest complex involved in mitochondrial oxidative phosphorylation, is composed of nuclear- and mitochondrial-encoded subunits. CI assembly occurs via the sequential addition of subdomains and modules. As CI is prone to oxidative damage, its subunits continually undergo proteolysis and turnover. We describe the mechanism by which CI abundance is regulated in a CI-deficient Arabidopsis thaliana mutant. Using a forward genetic approach, we determined that the CI Q-module domain subunit PSST interacts with FTSH PROTEASE 3 (FTSH3) to mediate the disassembly of the matrix arm domain for proteolysis and turnover as a means of protein quality control. We demonstrated the direct interaction of FTSH3 with PSST and identified the amino acid residues required for this interaction. The ATPase function of FTSH3, rather than its proteolytic activity, is required for this interaction, as its mutation was compensated for by a proteolytically inactive form of FTSH3. This study reveals the mechanistic process by which FTSH3 recognizes CI for degradation at amino acid resolution.

IN A NUTSHELL.

Background: Oxidative phosphorylation (OXPHOS) is the central process of aerobic respiration in plant mitochondria. Complex I (CI), the first entry point and largest complex of the OXPHOS pathway, begins the OXPHOS process by oxidizing the high-energy intermediate NADH and transferring electrons to the mobile electron carrier ubiquinone. High redox activity and constant exposure to reactive oxygen species render CI subunits prone to oxidative damage, thereby resulting in a high turnover rate. Despite recent advances in the structural determination of plant CI, how CI degradation and turnover is regulated remains enigmatic.

Question: Which factors determine CI turnover, and how is this process mechanistically regulated?

Findings: Using 2 independent Arabidopsis thaliana EMS mutants generated in a CI-defective background, we show that FTSH3, a mitochondrial matrix-facing inner membrane-bound protease, facilitates the unfolding of CI matrix arm subunits for degradation and turnover. For this function, FTSH3 interacts directly with PSST, a CI matrix arm domain subunit. This interaction is mediated by the ATPase domain of FTSH3 and the N-terminal domain of PSST. Mutations in these domains prevent the interaction between these 2 factors, slowing the turnover rate of matrix arm subunits, and resulting in enhanced CI subunit abundance and activity.

Next steps: We plan to assess the specificity of FTSH3 in regulating the turnover of other OXPHOS complexes, as well as its modulation of the activities of other mitochondrial proteases. By identifying the role of FTSH3 with regard to other OXPHOS subunits and proteases, we can provide a more comprehensive view of how mitochondrial OXPHOS protein turnover is regulated.

Introduction

Mitochondrial ATP is produced through oxidative phosphorylation (OXPHOS) by the combined action of multi-subunit protein complexes on the inner membrane and the mobile electron carriers ubiquinone and cytochrome c. Five distinct protein complexes are located in the mitochondrial inner membrane: NADH:ubiquinone oxidoreductase—Complex I (CI), succinate dehydrogenase—Complex II (CII), cytochrome bc1—Complex III (CIII), cytochrome c oxidase—Complex IV (CIV), and ATP synthase. NADH generated from the TCA cycle is oxidized by CI, followed by the transfer of electrons via mobile carriers through ubiquinone to CIII and cytochrome c to CIV with the reduction of oxygen to water. This process creates a proton gradient necessary for ATP synthesis.

CI is the first site of electron transfer and also the largest complex of the OXPHOS pathway, consisting of 47 to 51 subunits in plants (Soufari et al. 2020; Klusch et al. 2021, 2023; Maldonado et al. 2023). Functionally, plant CI can be divided into 4 modules: the NADH-binding module (N), ubiquinone-binding module (Q), proton-pumping (P) module, and carbonic anhydrase (CA) module. NQ modules form the hydrophilic matrix arm domain that protrudes into the mitochondrial matrix, which contains subunits involved in electron transfer from NADH to ubiquinone. The membrane arm is composed of the proximal (PP) and distal proton-pumping (PD) modules, and the CA module is matrix-facing and anchored to the inner membrane. CI assembly requires a sequential multi-step pathway, which is assisted by various assembly factors to form submodules and assembly intermediates (Schimmeyer et al. 2016; Ivanova et al. 2019; Ligas et al. 2019; Maldonado et al. 2020; Soufari et al. 2020).

The matrix arm domain, which is essential for overall CI activity, contains essential cofactors. These include an FMN prosthetic group covalently bound in a 51 kDa subunit for NADH-binding, a ubiquinone-binding domain formed by the subunits NAD1 (NADH DEHYDROGENASE 1) and PSST, and 8 iron–sulfur (Fe–S) clusters along the NQ modules required for electron transfer (Soufari et al. 2020). Redox reactions and electron transfer generate reactive oxygen species (Hirst and Roessler 2016), which render the matrix arm highly susceptible to oxidative damage (Szczepanowska et al. 2020). Consequently, matrix arm domain subunits have been shown to exhibit some of the highest protein turnover rates compared to the membrane module subunits and to other mitochondrial proteins (Nelson et al. 2013; Li et al. 2017; Szczepanowska et al. 2020). The observed selective turnover of matrix arm domain subunits suggests the presence of a submodule disassembly pathway, allowing for selective proteolysis of damaged subunits and thus conservation of energy.

Previous studies have implicated a number of mitochondrial proteases in in organello protein degradation. Plant mitochondria harbor several classes of proteases that belong to the AAA+ (ATPase-associated with various cellular activity) superfamily, which require ATP for activity (van Wijk 2015; Opalińska and Jańska 2018; Ghifari and Murcha 2022). Mitochondrial AAA+ proteases such as LON1 (long filamentous phenotype-1), caseinolytic protease (CLP), and filamentous temperature sensitive H (FTSH) proteases have all been shown to be involved in the degradation and turnover of OXPHOS complexes (Li et al. 2017; Huang et al. 2020; Petereit et al. 2020; Ivanova et al. 2021). However, in contrast to the well-established cytosolic ubiquitin-proteasome degradation system (Vierstra 2009), the mechanisms of OXPHOS complex disassembly, degradation, and turnover remain poorly understood.

Using a forward genetic screen of Arabidopsis (Arabidopsis thaliana) mutants in the CI-defective complex I assembly factor 1 (ciaf1) mutant background (Ivanova et al. 2019), we previously determined that the inner membrane-bound AAA+ protease FTSH PROTEASE 3 (FTSH3) plays a role in regulating CI activity and abundance by generating and analyzing a revertant named restoration of mitochondrial biogenesis-1 (rmb1) (Ivanova et al. 2021). This revertant contained a mutation in FTSH3 within the ATPase domain (P415L) and displayed a restoration of CI activity and abundance. Biochemical and genetic analysis of rmb1 ciaf1 indicated that FTSH3 is involved in the disassembly of the matrix arm domain. Furthermore, we showed that the ATPase domain of FTSH3, and not the proteolytic domain, was responsible for CI matrix arm disassembly in the CI-defective background. The mutation in FTSH3 within the ATPase domain (P415L) was able to restore CI activity and the abundance of matrix arm domain subunits. This phenotypic restoration was reverted by complementation with FTSH3. Interestingly, complementation was also achieved using a proteolytically inactive form of FTSH3 (FTSH3TRAP), indicating that the proteolytic function of FTSH3 is not responsible for CI abundance and activity; instead, the ATPase function of FTSH3 plays a role in these traits (Ivanova et al. 2021).

However, the mechanism by which FTSH3 regulates CI disassembly was not elucidated in these studies. Here, we determined that FTSH3 is involved in the disassembly of the matrix arm domain of CI via a direct protein–protein interaction with PSST, a 20 kDa subunit located at the interface of the membrane and matrix module (Soufari et al. 2020; Klusch et al. 2021, 2023; Maldonado et al. 2023). We identified and characterized an additional independent ethyl-methane-sulfonate (EMS) revertant of ciaf1 that contains a mutation within the PSST gene (At5g11770) named restoration of mitochondrial biogenesis-2 (rmb2) ciaf1. rmb2 ciaf1 displayed a restoration of CI abundance, activity, and consequently plant growth, as previously observed for the FTSH3 revertant mutant rmb1 ciaf1 (Ivanova et al. 2021). Using a variety of protein–protein interaction assays, we observed a physical interaction between FTSH3 and PSST that could be abolished by rmb1 and/or rmb2 mutations (FTSH3P415L/PSSTS70F). We observed an increased abundance of matrix arm subunits in rmb2 ciaf1, as previously observed in rmb1 ciaf1, in addition to delayed protein turnover rates in both mutants. Therefore, combining genetic, biochemical, and proteomic analyses of 2 independent revertant lines that display restored CI abundance in a CI-defective background, we determined that FTSH3 plays a role in the disassembly of the matrix arm domain of CI via a direct interaction with the CI PSST subunit.

Results

The delayed growth of ciaf1 is restored by a mutation in PSST

To identify regulators of CI biogenesis, we carried out a forward genetic screen using a CI-defective mutant. This mutant is a knockdown mutant of the gene encoding CI assembly factor-1 (CIAF1), encoding a mitochondrial LYR domain-containing protein required for Fe–S cluster insertion into matrix arm domain subunits (Ivanova et al. 2019). T-DNA insertional knockdown mutants of CIAF1 exhibit an almost complete lack of CI due to the stalled assembly of the matrix arm domain and consequently, a severely delayed growth phenotype (Ivanova et al. 2019). This delayed growth phenotype, characteristic of many CI mutants, allowed for a visual screen of revertant mutants showing restored growth.

Here, we characterized an EMS mutant named rmb2 ciaf1 that exhibits a substantial and significant growth restoration compared to ciaf1 (Fig. 1, A and B, Supplemental Fig. S1). rmb2 ciaf1 displayed significantly improved growth parameters, including plant height, number of rosette leaves, and maximum rosette radius compared to ciaf1 throughout the growth cycle (Fig. 1, A and B, Supplemental Fig. S1). Whole-genome sequencing of segregating homozygous rmb2 ciaf1 populations identified the single nucleotide change of C to T at the chromosomal location of the gene designated At5g11770. The mutation corresponds to a single amino acid change from serine to phenylalanine at position 70 (S70F) in PSST, a 20 kDa CI subunit that is located in the matrix arm Q-module in close proximity to the membrane arm submodule (Soufari et al. 2020; Klusch et al. 2021, 2023; Maldonado et al. 2023). PSST, also known as NADH dehydrogenase ubiquinone Fe–S protein-7 (NDUFS7), was shown to be involved in ubiquinone-binding and reduction (Galemou Yoga et al. 2019; Soufari et al. 2020). The S70 residue is located at the N-terminal domain of the mature PSST protein, with the first 64 residues containing the N-terminal mitochondrial targeting peptide (Soufari et al. 2020; Klusch et al. 2021) (Fig. 1C).

Figure 1.

Figure 1.

Open in a new tab

Identification of Arabidopsis rmb2 ciaf1, an EMS-induced revertant restoration of mitochondrial biogenesis-2 (rmb2) mutant in the COMPLEX I ASSEMBLY FACTOR-1 (ciaf1) knockdown background. A) Phenotypic analysis of Col-0, ciaf1 (SALK_143656), and rmb2 ciaf1 grown on agar medium (half-strength Murashige–Skoog basal salts supplemented with 1.5% w/v sucrose) after 14 d (top panels) and in soil after 28 d (bottom panels), showing that developmentally delayed growth in ciaf1 can be partially restored by a mutation in rmb2 ciaf1. Scale bar = 1 cm, B) Growth parameter analyses: plant height, number of rosette leaves, and maximum rosette radius showing restoration of growth and development in rmb2 ciaf1 compared to ciaf1. Asterisks indicate significantly different values (mean ± Sd, n = 20 individual plants, single-factor ANOVA P < 0.05, followed by Tukey’s honestly significant different (HSD) post hoc test). Error bars represent the Sd of replicates are shown (n = 3), and C) Whole-genome sequencing of rmb2 ciaf1 identifying a single-point mutation (TCC-TTC) within the At5g11770 gene, which corresponds to a mutation of position 70 changing serine to phenylalanine (S70F) in the N-terminal region of PSST, a 20 kDa CI Q-module domain subunit.

CI activity and the abundance of individual subunits are restored in rmb2 ciaf1

To confirm that the mutation in PSST is responsible for the observed phenotypic restoration, we performed several crosses and generated complementation lines (Fig. 2A). First, we introduced the S70F mutation into a Col-0 background by filial crossing to produce rmb2 Col-0, resulting in no changes in plant growth compared to Col-0 (Fig. 2A, Supplemental Fig. S1). Second, we generated a complementation line by stably transforming rmb2 ciaf1 with PSST under the control of its native promoter (rmb2pro:PSST) (Fig. 2A). This complementation line exhibited in a reversion back to the small, developmentally delayed growth phenotype of ciaf1 (Fig. 2A, Supplemental Fig. S1), confirming that the PSST gene product is responsible for the observed phenotype in rmb2 ciaf1. We generated plants over-expressing PSST under the control of the cauliflower mosaic virus (CaMV) 35S promoter in the Col-0 background (35S:PSST), which displayed no significant changes in plant growth parameters compared to Col-0 (Fig. 2A, Supplemental Fig. S1).

Figure 2.

Figure 2.

Open in a new tab

Rmb2 ciaf1 exhibits partially restored CI activity and abundance. A) Phenotypic analysis of rmb2 ciaf1, rmb2 in the Col-0 background (rmb2 Col-0), rmb2 complemented with PSST under the control of its native promoter (rmb2pro:PSST), and PSST over-expression line (35S:PSST). Scale bar = 1 cm, B) BN-PAGE followed by in-gel CI activity staining of isolated mitochondria from all lines. The position and molecular weight of the OXPHOS complexes are indicated, C) Activity of CI relative to Col-0, as quantified from in-gel staining in B) (mean ± Sd, n = 3). Significantly different values compared to ciaf1 are indicated (*P < 0.05, **P < 0.001, single-factor ANOVA). Individual data points and error bars representing the Sd of replicates are shown (n = 3), D) Assay of CI activity measured in vitro from isolated mitochondria (mean ± Sd, n = 3). Significantly different values compared to ciaf1 are indicated (*P < 0.05, **P < 0.001, single-factor ANOVA). Individual data points and error bars representing the Sd of replicates are shown (n = 3), E) Immunodetection of isolated mitochondria from all lines, including CI NQ-module subunits (75 kDa, NDUFS4, PSST, highlighted in gray), other CI subunits (B14.7, CA2), other OXPHOS subunits (RISP, ATPβ), and mitochondrial AAA+ proteases (FTSH3, FTSH10, FTSH4). The outer membrane transport protein TOM40 was used as a loading control, and F) Densitometry measurements of immunoblot signals in E) relative to Col-0 for NDUFS4 (18 kDa), PSST (20 kDa), 75 kDa, B14.7, CA2, FTSH3, FTSH10, FTSH4, RISP, and ATPβ normalized to the TOM40 signal as a loading control (mean ± Sd, n = 3). Significantly different values compared to ciaf1 are indicated by asterisks * (single-factor ANOVA followed by Tukey’s HSD post hoc test). Individual data points and error bars representing the Sd of replicates are shown (n = 3).

We isolated mitochondria from all genotypes and measured CI abundance by blue-native polyacrylamide gel electrophoresis (BN-PAGE), followed by an in-gel NADH dehydrogenase activity assay (Schertl and Braun 2015). In-gel CI activity staining indicated a significant increase in CI abundance of approximately 2-fold in rmb2 ciaf1 compared to ciaf1 (Fig. 2, B and C) (P < 0.05, n = 3), whilst the complementation line (rmb2pro:PSST) exhibited in a decrease in CI abundance back to the levels observed in ciaf1 (Fig. 2, B and C). Introduction of rmb2 into the Col-0 background (rmb2 Col-0) and PSST over-expression (35S:PSST) in the Col-0 background did not result in any changes to CI abundance compared to Col-0 (Fig. 2, B and C).

To confirm these results, we performed an in vitro colorimetric plate-based assay to measure CI activity (Huang et al. 2015), which revealed similar changes. A significant decrease of approximately 50% in CI activity was measured in ciaf1 compared to Col-0, which was restored to approximately 80% of wild-type levels in rmb2 ciaf1. Complementation with PSST under its native promoter (rmb2pro:PSST) reverted CI activity back to the level observed in ciaf1 (Fig. 2D).

We performed immunoblotting of individual CI subunits using isolated mitochondria from each genotype. The abundance of N-module subunits NADH dehydrogenase ubiquinone iron-sulfur protein 4 (NDUFS4 or18 kDa) and 75 kDa increased significantly in rmb2 ciaf1 compared to ciaf1 (P < 0.05, n = 3) (Fig. 2, E and F). No significant difference in the abundance of CI membrane arm subunits such as B14.7 and carbonic anhydrase subunit 2 (CA2), CIII subunit Rieske iron–sulfur protein (RISP), or the ATP synthase subunit-β (ATPβ) (P < 0.05, n = 3) was identified in rmb2 ciaf1 compared to ciaf1 (Fig. 2, E and F). Additionally, no significant changes were observed in the abundance of the mitochondrial FTSH proteases, including the matrix-facing FTSH3 and FTSH10 and the intermembrane space (IMS)-facing FTSH4 protease in rmb2 ciaf1 compared to ciaf1 (Fig. 2, E and F). Translocase of the outer membrane 40 (TOM40), an outer membrane mitochondrial protein, was used as a loading control (Fig. 2, E and F).

rmb2 ciaf1 exhibits restored abundances of CI matrix arm subunits

To assess global changes to the proteome of rmb2 ciaf1, we performed high-resolution isoelectric focusing (HiRIEF) liquid chromatography-mass spectrometry (LC-MS) proteomics of rosette leaf tissue (Branca et al. 2014). Analysis of previously published data for ciaf1 indicated a decrease in the abundance of CI NQ-module subunits compared to Col-0 (Fig. 3, A and B, Supplemental Fig. S2) (Ivanova et al. 2019). Analysis of rmb2 ciaf1 revealed similar protein abundance profiles to those observed for rmb1:ciaf1 (Ivanova et al. 2021). rmb2 ciaf1 and rmb1 ciaf1 both exhibited an approximately 1.5-fold increased abundance of matrix arm subunits compared to ciaf1 (Fig. 3, B and C, Supplemental Fig. S2, Supplemental Data Set 1).

Figure 3.

Figure 3.

Open in a new tab

HiRIEF proteomics analysis of ciaf1, rmb1 ciaf1 (mutation in FTSH3P415L), and rmb2 ciaf1 (mutation in PSSTS70F). A) Heatmap showing log2 foldchange of protein abundance for OXPHOS subunits, as measured by HiRIEF proteomics, in ciaf1 compared to Col-0, rmb1 ciaf1, and rmb2 ciaf1, B) Relative abundance (foldchange) of CI subunits in ciaf1, rmb1 ciaf1, and rmb2 ciaf1. CI modules are indicated: N, Q, CA, PP, and PD modules, C) Topographical heatmap of log2 relative protein abundance fitted to the cauliflower CI structure (PDB: 7A23) (Soufari et al. 2020) in different lines. The dark gray regions represent subunits that were not reliably detected. Several matrix arm domain (NQ-module) subunits are indicated, D) Heatmap of log2 transformed foldchange of protein abundance showing that most up-regulated proteins in ciaf1 were present at normal (wild type) levels or down-regulated in rmb1 ciaf1 and rmb2 ciaf1 compared to their ciaf1 genetic background, E) GO enrichment analysis of down-regulated (reduced abundance) and up-regulated (increased abundance) proteins in rmb2:ciaf1 compared to ciaf1. The fill color represents –log10P-value of enrichment after Bonferroni correction. Protein groups with the highest changes are indicated. Significantly reduced: CRK, RLP, white rust resistance, TIR-NBS disease resistance proteins, HSP, CYP, GST, uridine diphosphate (UDP)/UDP glycosyl-transferase (UGT), and WRKY transcription factors. Significantly increased: translocase of inner chloroplastic envelope, twin-arginine translocase, acyl carrier proteins, PSI and PSII proteins, light-harvesting chlorophyll-binding proteins, ribosomal protein large and small, plastid-specific ribosomal protein, mitochondrial intermembrane-space assembly, HSP, mitochondrial carrier family protein, and CI subunits B14.5a and NADH dehydrogenase:ubiquinone subcomplex assembly factor NDUFAF3, and an unidentified CI LYR domain-containing protein.

Proteomic analysis showed that the abundance of stress-responsive proteins in rmb2 ciaf1 and rmb1 ciaf1 returned to normal (wild type) levels compared to ciaf1 (Fig. 3, D and E) (Ivanova et al. 2021). The levels of stress response proteins such as alternative oxidase 1A (AOX1A), cysteine-rich kinases (CRKs), receptor-like proteins (RLPs), toll/interleukin-1 receptor nucleotide-binding site (TIR-NBS) disease-resistant proteins, heat shock proteins (HSPs), cytochrome P450s (CYPs), glutathione-S-transferases (GSTs), UDP-glycosil transferases (UGTs), and Trp-Arg-Lys-Tyr (WRKY) domain-containing transcription factors, which were up-regulated in ciaf1, all decreased in rmb2 ciaf1 (Fig. 3E, Supplemental Data Set 1). The CI defect in ciaf1 and other CI mutants generally corresponds to up-regulated stress-responsive mechanisms (Dutilleul et al. 2003; Meyer et al. 2009; Ivanova et al. 2019); the return to the normal abundance of these proteins in rmb1 ciaf1 and rmb2 ciaf1 likely corresponds to the restoration of CI activity.

FTSH3 and PSST are interacting proteins, and mutations within the ATPase domain of FTSH3 and the N-terminal domain of PSST inhibit this interaction

As rmb1 ciaf1 and rmb2 ciaf1 display similar growth phenotypes, CI abundance and activity, and proteomic profiles, we tested the ability of FTSH3 and PSST to interact. To this end, we carried out co-immunoprecipitation (Co-IP), yeast-2-hybrid (Y2H) interaction assays, and bimolecular fluorescence complementation (BiFC) assays. Immunoprecipitations were carried out using mitochondria isolated from a transgenic line expressing the FLAG-tagged CI subunit B14.7 (B14.7FLAG) (Wang et al. 2012; Klusch et al. 2023). Incubation of digitonin-treated mitochondria with anti-FLAG Sepharose beads resulted in the precipitation of B14.7FLAG, as detected in the Co-IP sample using both anti-B14.7 and anti-FLAG antibodies (Fig. 4A). Immunodetection with antibodies raised against various CI subunits, including PSST, 18 kDa (NDUFS4), 75 kDa, NAD1, CA2, and CAL1, revealed that intact CI was precipitated accordingly (Fig. 4A). Immunodetection with anti-FTSH3 antibodies revealed the enrichment and presence of FTSH3 within the Co-IP sample (Fig. 4A), suggesting an interaction between FTSH3 and CI. No band was detected using an antibody against FTSH10, the partner protein of FTSH3 (Piechota et al. 2010), or FTSH4, an IMS-facing AAA+ protease (Maziak et al. 2021), suggesting the interaction with CI and FTSH3 is unique (Fig. 4A). No other OXPHOS subunits, including CII (SDH1), CIII (RISP), CIV (COX2), and CV (ATPβ), or the outer membrane protein TOM40 were precipitated with B14.7FLAG (Fig. 4A).

Figure 4.

Figure 4.

Open in a new tab

FTSH3 interacts with the CI matrix arm subunit PSST. A) Co-immunoprecipitation of Col-0 and B14.7FLAG mitochondria enriching CI. The presence of immuno-detected proteins in the input supernatant (supern.), first wash (wash-1), second wash (wash-2), and co-immunoprecipitated beads is indicated (left), along with the corresponding molecular weight markers (right), B) Y2H assays testing the protein interaction ability of proteolytically inactive FTSH3TRAP, FTSH10TRAP, and CI subunit PSST, with and without FTSH3P415L and PSSTS70F mutations. Successful diploid mating was identified by growth on the DDO SD medium. Positive interactions were defined by growth on the QDO SD medium. Serial dilutions of diploid yeast are indicated (right). Interactions between SV40 antigen with p53 and Lam were used as positive and negative controls, respectively. No interactions were observed using empty vectors, C) BiFC assays of proteolytically inactive FTSH3TRAP, FTSH10TRAP, and PSST, with and without FTSH3P415L and PSSTS70F mutations. Arabidopsis mesophyll protoplasts transfected with constructs encoding proteolytically inactive FTSH3TRAP/FTSH10TRAP and PSST with either C-terminal or N-terminal half of yellow fluorescent protein (cYFP/nYFP). FTSH3TRAP with wild-type PSST or FTSH10TRAP resulted in reconstituted YFP fluorescence. Combinations involving FTSH3P415L and PSSTS70F, as well as between FTSH10TRAP and PSST, resulted in no YFP signals. YFP fluorescence, chlorophyll (Chl) autofluorescence, bright field, and overlay of YFP and chlorophyll (YFP + Chl) are indicated. Scale bar = 20 μm.

To investigate whether FTSH3 directly interacts with PSST, we performed Y2H interaction assays using a proteolytically inactive form of FTSH3 (FTSH3TRAP). Yeast strains transformed with FTSH3TRAP and PSST were mated and tested for their ability to grow on quadruple dropout medium (QDO) and double dropout medium (DDO). FTSH3TRAP and PSST diploid yeast grew on QDO, indicating a protein–protein interaction (Fig. 4B). A similar mating, but with the rmb1 mutation (P415L) within FTSH3 (FTSH3TRAP/P415L), did not result in any growth on QDO. Similarly, when the rmb2 mutation was introduced within PSST (PSSTS70F), no interaction was observed with FTSH3TRAP (Fig. 4B). As FTSH3 can form hetero-hexamers with FTSH10, we also tested the ability of FTSH10TRAP to interact was PSST. No growth was observed on QDO (Fig. 4B). To test the interactive ability of FTSH3 and FTSH10, FTSH3TRAP, and FTSH10TRAP were mated and grown on QDO, resulting in growth, as expected. To test if the P415L mutation within FTSH3 altered this interaction ability, FTSH3TRAP/P415L and FTSH10TRAP were mated and growth was observed on QDO, suggesting that the P415L mutation within FTSH3 does not alter the ability of these proteins to interact (Fig. 4B). All diploid yeast strains grew on DDO medium, confirming successful mating, alongside the positive (SV40 + p53) and negative (SV40 + Lam) controls (Fig. 4B).

To confirm these protein–protein interactions in planta, we performed BiFC using Arabidopsis protoplasts. Co-transformation of PSST-cYFP and FTSH3TRAP-nYFP resulted in reconstituted fluorescence (Fig. 4C), whereas co-transformation of PSST-cYFP and nYFP-FTSH3TRAP did not (Fig. 4C). Co-transformation of PSST-nYFP and FTSH3TRAP-cYFP also resulted in reconstituted cytosolic fluorescence, indicating protein interactions, whereas co-transformation with mutated PSST (PSSTS70F), FTSH3TRAP/P415L, or both proteins did not (Fig. 4C). Co-transformation of PSST-cYFP with FTSH10TRAP-nYFP did not result in any reconstituted fluorescence, supporting the results obtained in the Y2H assays. Additionally, co-transformation with FTSH3TRAP and FTSH3TRAP/P415L resulted in reconstituted fluorescence when co-transformed with FTSH10TRAP (Fig. 4C).

Mutations within the ATPase domain of FTSH3 and the N-terminal domain of PSST slow the rate of protein turnover for some CI subunits

The disruption of mitochondrial AAA+ proteases has resulted in altered turnover rates for various OXPHOS subunits (Li et al. 2017; Szczepanowska et al. 2020). To assess protein turnover, we carried out stable isotope 15N-labeling of Col-0, ciaf1, rmb1 ciaf1, and rmb2 ciaf1 seedlings, followed by mitochondrial isolation and quadrupole time-of-flight mass spectrometry of soluble matrix fractions, resulting in the detection of protein turnover rates for 832 out of 1,547 mitochondrial proteins (Supplemental Data Set 2). We measured the degradation or turnover rate (KD) of individual proteins based on the foldchange of the protein abundance of 15N-labeled and unlabeled proteins over a period of time using a previously described equation (Li et al. 2017). Analysis of degradation rates (KD) of CI subunits from ciaf1 compared to Col-0 showed overall faster turnover rates for subunits located within the CI matrix NQ-module (Fig. 5A, Supplemental Fig. S3). This is consistent with the previous finding that NQ-module subunits are less abundant in ciaf1 than Col-0 due to their incorrect assembly and maturation (Ivanova et al. 2019). No notable changes in turnover rates were identified for other OXPHOS subunits (Ivanova et al. 2019) (Supplemental Fig. S3).

Figure 5.

Figure 5.

Open in a new tab

Mutations in rmb1 ciaf1 (FTSH3P415L) and rmb2 ciaf1 (PSSTS70F) result in slow turnover rates of CI subunits. A) Topographical heatmap of protein turnover rate KD (d−1) measured from matrix-soluble fraction fitted into the CI structure (PDB: 7A23) (Soufari et al. 2020). Heatmap shows subunits with relatively slow (blue, KD = 0.1) to relatively fast turnover rates (red, KD ≥ 0.9) (Li et al. 2017). Several subunits are indicated. The dark gray region represents subunits that were unreliably detected, B) Comparison of turnover rates KD (d−1) for individual CI subunits in ciaf1, rmb1 ciaf1, and rmb2 ciaf1 showing subunits with relatively slow (KD = 0.1) to relatively fast turnover rates (KD ≥ 0.9). Changes in turnover rates (ΔKD) showing the differences in turnover rates of rmb1 ciaf1 and rmb2 ciaf1 compared to ciaf1, with negative values indicating more stable subunits and positive values indicating less stable subunits. Specific CI modules are indicated: N, Q, CA, PP, and PD modules.

Analysis of turnover rates indicated slower degradation rates (KD) of NQ-module subunits in rmb1 ciaf1 compared to ciaf1, confirming a slowing of disassembly and turnover (Fig. 5A, Supplemental Fig. S3). Comparison of turnover rates (ΔKD), i.e. the difference in turnover rate between EMS mutants rmb1 ciaf1 and rmb2 ciaf1 and the background genotype ciaf1, showed that on average, CI subunits in both rmb1 ciaf1 and rmb2 ciaf1 had ΔKD of −0.12 d−1 compared to ciaf1 (Fig. 5B), which corresponds to an increased half-life of approximately 6 d on average (Fig. 5B). The mutation within the N-terminal domain of PSST (S70F) resulted in relatively slower turnover rates (ΔKD) of CI subunits and consequently a more stable CI compared to ciaf1 (Fig. 5A). This effect was more notable for NQ module subunits, such as PSST, B8, NAD7, NAD9, and 24 kDa, which exhibited slower turnover rates by approximately 1.5- to 2-fold compared to ciaf1 (Fig. 5B). In ciaf1, PSST has a relatively fast turnover rate (0.82 d−1, half-life 20.4 h) whilst in rmb1 ciaf1 and rmb2 ciaf1, PSST has substantially slower turnover rates of 0.41 d−1 (half-life 40 h) and 0.31 d−1 (half-life 54 h), respectively (Fig. 5B). The overall trend of slowed turnover rates in rmb1 ciaf1 and rmb2 ciaf1 suggests that mutations within FTSH3 or PSST independently slow the turnover of CI subunits, resulting in the accumulation of CI and the partial restoration of plant growth phenotypes (Figs. 1 and 3).

The rmb2 ciaf1 mutation within PSST likely affects its N-terminal domain structure

We examined the consequences of both the rmb1 ciaf1 and rmb2 ciaf1 mutations at the molecular level. As plant FTSH3 structures were not available, we modeled Arabidopsis FTSH3 using the human AFG3L2 structure as a template (PDB ID: 6NYY) (Puchades et al. 2019) (Supplemental Fig. S4). AFG3L2 is a matrix-facing AAA+ (m-AAA) protease closely related to bacterial FTSH and plant m-AAA proteases. AFG3L2 exhibits 64.3% similarity and 49.0% identity to Arabidopsis FTSH3 and contains the conserved ATPase and protease domains (Supplemental Fig. S4). The ATPase domain of each AFG3L2 monomer contains an ATP-binding pocket that powers the movement of two substrate-recognizing loops called pore-loop 1 and pore-loop 2 to translocate substrates into the proteolytic domain (Puchades et al. 2017, 2019) (Fig. 6A). AFG3L2 consists of homo-hexameric ATPase and protease domains and a peptide substrate moiety (Puchades et al. 2019), allowing PSST to be modeled as a putative substrate for FTSH3 homo-hexamer, based on the previously determined plant CI structure (PDB ID: 7A23) (Soufari et al. 2020) (Supplemental Fig. S5). The substrate-bound structure shows that PSST can be accommodated into the FTSH3 AAA+ pore and interact with the substrate-recognizing pore-loop 1 (Fig. 6A). Modeling suggests the N-terminal peptide of PSST has a diameter of 9.01 Ǻ and displays no unfavorable clash with substrate-recognizing phenylalanine residues (F395) (Puchades et al. 2019) (Fig. 6B). However, modeling of PSST harboring the S70F mutation suggests a larger helix size (11.50 Ǻ diameter), due to the potential of the bulky benzyl group of phenylalanine for unfavorable interactions with F395 (Fig. 6B). This may lead to thermodynamic instability (Galemou Yoga et al. 2019), preventing interaction between FTSH3 and PSSTS70F.

Figure 6.

Figure 6.

Open in a new tab

Interaction of FTSH3 and PSST allows for the disassembly and degradation of the CI NQ module. A) Homology model structure of FTSH3 (multi-colored ribbons) and the first 30-aa of PSST representing the N-terminal domain (ribbon and surface). The interaction was modeled based on human m-AAA FTSH homolog AFG3L2 with a bound peptide substrate (PDB: 6NYY) (Puchades et al. 2019), B) Molecular interaction analysis of FTSH3–PSST based on the homology modeled structure. PSST with 9.01 Å helical diameter (ribbon, left) fitted to FTSH3 AAA+ pore in the ATPase domain. Substrate-recognizing phenylalanine residues (F395) situated in pore-loop 1 from different monomers are indicated (balls and sticks). EMS-mutated rmb2:ciaf1 PSST (PSSTS70F) has a wider overall diameter (11.50 Å) (ribbon, right), resulting in an unfavorable clash with one of the F395 residues (dotted circle), and C) Proposed model of FTSH3-mediated regulation of the CI matrix arm domain in wild type, ciaf1, rmb1 ciaf1, and rmb2 ciaf1. The binding of PSST by FTSH3 enables its proteolytic activity against the CI matrix arm NQ-module. The assembly defect of the NQ-module in ciaf1 induces its unfolding, disassembly, and subsequent degradation, leading to lower abundance and overall activity of CI. The mutation of either FTSH3 ATPase domain in rmb1 ciaf1 (FTSH3P415L) or the N-terminal domain of PSST in rmb2 ciaf1 (PSSTS70F) abolishes their interaction, resulting in a slower turnover rate and the slight accumulation of NQ-module subunits, which partially restores CI activity.

Discussion

Here, we uncovered the mechanism for the regulation of CI turnover by the inner membrane-bound FTSH3 protease via an interaction with the CI matrix arm subunit PSST. Using the CI-defective mutant ciaf1, we generated 2 independent EMS-induced mutants that displayed restored CI activity and plant growth. We demonstrated that the 2 independent mutations, one within the ATPase domain of FTSH3P415L (rmb1 ciaf1) (Ivanova et al. 2021) and another within the N-terminal domain of PSSTS70F (rmb2 ciaf1), were responsible for the restoration of CI in ciaf1. Furthermore, protein interaction assays show that FTSH3, and not FTSH10, interacts with PSST. Mutations within the FTSH3 ATPase domain (FTSH3P415L) or the N-terminal domain of PSST (PSSTS70F) abolished this interaction (Fig. 4, B and C), demonstrating the importance of these residues and domains for substrate recognition and subsequent NQ-module disassembly and proteolytic degradation (Supplemental Figs. S4 and S5). These mutations resulted in slowed protein turnover rates of NQ-module subunits compared to ciaf1, resulting in the increased accumulation of these subunits, and ultimately increased CI abundance and activity (Fig. 6C).

We previously showed that complementation of rmb1 ciaf1 with FTSH3 resulted in a reverted phenotype typical of ciaf1, i.e. small and developmentally delayed plants. Complementation using a gene encoding proteolytically inactive FTSH3 also reverted this phenotype, indicating that the proteolytic function of FTSH3 is not required for the restoration of CI abundance in this case (Ivanova et al. 2021). Furthermore, the deletion of FTSH3 in the Col-0 background has no observable effect on CI (Kolodziejczak et al. 2018; Ivanova et al. 2021) under normal conditions. Although FTSH3 can form a hetero-hexamer with FTSH10 (Piechota et al. 2010), FTSH10 was not detected in our CI immunoprecipitations (Fig. 4A). Furthermore, no interaction of FTSH10 with PSST was observed using Y2H and BiFC assays (Fig. 4, C and D). It is likely that a homo-hexamer of FTSH3, and not an FTSH3/FTSH10 hetero-hexamer, is involved in PSST recognition and disassembly of the CI matrix arm. A similar situation was previously reported for the human FTSH3 ortholog AFG3L2. AFG3L2 assembles as a homo-hexamer, or as hetero-hexamer with paraplegin (SPG7), an AAA+ protease (Koppen et al. 2007; Puchades et al. 2019). However, only the AFG3L2 homo-hexamer is responsible for the maturation and degradation of OXPHOS components (Arlt et al. 1998; Koppen et al. 2007).

Studies in mammals have shown that damaged CI subunits are disassociated, degraded, and replaced with newly synthesized subunits to maintain CI activity (Szczepanowska et al. 2020). In plants, the NQ-module subunits display higher turnover rates compared to the membrane arm domain subunits and other mitochondrial OXPHOS subunits (Nelson et al. 2013; Li et al. 2017). We previously showed that when CI assembly is defective, FTSH3 plays a role in the disassembly of the NQ module domain, likely via its ATPase domain (Ivanova et al. 2021), which is required for substrate recognition and ATP-dependent protein unfolding, as previously observed in other AAA+ proteases (Puchades et al. 2017, 2019; Shin et al. 2020, 2021). In this study, we uncover the mechanism of how this occurs. We identified PSST as a substrate of FTSH3. PSST, an ortholog of human NDUFS7, is a 20 kDa Q-module subunit that is located within the interface of the matrix and membrane arm domain and is conserved across species (Baradaran et al. 2013; Fiedorczuk et al. 2016; Parey et al. 2019; Soufari et al. 2020; Klusch et al. 2021; Kolata and Efremov 2021) (Supplemental Fig. S4). Mutation of the PSST conserved loop residues in the aerobic yeast Yarrowia lipolytica resulted in reduced electron transfer to ubiquinone (Galemou Yoga et al. 2019), whilst the deletion of PSST in Arabidopsis resulted in severe developmental delays, confirming its essential function (Kühn et al. 2015). PSST is located adjacent to the membrane-bound PP module, forming the ubiquinone-binding tunnel at the interface of the Q module and Nad7 subunit in the PP module (Soufari et al. 2020; Klusch et al. 2021). The N-terminal domain of PSST is exposed to the matrix, particularly when the NQ module is not correctly assembled to the holocomplex (Supplemental Fig. S5), suggesting that PSST is accessible for recognition by FTSH3 for unfolding and subsequent proteolysis (Fig. 6C).

AAA+ proteases were previously shown to function in substrate unfolding in other species. Human YME1L, an FTSH4 ortholog, recognizes its substrates via an accessible N-terminal sequence (Shi et al. 2016). In addition, the accessibility of the substrate peptide through the AAA+ pore is required for protease activity (Shi et al. 2016). The protein unfolding activity of AAA+ proteases depends on ATP hydrolysis and the size of the unfolded polypeptide substrate (Martin et al. 2008). Homology modeling of the S70F mutation in PSST indicated an increased diameter of the helix (Fig. 6B), potentially resulting in inaccessibility through the FTSH3 AAA pore and altering the binding energy of an FTSH3–PSST interaction (Fig. 4).

A similar FTSH protease-mediated turnover mechanism has been proposed for the PSII repair cycle, where the reaction center D1 protein is directly regulated by FTSH protease homologs. The light-absorbing activity of D1 results in its subsequent photo-damage and rapid turnover (Silva et al. 2003). The loss of the thylakoid membrane-associated FTSH2 and FTSH5 resulted in the accumulation of damaged D1 proteins (Kato et al. 2009). In the cyanobacterium Synechocystis sp. PCC 6803, an FTSH homolog (slr0228) was found to play a direct role in the binding, unfolding, and degradation of D1 during the early stage of the PSII repair cycle (Silva et al. 2003).

In mammals, CLPXP, another AAA+ protease, was shown to actively degrade matrix-facing N-module domain subunits (Szczepanowska et al. 2020). Immunoprecipitations revealed that the matrix-localized CLPXP interacted specificity with the matrix-facing NQ subunits NDUFS1, NDUFV1, and NDUFV2 (orthologous to plant 75, 51, and 24 kDa, respectively) (Subrahmanian et al. 2016; Szczepanowska et al. 2020). Similarly, in Arabidopsis, knockdown of the gene encoding the CLPXP subunit CLPP2 resulted in the accumulation of the 75, 51, and 24 kDa subunits and the accumulation of a matrix submodule containing the 24 and 51 kDa subunits (Petereit et al. 2020), pointing to a role for CLPXP in the proteolysis of CI.

Compared to other AAA+ proteases, FTSH homologs exhibit weak protein unfolding activity, as measured by their efficiency in converting ATP hydrolysis to protein degradation (Shi et al. 2016). Indeed in bacteria, FTSH was also proposed to exhibit weak unfoldase activity (Kihara et al. 1999; Herman et al. 2003). Rather, bacterial FTSH recognized substrates when misfolded or disassembled from its complex, allowing it to be unfolded and degraded. Therefore, FTSH regulates the abundance of membrane-bound proteins depending on their folding state or their association with a complex (Kihara et al. 1995, 1999; Herman et al. 2003), allowing FTSH proteases to recognize damaged proteins in membrane-bound complexes (Herman et al. 2003). The disassembly function of FTSH3 in Arabidopsis was only identified in a dysfunctional CI mutant (Ivanova et al. 2021), which is consistent with the proposal that FTSH proteases recognize unfolded substrates.

In conclusion, we uncovered the mechanism of CI turnover by FTSH3. Through its interaction with PSST, FTSH3 plays a role in the disassociation of the NQ module from holo-CI, allowing for the selective degradation and turnover of damaged and non-functional subunits.

Materials and methods

Plant material, growth, and phenotyping

All A. thaliana plants used in this study were in the Col-0 ecotype background. Seeds were surface-sterilized with chlorine gas and sown on half-strength Murashige–Skoog (½ MS) liquid or agar medium (pH 5.7) supplemented with 1.5% (w/v) sucrose and 0.1% (v/v) Gamborg B5 vitamins. A peat moss, vermiculite, and perlite (3:1:1) mix was used for growth in the soil. All plants were grown under a long-day photoperiod (16 h light, 8 h dark) with 100 µmol m−2 s light intensity (with tubular fluorescence lighting) at 22 °C and 60% humidity. Growth parameters were measured from individual seedlings or plants.

Forward genetics and whole-genome sequencing

EMS treatment was used to mutagenize the CI assembly factor 1 knockdown mutant ciaf1 (SALK_143656) as previously described (Weigel and Glazebrook 2006; Ivanova et al. 2021). The progeny of approximately 2,000 individual M1 plants were grown on ½ MS agar medium and visually screened to identify plants exhibiting substantially increased growth compared to ciaf1 (Ivanova et al. 2019). Candidate lines were confirmed in the M3 generation following 3 rounds of back-crossing to ciaf1. Six homozygous rmb2 ciaf1 lines from the segregating back-crossed population were used for whole-genome sequencing.

Genomic DNA was isolated from the samples using a DNeasy Total DNA Purification Maxi Kit (Qiagen) according to the manufacturer's instructions. Sequencing libraries were generated from total genomic DNA using a Nextera DNA Flex Library Prep Kit (Illumina) and sequenced with a 70-bp read length on the NextSeq 500 system (Illumina). Whole-genome sequencing was performed using the HiSeq 2500 platform (Illumina). Sequencing reads were aligned to the Arabidopsis reference genome (Araport11) using bowtie2 v2.3.4.1 (Langmead and Salzberg 2012). Mutations were identified using the “mpileup” and “call” commands of bfctools v1.4 with default parameters (http://www.htslib.org/doc/bcftools.html). A single nucleotide C to T mutation on chromosome 5 within At5g11770 was present in all 6 homozygous revertant lines.

Cloning, plasmid construction, and line generation

The PSST (At5g11770), B14.7 (At2g42210), and FTSH10 (At1g07510) protein-encoding sequences (CDS) were cloned from Col-0 complementary DNA (cDNA). Cloning was carried using Gateway site-specific recombination (Invitrogen) (Katzen 2007) into the pDONR207 entry vector using gene-specific primers (Supplemental Table S1). Site-directed mutagenesis was performed using a QuikChange kit (Stratagene) to generate FTSH3P415L, PSSTS70F, FTSH3TRAP (H586G), and FTSH10TRAP (H586G) with primers listed in Supplemental Table S1. The fragments were recombined into the following plasmids: pH2GW7 (driven by the CaMV 35S promoter) for over-expression lines (Karimi et al. 2002); pGWB11 for C-terminal FLAG-tagged B14.7 (Nakagawa et al. 2007); pGBKCg/pGBKT7 and pGADCg/pGADT7g for yeast-2-hybrid assays (Stellberger et al. 2010); and pSAT4-DEST-nEYFP-C1 (pE3136), pSAT4-DEST-nEYFP-N1 (pE3134), pSAT5-DEST-cEYFP-C1 (pE3130), and pSAT5-DEST-cEYFP-N1 (pE3132) for BiFC assays (Citovsky et al. 2006). The full-length PSST gene from genomic DNA, including the 1 kb upstream promoter region, was cloned into pCAMBIA1300 using NEBuilder HiFi DNA assembly master mix (NEBioLabs). Complementation and over-expression lines were generated using Agrobacterium tumefaciens via the floral dip method (Karimi et al. 2002).

RT-qPCR

Total RNA was isolated from 14-d-old agar-grown seedlings using a FavorPrep total RNA purification mini kit (Favorgen). cDNA was generated using a high-capacity cDNA reverse transcription kit (Applied Biosystem). RT-qPCR standards were generated from the Col-0 cDNA pool using MyTaq polymerase mix (Bioline) and purified using a Favorprep PCR Purification Kit (Favorgen) according to the manufacturer's instructions with the primers listed in Supplemental Table S1. RT-qPCR was carried out in a LightCycler 480 (Roche) using SYBR Green master mix (Roche) and gene-specific primers (Supplemental Table S1). ACTIN2 (At3g18780) was used as a control. All measurements were technically replicated at least twice on RNA isolated from 3 independent seedling populations.

BiFC assay

BiFC assays were carried out using transiently transformed isolated Arabidopsis leaf mesophyll protoplasts as previously described (Yoo et al. 2007; Wu et al. 2009). Briefly, 12 to 15 fully expanded 3- to 4-wk-old rosette leaves were harvested and sandwiched with tape to remove the lower epidermis layer. Stripped leaves were incubated in a freshly prepared enzyme solution (0.5% w/v cellulase, 0.25% w/v pectolyase, 0.4 M mannitol, 20 mm KCl, 10 mm CaCl2, 20 mm MES pH 5.7, 0.1% w/v BSA, sterilized through a 0.45 μm syringe filter) for 1 h at 40 rpm in the light. Protoplasts were harvested by centrifugation at 100 × g 4 °C for 12 min, washed twice with W5 solution (154 mm NaCl, 125 mm CaCl2, 5 mm KCl, 5 mm glucose, 2 mm MES pH 5.7), and incubated on ice for 30 min in the dark. Protoplast concentration was adjusted to approximately 105 cells mL−1 in 2 mL MMG solution (0.4 m mannitol, 15 mm MgCl2, 5 mm glucose, 4 mm MES pH 5.7, sterilized through a 0.45 μm syringe filter). To transform protoplasts, 10 μg of plasmid (in a 10 μL total volume) and 110 μL freshly prepared sterile polyethylene glycol (PEG) solution (40% w/v PEG, 0.1 M CaCl2, 0.2 M mannitol) were added to 500 μL protoplasts. The mixture was incubated in the dark for 20 min, washed twice with 440 μL W5 solution (centrifuged 100 × g for 2 min), and resuspended in 500 μL W5 solution. The transformed mixture was incubated overnight (12 h) in the dark and visualized under a BX61 fluorescence microscope (Olympus). Fluorescence was visualized with the excitation/emission wavelengths of 540 to 550/565 to 575 nm for chloroplast autofluorescence (TRITC) and 510 to 520/525 to 535 nm for enhanced yellow fluorescent protein.

Y2H assay

Y2H assays were carried out as previously described (Ivanova et al. 2019). Bait constructs containing DNA-binding domain (pGBKCg/PGBKT7) (Stellberger et al. 2010) were transformed into yeast strain Y187, while prey constructs containing the activation domain/AD (pGADCg/pGADT7) (Stellberger et al. 2010) were transformed into yeast strain AH109. Mating was carried out in a 96-well plate according to the manufacturer's instructions (Matchmaker GAL4 Two Hybrid System (Clontech/TakaraBio)). Mated diploid strains were serially diluted and plated on –Leu–Trp DDO and –Leu–Trp–His–Ade QDO synthetic defined (SD) agar medium and incubated for 4 d at 30 °C. Interactions were considered positive if growth on QDO was observed in 3 independent yeast transformations and mating.

Mitochondrial isolation, gel electrophoresis, and immunoblotting

Mitochondria were isolated from 14-d-old seedlings grown in pots in liquid ½ MS medium with 200 rpm shaking as described previously (Duncan et al. 2017). For BN-PAGE, samples were prepared by solubilizing mitochondria with 5% (w/v) digitonin as previously described (Eubel et al. 2005). Samples were resolved using precast NativePAGE 4% to 16% Bis-Tris gels (Invitrogen). CI in-gel activity staining was carried out as previously described (Schertl and Braun 2015). Protein separation and identification were performed by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) (4% to 12% Bis-polyacrylamide gel) of solubilized intact mitochondria, followed by immunoblotting. After SDS–PAGE, the proteins were transferred onto a PVDF membrane and immunodetection was carried out using antibodies against 75 kDa (in 1:500 dilution) (PhytoAB), NDUFS4 (18 kDa; in 1:5,000 dilution) (Meyer et al. 2009), PSST (20 kDa; in 1:2,000 dilution) (Meyer et al. 2011), B14.7 (in 1:5,000 dilution) (Wang et al. 2012), NAD1 (in 1:1,000 dilution) (PhytoAB), CAL1 (in 1:1,000 dilution) (Fromm et al. 2016), CA2 (in 1:500 dilution) (Perales et al. 2005), SDH1 (in 1:1,000 dilution) (Peters et al. 2012), RISP (in 1:5,000 dilution) (Carrie et al. 2010), COX2 (in 1:1,000 dilution) (Agrisera), ATPβ (in 1:5,000 dilution) (Agrisera), TOM40 (in 1:5,000 dilution) (Carrie et al. 2010), FTSH3 (in 1:3,000 dilution) (Kolodziejczak et al. 2007), FTSH4 (in 1:500 dilution) (Urantowka et al. 2005), and FTSH10 (in 1:1,000 dilution) (Piechota et al. 2010). For densitometric measurement of band intensity, the blots were analyzed in ImageJ (Schneider et al. 2012). In-gel CI activity assays and immunoblotting were done on mitochondria from 3 independent mitochondrial isolations.

Immunoprecipitation

Mitochondria were isolated from B14.7FLAG tagged lines. Anti-FLAG (anti-DYKDDDK) monoclonal M2 Sepharose beads (Sigma Aldrich) were used to bind and precipitate B14.7FLAG along with any associated proteins. Freshly isolated mitochondria were ruptured with 5% (w/v) digitonin in 0.5 mL lysis buffer containing 20 mm Tris-Cl, 50 mm NaCl, 0.1 mm EDTA, 10% (v/v) glycerol, and 1 mm phenylmethanesulfonyl fluoride, and 10 μL of protease inhibitor (protease and phosphatase inhibitor cocktail dissolved in 1 mL water) (Sigma Aldrich) and incubated on ice for 30 min. The lysed mitochondrial mixture was centrifuged at 20,000 × g for 15 min at 4 °C. The supernatant was removed, added to 150 μL of prewashed anti-FLAG tag beads, and incubated for 16 h at 4 °C on a rotary shaker. The supernatant was removed by centrifugation at 1,000 × g for 1 min, and the beads were washed 3 times in 1 mL 1 × PBS buffer (140 mm NaCl, 10.1 mm Na2HPO4, 1.76 mm KH2PO4, and 2.68 mm KCl). The supernatants, wash fractions, and beads were stored at −80 °C prior to SDS–PAGE and immunoblotting.

CI activity assay

CI activity (NADH:ubiquinone reductase) was measured as previously described (Huang et al. 2015). The decrease of A420 was recorded after adding 1 μg mitochondrial sample to 200 μL of reaction solution containing 50 mm Tris–HCl buffer pH 7.2, 50 mm NaCl, 0.2 mm deamino-NADH, and 1 mm [Fe(CN)6]3− (ε 1.03 mm−1 cm−1 at 420 nm). CI activity was determined as the rate of [Fe(CN)6]3− reduction as it received electrons from NADH oxidation by CI. Activity assays were done on mitochondria from 3 independent mitochondrial isolations.

HiRIEF proteomics analysis

Steady-state proteomics of Col-0, ciaf1-1, rmb1::ciaf1, and rmb2::ciaf1 plants were carried out by HiRIEF followed by LC-MS (Branca et al. 2014) as described previously (Ivanova et al. 2019). The heatmap of significant changes was generated in R using the ggplot2 package with Euclidian distribution and Ward's hierarchical clustering method. Gene ontology (GO) term enrichment was carried out using the ClueGO plugin in Cytoscape (Shannon et al. 2003; Bindea et al. 2009), with 2-sided hypergeometric enrichment and Bonferroni P-value correction. Relative abundances of proteins were mapped to CI the structure from cauliflower (Brassica oleracea var. botrytis; PDB: 7A23) (Soufari et al. 2020). Protein structures were visualized in UCSF ChimeraX (Pettersen et al. 2021), using data sets available in Supplemental Data Set 1.

Proteomics analysis to measure turnover rate

Heavy nitrogen labeling of Arabidopsis seedlings was carried out as previously described (Li et al. 2017). Briefly, seeds were grown in 80 mL of liquid ½ MS medium for 10 d. The seedlings were washed with sterile liquid ½ MS medium without nitrogen 3 times, placed in modified 15N growth medium containing ½ MS medium without nitrogen, 0.96 g L−115NH415NO3, and 0.83 g/L K15NO3, and grown for another 4 d. Fourteen-day-old seedlings were harvested and mitochondria isolated as described previously (Duncan et al. 2017). Matrix fractions were collected from the mitochondrial samples (corresponding to 500 μg protein content) by performing 3× freeze–thaw cycles (20 min at −20 °C, followed by 20 min at 4 °C), vortexed at 4 °C, and centrifuged at 20,000 × g for 30 min at 4 °C. Trypsin digestion, identification, and quantification of peptides using mass spectrometry were performed as previously described (Li et al. 2017).

Sequence alignment and structural homology modeling

FTSH3, FTSH10, and PSST homolog sequences were retrieved from UniProt (https://www.uniprot.org/) (The UniProt Consortium 2017). Sequence alignments were carried out using Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/) (Sievers et al. 2011). The Arabidopsis FTSH3 structure was modeled based on the full protein sequence retrieved from The Arabidopsis Information Resource (TAIR) (https://www.arabidopsis.org/) (Huala et al. 2001). Homology modeling was carried out using Phyre2 (Kelley et al. 2015), with the human m-AAA protease AFG3L2 structure (PDB: 6NYY) as the reference template (Puchades et al. 2019). The N-terminal domain structure of PSST was retrieved from the cauliflower CI structure (PDB: 7A23) (Soufari et al. 2020). Structural visualization, surface generation, structural superimposition, and in silico point mutations were carried out in Maestro (Schrödinger, LLC) and ChimeraX (Pettersen et al. 2021).

Statistical analysis

All replicates used in this study were described in each respective Materials and Methods section. Statistically significant differences between multiple means were determined using single-factor ANOVA (95% confidence level, P < 0.05). Significantly different means were then determined using Tukey’s honestly significant difference (HSD) post hoc test. Statistical analyses are provided in Supplemental Data Set 3.

Accession numbers

Proteomics data used in this study have been deposited in the ProteomeXchange Consortium via the Proteomics Identification Database (PRIDE) partner repository (https://www.ebi.ac.uk/pride/) with the project identifier PXD011795. Proteomics data for ciaf1 and rmb1 ciaf1 were generated in our previous studies (Ivanova et al. 2019, 2021). Protein abundance data for Col-0 (WT), ciaf1, rmb1 ciaf1, and rmb2 ciaf1 can be accessed in the annotated protein table (Arabidopsis_Proteomics_annotated_protein_table.xslx) as Col0, ciaf1-1, Mut1, and Mut3, respectively. Sequence data associated with this study can be found in TAIR database (https://www.arabidopsis.org/) with the following Arabidopsis Genome Initiative identifiers: AT5G37510 (75 kDa), AT3G18780 (ACTIN2), AT5G08670 (ATPβ), AT2G42210 (B14.7), AT1G47260 (CA2), AT5G63510 (CAL1), AT1G76060 (CIAF1), ATMG00160 (COX2), AT2G29080 (FTSH3), AT2G26140 (FTSH4), AT1G07510 (FTSH10), ATMG01275 (NAD1), AT5G67590 (18 kDa/NDUFS4), AT5G11770 (20 kDa/PSST), AT3G20000 (TOM40), AT5G13430 (RISP), and AT5G66760 (SDH1-1).

Supplementary Material

koad128_Supplementary_Data
Click here for additional data file. (13.4MB, zip)

Acknowledgments

We thank Dr. Elke Ströher and Dr. Owen Duncan (WA Proteomics Facility, Centre for Microscopy, Characterisation, and Analysis, UWA) for their assistance in proteomic turnover rate analysis.

Contributor Information

Abi S Ghifari, School of Molecular Sciences & ARC Centre of Excellence in Plant Energy Biology, The University of Western Australia, Perth, WA 6009, Australia.

Aneta Ivanova, School of Molecular Sciences & ARC Centre of Excellence in Plant Energy Biology, The University of Western Australia, Perth, WA 6009, Australia.

Oliver Berkowitz, Department of Animal, Plant and Soil Science, School of Life Science, ARC Centre of Excellence in Plant Energy Biology, La Trobe University, Bundoora, VIC 3086, Australia.

James Whelan, College of Life Science, Zhejiang University, Hangzhou, Zhejiang 310058, PR China.

Monika W Murcha, School of Molecular Sciences & ARC Centre of Excellence in Plant Energy Biology, The University of Western Australia, Perth, WA 6009, Australia.

Author Contributions

A.S.G., A.I., and M.W.M. performed experiments and analyzed the data. A.S.G. prepared all figures. O.B. performed the genomic sequencing. J.W. and M.W.M. designed the experiments and supervised the study. A.S.G. and M.W.M. wrote the manuscript with input from all co-authors.

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. Phenotypic analysis of generated mutants.

Supplemental Figure S2. Protein abundance of OXPHOS complex subunits in ciaf1, rmb1 ciaf1, and rmb2 ciaf1.

Supplemental Figure S3. Turnover rates of OXPHOS complex subunits in ciaf1, rmb1 ciaf1, and rmb2 ciaf1.

Supplemental Figure S4. Sequence alignment and structural homology modeling of FTSH3.

Supplemental Figure S5. Sequence alignment and structural features of PSST.

Supplemental Table S1. Primers used in this study.

Supplemental Data Set 1. Proteomics analysis of OXPHOS subunits.

Supplemental Data Set 2. Turnover analysis of mitochondrial proteins.

Supplemental Data Set 3. Statistical analysis using ANOVA and Tukey’s HSD post hoc test.

Funding

This work was supported by the Australian Research Council (ARC) Future Fellowship to M.W.M. (FT130100112), ARC Centre of Excellence in Plant Energy Biology (CE140100008) to J.W., and Discovery Project funding to M.W.M. (DP200101922) and J.W. and M.W.M. (DP210103258). A.S.G. is funded by the Australian Government Research Training Program and University Postgraduate Award at The University of Western Australia.

References

  1. Arlt H, Steglich G, Perryman R, Guiard B, Neupert W, Langer T. The formation of respiratory chain complexes in mitochondria is under the proteolytic control of the m-AAA protease. EMBO J. 1998:17(16):4837–4847. 10.1093/emboj/17.16.4837 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Baradaran R, Berrisford JM, Minhas GS, Sazanov LA. Crystal structure of the entire respiratory complex I. Nature. 2013:494(7438):443–448. 10.1038/nature11871 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bindea G, Mlecnik B, Hackl H, Charoentong P, Tosolini M, Kirilovsky A, Fridman WH, Pagès F, Trajanoski Z, Galon J. ClueGO: a Cytoscape plug-in to decipher functionally grouped gene ontology and pathway annotation networks. Bioinformatics. 2009:25(8):1091–1093. 10.1093/bioinformatics/btp101 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Branca RMM, Orre LM, Johansson HJ, Granholm V, Huss M, Pérez-Bercoff A, Forshed J, Käll L, Lehtiö J. HiRIEF LC-MS enables deep proteome coverage and unbiased proteogenomics. Nat Methods. 2014:11(1):59–62. 10.1038/nmeth.2732 [DOI] [PubMed] [Google Scholar]
  5. Carrie C, Giraud E, Duncan O, Xu L, Wang Y, Huang S, Clifton R, Murcha M, Filipovska A, Rackham O, et al. Conserved and novel functions for Arabidopsis thaliana MIA40 in assembly of proteins in mitochondria and peroxisomes. J Biol Chem. 2010:285(46):36138–36148. 10.1074/jbc.M110.121202 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Citovsky V, Lee L-Y, Vyas S, Glick E, Chen M-H, Vainstein A, Gafni Y, Gelvin SB, Tzfira T. Subcellular localization of interacting proteins by bimolecular fluorescence complementation in planta. J Mol Biol. 2006:362(5):1120–1131. 10.1016/j.jmb.2006.08.017 [DOI] [PubMed] [Google Scholar]
  7. Duncan O, Millar AH, Taylor NL. Isolation of mitochondria, their sub-organellar compartments, and membranes. In: Taylor NL, Millar AH, editors. Isolation of plant organelles and structures: methods and protocols. New York: (NY: ): Springer; 2017. p. 83–96. [DOI] [PubMed] [Google Scholar]
  8. Dutilleul C, Garmier M, Noctor G, Mathieu C, Chétrit P, Foyer CH, De Paepe R. Leaf mitochondria modulate whole cell redox homeostasis, set antioxidant capacity, and determine stress resistance through altered signaling and diurnal regulation. Plant Cell. 2003:15(5):1212–1226. 10.1105/tpc.009464 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Eubel H, Braun HP, Millar AH. Blue-native PAGE in plants: a tool in analysis of protein-protein interactions. Plant Methods. 2005:1(1):1–13. 10.1186/1746-4811-1-11 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Fiedorczuk K, Letts JA, Degliesposti G, Kaszuba K, Skehel M, Sazanov LA. Atomic structure of the entire mammalian mitochondrial complex I. Nature. 2016:538(7625):406–410. 10.1038/nature19794 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Fromm S, Senkler J, Eubel H, Peterhänsel C, Braun H-P. Life without complex I: proteome analyses of an Arabidopsis mutant lacking the mitochondrial NADH dehydrogenase complex. J Exp Bot. 2016:67(10):3079–3093. 10.1093/jxb/erw165 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Galemou Yoga E, Haapanen O, Wittig I, Siegmund K, Sharma V, Zickermann V. Mutations in a conserved loop in the PSST subunit of respiratory complex I affect ubiquinone binding and dynamics. Biochim Biophys Acta Bioenerg. 2019:1860(7):573–581. 10.1016/j.bbabio.2019.06.006 [DOI] [PubMed] [Google Scholar]
  13. Ghifari AS, Murcha MW. Proteolytic regulation of mitochondrial oxidative phosphorylation components in plants. Biochem Soc Trans. 2022:50(3):1119–1132. 10.1042/BST20220195 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Herman C, Prakash S, Lu CZ, Matouschek A, Gross CA. Lack of a robust unfoldase activity confers a unique level of substrate specificity to the universal AAA protease FTSH. Mol Cell. 2003:11(3):659–669. 10.1016/S1097-2765(03)00068-6 [DOI] [PubMed] [Google Scholar]
  15. Hirst J, Roessler MM. Energy conversion, redox catalysis and generation of reactive oxygen species by respiratory complex I. Biochim Biophys Acta Bioenerg. 2016:1857(7):872–883. 10.1016/j.bbabio.2015.12.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Huala E, Dickerman AW, Garcia-Hernandez M, Weems D, Reiser L, LaFond F, Hanley D, Kiphart D, Zhuang M, Huang W, et al. The Arabidopsis Information Resource (TAIR): a comprehensive database and web-based information retrieval, analysis, and visualization system for a model plant. Nucleic Acids Res. 2001:29(1):102–105. 10.1093/nar/29.1.102 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Huang S, Lee CP, Millar AH. Activity assay for plant mitochondrial enzymes. In: Whelan J, Murcha MW, editors. Plant mitochondria: methods and protocols. New York: (NY: ): Springer; 2015. p. 139–149. [DOI] [PubMed] [Google Scholar]
  18. Huang S, Petereit J, Millar AH. Loss of conserved mitochondrial CLPP and its functions lead to different phenotypes in plants and other organisms. Plant Signal Behav. 2020:15(12):1831789. 10.1080/15592324.2020.1831789 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Ivanova A, Ghifari AS, Berkowitz O, Whelan J, Murcha MW. The mitochondrial AAA protease FTSH3 regulates Complex I abundance by promoting its disassembly. Plant Physiol. 2021:186(1):599–610. 10.1093/plphys/kiab074 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Ivanova A, Gill-Hille M, Huang S, Branca R, Kmiec B, Teixeira PF, Lehtiö J, Whelan J, Murcha MW. A mitochondrial LYR protein is required for complex I assembly. Plant Physiol. 2019:181(4):1632–1650. 10.1104/pp.19.00822 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Karimi M, Inzé D, Depicker A. GATEWAY vectors for Agrobacterium-mediated plant transformation. Trends Plant Sci. 2002:7(5):193–195. 10.1016/S1360-1385(02)02251-3 [DOI] [PubMed] [Google Scholar]
  22. Kato Y, Miura E, Ido K, Ifuku K, Sakamoto W. The variegated mutants lacking chloroplastic FTSHs are defective in D1 degradation and accumulate reactive oxygen species. Plant Physiol. 2009:151(4):1790–1801. 10.1104/pp.109.146589 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Katzen F. Gateway® recombinational cloning: a biological operating system. Expert Opin Drug Discov. 2007:2(4):571–589. 10.1517/17460441.2.4.571 [DOI] [PubMed] [Google Scholar]
  24. Kelley LA, Mezulis S, Yates C, Wass M, Sternberg M. The Phyre2 web portal for protein modelling, prediction, and analysis. Nat Protoc. 2015:10(6):845–858. 10.1038/nprot.2015.053 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kihara A, Akiyama Y, Ito K. FTSH is required for proteolytic elimination of uncomplexed forms of SecY, an essential protein translocase subunit. Proc Natl Acad Sci U S A. 1995:92(10):4532–4536. 10.1073/pnas.92.10.4532 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kihara A, Akiyama Y, Koreaki I. Dislocation of membrane proteins in FTSH-mediated proteolysis. EMBO J. 1999:18(11):2970–2981. 10.1093/emboj/18.11.2970 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Klusch N, Dreimann M, Senkler J, Rugen N, Kühlbrandt W, Braun H-P. Cryo-EM structure of the respiratory I + III2 supercomplex from Arabidopsis thaliana at 2 Å resolution. Nat Plants. 2023:9(1):142–156. 10.1038/s41477-022-01308-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Klusch N, Senkler J, Yildiz Ö, Kühlbrandt W, Braun H-P. A ferredoxin bridge connects the two arms of plant mitochondrial complex I. Plant Cell. 2021:33(6):2072–2091. 10.1093/plcell/koab092 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kolata P, Efremov RG. Structure of Escherichia coli respiratory complex I reconstituted into lipid nanodiscs reveals an uncoupled conformation. Elife. 2021:10:e68710. 10.7554/eLife.68710 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kolodziejczak M, Gibala M, Urantowka A, Janska H. The significance of Arabidopsis AAA proteases for activity and assembly/stability of mitochondrial OXPHOS complexes. Physiol Plant. 2007:129(1):135–142. 10.1111/j.1399-3054.2006.00835.x [DOI] [Google Scholar]
  31. Kolodziejczak M, Skibior-Blaszczyk R, Janska H. m-AAA complexes are not crucial for the survival of Arabidopsis under optimal growth conditions despite their importance for mitochondrial translation. Plant Cell Physiol. 2018:59(5):1006–1016. 10.1093/pcp/pcy041 [DOI] [PubMed] [Google Scholar]
  32. Koppen M, Metodiev MD, Casari G, Rugarli EI, Langer T. Variable and tissue-specific subunit composition of mitochondrial m-AAA protease complexes linked to hereditary spastic paraplegia. Mol Cell Biol. 2007:27(2):758–767. 10.1128/MCB.01470-06 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Kühn K, Obata T, Feher K, Bock R, Fernie AR, Meyer EH. Complete mitochondrial complex I deficiency induces an up-regulation of respiratory fluxes that is abolished by traces of functional complex I. Plant Physiol. 2015:168(4):1537–1549. 10.1104/pp.15.00589 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nat Methods. 2012:9(4):357–359. 10.1038/nmeth.1923 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Li L, Nelson C, Fenske R, Trösch J, Pružinská A, Millar AH, Huang S. Changes in specific protein degradation rates in Arabidopsis thaliana reveal multiple roles of Lon1 in mitochondrial protein homeostasis. Plant J. 2017:89(3):458–471. 10.1111/tpj.13392 [DOI] [PubMed] [Google Scholar]
  36. Ligas J, Pineau E, Bock R, Huynen MA, Meyer EH. The assembly pathway of complex I in Arabidopsis thaliana. Plant J. 2019:97(3):447–459. 10.1111/tpj.14133 [DOI] [PubMed] [Google Scholar]
  37. Maldonado M, Fan Z, Abe KM, Letts JA. Plant-specific features of respiratory supercomplex I + III2 from Vigna radiata. Nat Plants. 2023:9(1):157–168. 10.1038/s41477-022-01306-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Maldonado M, Padavannil A, Zhou L, Guo F, Letts JA. Atomic structure of a mitochondrial complex I intermediate from vascular plants. Elife. 2020:9:e56664. 10.7554/eLife.56664 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Martin A, Baker TA, Sauer RT. Protein unfolding by a AAA+ protease is dependent on ATP-hydrolysis rates and substrate energy landscapes. Nat Struct Mol Biol. 2008:15(2):139–145. 10.1038/nsmb.1380 [DOI] [PubMed] [Google Scholar]
  40. Maziak A, Heidorn-Czarna M, Weremczuk A, Janska H. FTSH4 and OMA1 mitochondrial proteases reduce moderate heat stress-induced protein aggregation. Plant Physiol. 2021:187(2):769–786. 10.1093/plphys/kiab296 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Meyer EH, Solheim C, Tanz SK, Bonnard G, Millar AH. Insights into the composition and assembly of the membrane arm of plant complex I through analysis of subcomplexes in Arabidopsis mutant lines. J Biol Chem. 2011:286(29):26081–26092. 10.1074/jbc.M110.209601 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Meyer EH, Tomaz T, Carroll AJ, Estavillo G, Delannoy E, Tanz SK, Small ID, Pogson BJ, Millar AH. Remodeled respiration in ndufs4 with low phosphorylation efficiency suppresses Arabidopsis germination and growth and alters control of metabolism at night. Plant Physiol. 2009:151(2):603–619. 10.1104/pp.109.141770 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Nakagawa T, Kurose T, Hino T, Tanaka K, Kawamukai M, Niwa Y, Toyooka K, Matsuoka K, Jinbo T, Kimura T. Development of series of gateway binary vectors, pGWBs, for realizing efficient construction of fusion genes for plant transformation. J Biosci Bioeng. 2007:104(1):34–41. 10.1263/jbb.104.34 [DOI] [PubMed] [Google Scholar]
  44. Nelson CJ, Li L, Jacoby RP, Millar AH. Degradation rate of mitochondrial proteins in Arabidopsis thaliana cells. J Proteome Res. 2013:12(7):3449–3459. 10.1021/pr400304r [DOI] [PubMed] [Google Scholar]
  45. Opalińska M, Jańska H. AAA proteases: guardians of mitochondrial function and homeostasis. Cells. 2018:7(10):163. 10.3390/cells7100163 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Parey K, Haapanen O, Sharma V, Köfeler H, Züllig T, Prinz S, Siegmund K, Wittig I, Mills DJ, Vonck J, et al. High-resolution cryo-EM structures of respiratory complex I: mechanism, assembly, and disease. Sci Adv. 2019:5(12):eaax9484. 10.1126/sciadv.aax9484 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Perales M, Eubel H, Heinemeyer J, Colaneri A, Zabaleta E, Braun H-P. Disruption of a nuclear gene encoding a mitochondrial gamma carbonic anhydrase reduces complex I and supercomplex I + III2 levels and alters mitochondrial physiology in Arabidopsis. J Mol Biol. 2005:350(2):263–277. 10.1016/j.jmb.2005.04.062 [DOI] [PubMed] [Google Scholar]
  48. Petereit J, Duncan O, Murcha MW, Fenske R, Cincu E, Cahn J, Pružinská A, Ivanova A, Kollipara L, Wortelkamp S, et al. Mitochondrial CLPP2 assists coordination and homeostasis of respiratory complexes. Plant Physiol. 2020:184(1):148–164. 10.1104/pp.20.00136 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Peters K, Niessen M, Peterhansel C, Spath B, Holzle A, Binder S, Marchfelder A, Braun H-P. Complex I-complex II ratio strongly differs in various organs of Arabidopsis thaliana. Plant Mol Biol. 2012:79(3):273–284. 10.1007/s11103-012-9911-4 [DOI] [PubMed] [Google Scholar]
  50. Pettersen EF, Goddard TD, Huang CC, Meng EC, Couch GS, Croll TI, Morris JH, Ferrin TE. UCSF Chimerax: structure visualization for researchers, educators, and developers. Protein Sci. 2021:30(1):70–82. 10.1002/pro.3943 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Piechota J, Kolodziejczak M, Juszczak I, Sakamoto W, Janska H. Identification and characterization of high molecular weight complexes formed by matrix AAA proteases and prohibitins in mitochondria of Arabidopsis thaliana. J Biol Chem. 2010:285(17):12512–12521. 10.1074/jbc.M109.063644 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Puchades C, Ding B, Song A, Wiseman RL, Lander GC, Glynn SE. Unique structural features of the mitochondrial AAA+ protease AFG3L2 reveal the molecular basis for activity in health and disease. Mol Cell. 2019:75(5):1073–1085.e6. 10.1016/j.molcel.2019.06.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Puchades C, Rampello AJ, Shin M, Giuliano CJ, Wiseman RL, Glynn SE, Lander GC. Structure of the mitochondrial inner membrane AAA+ protease YME1 gives insight into substrate processing. Science (80-.). 2017:358(6363):eaao0464. 10.1126/science.aao0464 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Schertl P, Braun H-P. Activity measurements of mitochondrial enzymes in native gels. In: Whelan J, Murcha MW, editors. Plant mitochondria: methods and protocols. New York: (NY: ): Springer; 2015. p. 131–138. [DOI] [PubMed] [Google Scholar]
  55. Schimmeyer J, Bock R, Meyer EH. l-Galactono-1,4-lactone dehydrogenase is an assembly factor of the membrane arm of mitochondrial complex I in Arabidopsis. Plant Mol Biol. 2016:90(1–2):117–126. 10.1007/s11103-015-0400-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012:9(7):671–675. 10.1038/nmeth.2089 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, Ideker T. Cytoscape: a software environment for integrated models. Genome Res. 2003:13(11):2498–2504. 10.1101/gr.1239303 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Shi H, Rampello AJ, Glynn SE. Engineered AAA+ proteases reveal principles of proteolysis at the mitochondrial inner membrane. Nat Commun. 2016:7(1):13301. 10.1038/ncomms13301 [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Shin M, Puchades C, Asmita A, Puri N, Adjei E, Wiseman RL, Karzai AW, Lander GC. Structural basis for distinct operational modes and protease activation in AAA+ protease Lon. Sci Adv. 2020:6(21):eaba8404. 10.1126/sciadv.aba8404 [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Shin M, Watson ER, Song AS, Mindrebo JT, Novick SJ, Griffin PR, Wiseman RL, Lander GC. Structures of the human LONP1 protease reveal regulatory steps involved in protease activation. Nat Commun. 2021:12(1):3239. 10.1038/s41467-021-23495-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H, Remmert M, Soding J, et al. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol. 2011:7(1):539. 10.1038/msb.2011.75 [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Silva P, Thompson E, Bailey S, Kruse O, Mullineaux CW, Robinson C, Mann NH, Nixon PJ. FTSH is involved in the early stages of repair of photosystem II in Synechocystis sp PCC 6803. Plant Cell. 2003:15(9):2152–2164. 10.1105/tpc.012609 [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Soufari H, Parrot C, Kuhn L, Waltz F, Hashem Y. Specific features and assembly of the plant mitochondrial complex I revealed by cryo-EM. Nat Commun. 2020:11(1):5195. 10.1038/s41467-020-18814-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Stellberger T, Häuser R, Baiker A, Pothineni VR, Haas J, Uetz P. Improving the yeast two-hybrid system with permutated fusions proteins: the Varicella Zoster Virus interactome. Proteome Sci. 2010:8(1):8. 10.1186/1477-5956-8-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Subrahmanian N, Remacle C, Hamel PP. Plant mitochondrial Complex I composition and assembly: a review. Biochim Biophys Acta Bioenerg. 2016:1857(7):1001–1014. 10.1016/j.bbabio.2016.01.009 [DOI] [PubMed] [Google Scholar]
  66. Szczepanowska K, Senft K, Heidler J, Herholz M, Kukat A, Höhne MN, Hofsetz E, Becker C, Kaspar S, Giese H, et al. A salvage pathway maintains highly functional respiratory complex I. Nat Commun. 2020:11(1):1643. 10.1038/s41467-020-15467-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. The UniProt Consortium . UniProt: the universal protein knowledgebase. Nucleic Acids Res. 2017:45(D1):D158–D169. 10.1093/nar/gkw1099 [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Urantowka A, Knorpp C, Olczak T, Kolodziejczak M, Janska H. Plant mitochondria contain at least two i-AAA-like complexes. Plant Mol Biol. 2005:59(2):239–252. 10.1007/s11103-005-8766-3 [DOI] [PubMed] [Google Scholar]
  69. van Wijk KJ. Protein maturation and proteolysis in plant plastids, mitochondria, and peroxisomes. Annu Rev Plant Biol. 2015:66(1):75–111. 10.1146/annurev-arplant-043014-115547 [DOI] [PubMed] [Google Scholar]
  70. Vierstra RD. The ubiquitin-26S proteasome system at the nexus of plant biology. Nat Rev Mol Cell Biol. 2009:10(6):385–397. 10.1038/nrm2688 [DOI] [PubMed] [Google Scholar]
  71. Wang Y, Carrie C, Giraud E, Elhafez D, Narsai R, Duncan O, Whelan J, Murcha MW. Dual location of the mitochondrial preprotein transporters B14.7 and Tim23-2 in complex I and the TIM17:23 complex in Arabidopsis links mitochondrial activity and biogenesis. Plant Cell. 2012:24(6):2675–2695. 10.1105/tpc.112.098731 [DOI] [PMC free article] [PubMed] [Google Scholar]
  72. Weigel D, Glazebrook J. EMS mutagenesis of Arabidopsis seed. Cold Spring Harb Protoc. 2006:2006(5):pdb.prot4621. 10.1101/pdb.prot4621 [DOI] [PubMed] [Google Scholar]
  73. Wu F-H, Shen S-C, Lee L-Y, Lee S-H, Chan M-T, Lin C-S. Tape-Arabidopsis sandwich - a simpler Arabidopsis protoplast isolation method. Plant Methods. 2009:5(1):16. 10.1186/1746-4811-5-16 [DOI] [PMC free article] [PubMed] [Google Scholar]
  74. Yoo S-D, Cho Y-H, Sheen J. Arabidopsis mesophyll protoplasts: a versatile cell system for transient gene expression analysis. Nat Protoc. 2007:2(7):1565–1572. 10.1038/nprot.2007.199 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

koad128_Supplementary_Data
Click here for additional data file. (13.4MB, zip)

Articles from The Plant Cell are provided here courtesy of Oxford University Press

RESOURCES