Temperature-driven changes in membrane fluidity differentially impact FILAMENTATION TEMPERATURE-SENSITIVE H2-mediated photosystem II repair

Abstract

The Arabidopsis (Arabidopsis thaliana) yellow variegated2 (var2) mutant, lacking functional FILAMENTATION TEMPERATURE-SENSITIVE H2 (FtsH2), an ATP-dependent zinc metalloprotease, is a powerful tool for studying the photosystem II (PSII) repair process in plants. FtsH2, forming hetero-hexamers with FtsH1, FtsH5, and FtsH8, plays an indispensable role in PSII proteostasis. Although abiotic stresses like cold and heat increase chloroplast reactive oxygen species (ROS) and PSII damage, var2 mutants behave like wild-type plants under heat stress but collapse under cold stress. Our study on transgenic var2 lines expressing FtsH2 variants, defective in either substrate extraction or proteolysis, reveals that cold stress causes an increase in membrane viscosity, demanding more substrate extraction power than proteolysis by FtsH2. Overexpression of FtsH2 lacking substrate extraction activity does not rescue the cold-sensitive phenotype, while overexpression of FtsH2 lacking protease activity does in var2, with other FtsH isomers present. This indicates that FtsH2's substrate extraction activity is indispensable under cold stress when membranes become more viscous. As temperatures rise and membrane fluidity increases, substrate extraction activity from other isomers suffices, explaining the var2 mutant's heat stress resilience. These findings underscore the direct effect of membrane fluidity on the functionality of the thylakoid FtsH complex under stress. Future research should explore how membrane fluidity impacts proteostasis, potentially uncovering strategies to modulate thermosensitivity.

Introduction

Sessile plants cannot evade the effects of changing environments; consequently, they have evolved complex adaptations and survival mechanisms to withstand adverse conditions, such as extreme temperatures. In the 1970s, SJ Singer and GL Nicolson proposed the fluid mosaic model to describe the overall organization and structure of proteins and lipids in biological membranes, aligning with thermodynamic constraints. They also highlighted that temperature variations impact membrane fluidity (Singer and Nicolson 1972). Later research revealed that low temperatures could influence the degree of unsaturation of fatty acids, the composition of glycerides, and the distribution of saturated and unsaturated fatty acids within lipid molecules (Lynch and Thompson 1982; Murata and Los 1997). Low temperatures unquestionably reduce membrane fluidity, while heat stress increases it (Singer and Nicolson 1972; Sinensky 1974; Cossins et al. 1978; Horvath et al. 1998; Vigh et al. 1998; Sangwan et al. 2001; Los and Murata 2004; Cano-Ramirez et al. 2021). The physical state of membrane lipids also directly regulates the activity of membrane-bound proteins (Sugiura et al. 1994; Tokishita and Mizuno 1994; Sukharev 1999; Wood 1999; Hohmann 2002; Los and Murata 2004). Hence, the stiffening of membrane lipids in response to low-temperature stress is likely the primary trigger for the corresponding adaptive cellular response (Monroy and Dhindsa 1995; Orvar et al. 2000; Suzuki et al. 2000a, 2000b; Sangwan et al. 2001, 2002; Los and Murata 2004).

The FtsH (FILAMENTOUS TEMPERATURE-SENSITIVE H) was first identified in Escherichia coli K-12 (Santos and De Almeida 1975; Begg et al. 1992). In Arabidopsis (Arabidopsis thaliana), 12 kinds of FtsH have been identified thus far. FtsH3, FtsH4, and FtsH10 are localized in the mitochondria, while the others are found in the chloroplast. Notably, FtsH1, FtsH2, FtsH5, and FtsH8 assemble into a hetero-hexamer on the thylakoid membrane (Lindahl et al. 1996; Chen et al. 2000; Takechi et al. 2000; Sakamoto et al. 2003; Yu et al. 2004, 2005; Zaltsman et al. 2005). Based on sequence similarity, FtsH1/FtsH5 and FtsH2/FtsH8 are classified into types A and B, respectively. Among these, FtsH2 exhibits the highest transcriptional abundance and steady-state protein levels, followed by FtsH5, FtsH1, and FtsH8 (Sinvany-Villalobo et al. 2004).

Multiple studies have shown that reduced levels of the FtsH complex, particularly due to the absence of its most abundant component, FtsH2, result in significant defects in photosystem II (PSII) repair and chloroplast biogenesis (Lindahl et al. 2000; Bailey et al. 2002; Kato et al. 2007; Sakamoto et al. 2009). The loss of FtsH5 also leads to notable phenotypic effects, though less severe than those observed with FtsH2, and more significant than those seen in mutants lacking FtsH1 or FtsH8 (Martinez-Zapater 1993; Chen et al. 2000; Sakamoto et al. 2003; Yu et al. 2004, 2005; Zaltsman et al. 2005). Therefore, the established functions of the FtsH complex, comprising FtsH1, FtsH2, FtsH5, and FtsH8, include promoting chloroplast biogenesis and participating in the degradation of the core components of PSII, namely D1 and D2 (Lindahl et al. 2000; Bailey et al. 2002; Duan et al. 2019). Meanwhile, it has been noted that ftsh2 [referred to as yellow variegated2 (var2)] displays more pronounced leaf variegation and reduced chlorophyll levels under cold stress compared to ftsh5 (referred to as var1). In contrast, neither ftsh1 nor ftsh8 exhibit distinguishable phenotypes relative to wild-type (WT) plants (Sakamoto et al. 2003, 2004).

Each FtsH1, FtsH2, FtsH5, or FtsH8 subunit consists of an N-terminal transmembrane domain that anchors the protein in the thylakoid membrane, a highly conserved AAA + ATPase (ATPases Associated with diverse cellular Activities) domain that provides the energy for substrate unfolding and retro-translocation to a proteolytic chamber via ATP hydrolysis, and a protease domain that catalyzes the degradation of substrate proteins (Tomoyasu et al. 1993, 1995; Ogura and Wilkinson 2001; Janska et al. 2013; Puchades et al. 2017). The N-terminal transmembrane domains of the FtsH subunits ensure proper membrane localization and help stabilize the complex within the lipid bilayer of the thylakoid membrane. The AAA + ATPase domains of the 6 subunits come together to form a central channel, through which substrate proteins are translocated from the membrane to the proteolytic chamber. The protease domains, located at the C-terminal regions of the subunits, form the proteolytic chamber where substrate proteins are degraded. The FtsH hexamer is highly enriched in the nonappressed regions of the grana, where it degrades the damaged PSII core proteins (Lindahl et al. 1996; Rodrigues et al. 2011; Wang et al. 2016). Such significant advancements have been made in understanding these domains, with mutations FtsH2^G267D and FtsH2^H488L leading to the functional inactivation of the AAA + ATPase and protease domains, respectively (Sakamoto et al. 2004; Zhang et al. 2010; Duan et al. 2019; Liu et al. 2023).

The reactive oxygen species (ROS)-driven oxidative posttranslational modifications of PSII core proteins consistently activate the PSII repair cycle, a complex process that necessitates the translation of plastid-encoded proteins and the reassembling of the PSII core (Zhang et al. 2000; Baena-Gonzalez and Aro 2002; Nowaczyk et al. 2010; Weisz et al. 2017; Kim 2019; Dogra et al. 2019b; Kato et al. 2023). This cycle involves several intricate steps: the partial disassembly of inactive PSII in the thylakoid membrane, the proteolysis of damaged core proteins by the FtsH hexamer, the precise insertion of new D1 proteins, and the reassembly of functional PSII units (Lindahl et al. 2000; Zhang et al. 2000; Bailey et al. 2002; Nowaczyk et al. 2010). This process depends on the coordinated release and re-insertion of pigments, as their free forms act as photosensitizers (Krieger-Liszkay et al. 2008; Cakiroglu et al. 2023). Chlorophyll-binding proteins such as LHC (LIGHT-HARVESTING COMPLEX)-LIKE proteins and WSCP (WATER-SOLUBLE CHLOROPHYLL PROTEIN) protect chloroplasts from oxidative damage, adding another layer of complexity to the cycle (Levin and Schuster 2023; Lee and Kim 2024). The phosphorylation of D1 proteins is also essential for the PSII repair cycle during photoinhibition and recovery (Vainonen et al. 2005). While D1 phosphorylation mainly occurs in the grana core, the actual repair processes occur in the grana margin, where the FtsH protease degrades damaged D1 proteins (Rodrigues et al. 2011; Wang et al. 2016). A recent study demonstrated that the oxidation of the N-terminal tryptophan residue of D1 acts as an N-degron signal for FtsH-mediated degradation in Chlamydomonas reinhardtii, filling the gap in understanding the initiation of the PSII repair cycle (Kato et al. 2023). Additionally, the decomposition of the oxidized unsaturated fatty acids is found to have a negative impact on the PSII repair cycle by impeding the dimerization of repaired PSII monomers, hindering the reassembly of PSII supercomplexes on grana stacks (Ji et al. 2023). Moreover, the EXECUTER1 protein, mostly enriched in the grana margin and involved in the singlet oxygen-triggered retrograde signaling, undergoes rapid turnover by FtsH hexamer, an integral part of initiating ¹O₂ signaling in Arabidopsis (Wang et al. 2016; Dogra et al. 2019b, 2022).

Independent research by several groups has significantly advanced our understanding of the PSII repair cycle, yet the impact of lipid fluidity on this cycle remains an important area of investigation. A previous study found that var2 mutants are highly susceptible to cold stress but can endure heat stress, whereas WT plants can adapt to both stress conditions (Sakamoto et al. 2004; Liu et al. 2019). Both of these abiotic stress factors lead to ROS bursts in chloroplasts (van Buer et al. 2016; Medina et al. 2021), which makes it puzzling that the absence of FtsH2 has no effect under heat stress conditions. This intriguing paradox is the focus of our research, as we aim to uncover the underlying mechanism of var2 mutant responses to cold and heat stress. We believe that the findings of this study are important for understanding PSII repair and may also provide insights into how cold stress affects the proteostasis of other integral membrane proteins.

Results

No mutant lacking the isomer of the thylakoid FtsH hexamer exhibits heat sensitivity

While the FtsH hexamer on the thylakoid membrane is involved in PSII repair and chloroplast biogenesis, FtsH11 is located on the inner membrane of the chloroplast and is indispensable to sustaining thermotolerance in Arabidopsis (Chen et al. 2006; Adam et al. 2019; Luo and Kim 2021; Yue et al. 2023). Unlike the well-characterized topology of the FtsH hexamer in the thylakoid membrane, the topology of FtsH11 remains to be elucidated. Nevertheless, similar to the Escherichia coli ftsh mutant strain, the Arabidopsis ftsh11 mutant is highly sensitive to heat stress (Chen et al. 2006). Therefore, we included the ftsh11 mutant as a control to assess the heat sensitivity of Arabidopsis plants. The effects of the loss of FtsH1, FtsH2, FtsH5, FtsH8, and FtsH11 were compared across normal, cold, and heat stress conditions (Fig. 1A). As shown in Fig. 1B, all genotypes were T-DNA knockout mutants as initially characterized by Sakamoto et al. (2003). After 5 d of growth under normal temperature conditions, no significant phenotypic differences were observed among the genotypes, except for slightly lower photosynthetic efficiency, as manifested by maximum photochemical efficiency of PSII (Fv/Fm), in var2 and ftsh11 mutants (Fig. 1, C and D). After an additional 5-d treatment under heat stress, ftsh11 displayed significant heat sensitivity compared to WT seedlings (Fig. 1, C and D). In contrast, var2 and var1 showed specific sensitivity to cold stress, while ftsh1 and ftsh8 behaved similarly to WT seedlings under both stress conditions (Fig. 1, C and D). This observation aligns with the known abundance of FtsH2 and FtsH5 in the hexamer's stoichiometry (Sinvany-Villalobo et al. 2004). When comparing var2 and var1, both of which respond to cold stress, var2 exhibited higher sensitivity in terms of Fv/Fm (Fig. 1D; Supplementary Fig. S1).

Heat stress rescues the defect in PSII activity in var2

It was previously shown that inactivating the AAA + ATPase domain leads to a leaf variegation phenotype, while inactivating the protease domain does not, although it significantly impacts the degradation rate of its substrate during photoinhibition (Zhang et al. 2010). Rather than deleting the AAA + ATPase or protease domains entirely, the research team has used single amino acid substitutions (FtsH2^G267D for the AAA + ATPase domain and FtsH2^H488L for the protease domain) to minimize substantial changes in protein structure (Zhang et al. 2010). Thus, we decided to explore the roles of AAA + ATPase and protease domains under cold versus heat stress conditions by using different var2 alleles and their transgenic lines expressing FtsH2, FtsH2^G267D, or FtsH2^H488L under the control of 35S promoter (Fig. 2, A to C). A schematic view of the allelic mutations at the FtsH2 locus, including var2-1 (a nonsense mutation, Q597*) and var2-9 [a missense mutation, G267D; also called var2-3 in Chen et al. (2000)] in Arabidopsis is shown in Fig. 2B (Chen et al. 2000; Sakamoto et al. 2002). The transgenic var2 (T-DNA knockout mutant) or var2-1 lines expressing comparable levels of FtsH2 or its variants across multiple independent lines were selected for further study (Supplementary Fig. S2, A to C). The absence of FtsH2 was confirmed by an immunoblot assay using an anti-VAR2 antibody, which showed slight cross-reactivity with the FtsH8 isomer in var2 and var2-1 (Sakamoto et al. 2003; Kato et al. 2018). Additionally, the presence of FtsH2^G267D was confirmed in the var2-9 mutant (Supplementary Fig. S2B). All genotypes, including ftsh11, were germinated and treated under normal, cold, or heat conditions, as shown in Fig. 2A. After 5 d of growth under normal conditions, no discernible phenotypic differences were observed among the genotypes (Fig. 2D). However, var2, var2-9, var2-1, and FtsH2^G267D var2-1 exhibited notably reduced Fv/Fm relative to WT (Fig. 2E; Supplementary Fig. S3). In contrast, overexpression of either FtsH2 or FtsH2^H488L rescued the decreased Fv/Fm levels in both var2 and var2-1. After an additional 5 d of exposure to cold or heat stress conditions, var2, var2-9, var2-1, and FtsH2^G267D var2-1 displayed a chlorotic phenotype under cold stress, with significantly reduced Fv/Fm levels. In sharp contrast, only ftsh11 showed a significant and rapid decrease in Fv/Fm and chlorotic phenotype under heat stress (Fig. 2, D and E; Supplementary Fig. S3; Chen et al. 2006).

Of particular interest, the lower Fv/Fm observed in all var2 alleles and FtsH2^G267D var2-1 at normal temperatures was fully restored under heat stress, suggesting a potential impact of increased lipid fluidity, induced by heat, on the recovery of PSII repair in var2 mutants (Fig. 2E; Supplementary Fig. S3).

Hexamer lacking functional FtsH2 is sufficient for D1 turnover under heat stress

It has been over 20 years since D1 was discovered as a substrate for the FtsH hexamer containing FtsH1, FtsH2, FtsH5, and FtsH8 (Lindahl et al. 2000). The D1 degradation, which is a prerequisite step for PSII repair involving multiple steps (Nixon et al. 2010; Theis and Schroda 2016; Kato and Sakamoto 2018), is a critical factor in studying the mode of action of FtsH hexamer variants, such as those containing FtsH2, FtsH2^G267D, or FtsH2^H488L. Therefore, examining the D1 turnover rate is essential for understanding the significance of AAA + ATPase or protease activity in PSII repair in var2 mutants under normal, cold, or heat stress conditions.

As such, we utilized the aforementioned genotypes, such as WT, var2-1, FtsH2^G267D var2-1, FtsH2^H488L var2-1, and FtsH2 var2-1 to examine the turnover rate of D1 under normal, cold, and heat stress conditions (Fig. 3; Supplementary Fig. S4). Given that the loss of FtsH2, which reduces the levels of the FtsH hexamer, also slows down the D2 turnover rate (Bailey et al. 2002; Yu et al. 2004; Duan et al. 2019), we examined its turnover rate as well. To observe the turnover rate of D1 and D2 proteins over time, we treated seedlings with lincomycin to inhibit chloroplast translation prior to exposure to normal, cold, or heat conditions. The immunoblot results showed that the degradation rate of D1 and D2 in WT rapidly increases under heat stress conditions. What drew our attention was that, under moderate temperature conditions, var2-1 and FtsH2^G267D var2-1 exhibited slightly slower degradation rates for D1 and D2 compared to other genotypes. This difference was more pronounced under cold stress but disappeared under heat stress (Fig. 3, A and B). Notably, the overexpression of proteolytically inactive FtsH2^H448L in var2-1 results in D1 and D2 turnover rates almost comparable to those of FtsH2 var2-1. This result indicates that having an active AAA + ATPase of FtsH2 in the hexamer is indispensable for promoting D1 and D2 turnover under cold stress conditions.

Temperature-dependent role of FtsH's ATPase function in membrane protein extraction

The fluid mosaic model, introduced by Singer and Nicolson (1972), transformed our understanding of cell membranes by illustrating how cold stress reduces membrane fluidity while heat stress increases it. We applied this principle to generate a working model of FtsH2 variants in var2 mutants, which is crucial for understanding how membrane fluidity affects PSII proteostasis under varying temperature conditions.

Our models, based on the presented data, explain why var2 mutants behave like WT under heat stress but fail under cold stress, despite both conditions causing ROS bursts and PSII damage in chloroplasts (Yamamoto et al. 2008; Yamashita et al. 2008; van Buer et al. 2016; Medina et al. 2021; Wei et al. 2022). The key conclusion is that the degradation rate of substrate proteins by the hetero-hexamer increases with temperature as membrane fluidity rises, regardless of genotype. This finding aligns with the membrane transition model proposed by Los and Murata (2004), which describes a shift from rigidity during cold stress to a liquid-crystalline phase at normal temperatures and further fluidization at higher temperatures.

With increasing temperature, extracting membrane proteins becomes easier due to heightened membrane fluidity, as observed even in FtsH2 ^G267D var2-1 (Fig. 4). The AAA + ATPase activity of other FtsH isomers in the hetero-hexamer appears sufficient to extract substrates under these conditions. However, under cold stress-induced membrane viscosity, the activity by other isomers is insufficient, underscoring the critical role of the AAA + ATPase domain of the primary isomer FtsH2 (Fig. 4).

Cold stress upregulates FtsH hexamer under decreased membrane fluidity

Based on the data from Fig. 3, the degradation rate of PSII RC proteins increases with higher temperatures, even in var2 null mutants, suggesting that membrane fluidization is sufficient for substrate extraction and subsequent turnover, despite the reduced levels of the hetero-hexamer lacking FtsH2. We then hypothesized that cold stress might require more feedback upregulation of FtsH proteins to compensate for a decrease in membrane fluidity. To investigate this, we examined the abundance of transcripts of FtsH1, FtsH2, FtsH5, and FtsH8 in WT, var2-1, FtsH2^G267D var2-1, FtsH2^H488L var2-1, and FtsH2 var2-1 after 5 d under normal conditions followed by 24 h of exposure to normal, cold, or heat conditions. Reverse transcription-quantitative PCR (RT-qPCR) analysis revealed heightened levels of FtsH1, FtsH2, FtsH5, and FtsH8 transcripts under cold stress across all genotypes compared to normal or heat conditions, along with the concurrent upregulation of cold-responsive genes, such as COLD AND ABA INDUCIBLE PROTEIN KIN1 (KIN1) and RESPONSIVE TO DESICCATION 29A (RD29A) (Fig. 5, A and B). However, no significant differences were observed under heat stress, despite the significant upregulation of heat-responsive genes, including HEAT SHOCK PROTEIN 101 (HSP101) and HEAT SHOCK TRANSCRIPTION FACTOR A2 (HSFA2) (Fig. 5, A and B). Consistent with these expression patterns, immunoblot analyses using VAR1 and VAR2 antibodies, which cross-react with FtsH1/FtsH5 and FtsH2/FtsH8, respectively, confirmed that FtsH protein levels are specifically upregulated by cold stress (Fig. 5, C and D).

Discussion

Chloroplasts” initial reaction to external stimuli, such as heat and cold stress, is characterized by an increased generation of ROS due to excess light over photochemical efficiency and overreduction of the electron transport chain in the thylakoid membrane (Yamashita et al. 2008; van Buer et al. 2016; Medina et al. 2021; Wei et al. 2022; Lee and Kim 2024). These ROS oxidize unsaturated fatty acids in the thylakoid membrane, leading to lipid decomposition and the formation of reactive carbonyl species (RCS) (Mano 2012). These RCS molecules can modify proteins, alter their functions, and act as signaling molecules, although their roles in plants remain largely unexplored. Besides, ROS produced through the photosynthetic apparatus and pigment molecules in the thylakoid membrane negatively impact net photosynthesis with a concurrent increase of oxidative modifications of target proteins (such as PSII RC proteins) (Nishiyama et al. 2006; Mizusawa and Wada 2012; Pospisil 2012; Dogra and Kim 2019; Dogra et al. 2019a, 2019b; Kato et al. 2023).

Heat and cold stress elevate ROS levels in chloroplasts, resulting in photodamages of photosynthetic apparatus (Yamamoto et al. 2008; Yamashita et al. 2008; van Buer et al. 2016; Medina et al. 2021; Wei et al. 2022). Among the chloroplast proteases, the thylakoid membrane-bound FtsH hexamer is crucial for the PSII repair cycle (Lindahl et al. 2000; Bailey et al. 2002; Zhang et al. 2010; Duan et al. 2019). FtsH2 and FtsH5 are particularly significant due to their higher proportions within the hexamer relative to the levels of FtsH1 and FtsH8 (Sakamoto et al. 2003; Sinvany-Villalobo et al. 2004; Yu et al. 2004, 2005). Indeed, mutants lacking FtsH2 (var2) and FtsH5 (var1) exhibit cold sensitivity, with FtsH2 mutants experiencing more severe damage under cold stress. However, both mutants perform similarly to WT, ftsh1, and ftsh8 plants under heat stress (Fig. 1, C and D; Liu et al. 2019). This observation prompted an investigation into why the absence of the major isomer in the FtsH hexamer has no notable impact under heat stress, despite the known accumulation of ROS in chloroplasts.

Remarkably, the basal defect in maximum PSII quantum yield in var2 mutants was rescued under heat stress conditions but exacerbated under cold stress conditions (Fig. 1, C and D; Supplementary Fig. S1). Cold sensitivity was greatly mitigated by overexpressing either FtsH2 or FtsH2^H488L (defective in proteolysis activity) but remained unchanged by overexpressing FtsH2^G267D (defective in substrate-unfolding activity). This detailed analysis indicates that, under cold stress conditions, the AAA + ATPase activity of FtsH2 is crucial for maintaining cold tolerance rather than its protease activity within the hexamer (Fig. 4). According to the fluid mosaic model, our results suggest that decreased membrane fluidity under cold stress necessitates increased activity of the AAA + ATPase domain to extract substrates from the viscous membrane. In contrast, increased membrane fluidity under heat stress diminishes the AAA + ATPase domain's significance, and the protease activity of other isomers suffices (Fig. 4).

Beyond membrane fluidity, recent reports have shown that oxidative posttranslational modification of D1 proteins serves as a protein degradation signal, activating the PSII repair cycle in Chlamydomonas (Kato et al. 2023), while lipid decomposition products hinder the repair cycle (Ji et al. 2023). Therefore, both membrane fluidity and ROS-driven lipid and protein oxidations profoundly impact protein proteostasis across the thylakoid membrane. These findings significantly advance our understanding of chloroplast proteostasis and open new research avenues on the role of membrane integrity/dynamics in the repair of the photosynthetic apparatus. It also remains unexplored whether membrane fluidity can affect chloroplast-to-nucleus retrograde signaling, given that the activity of FtsH hexamers has been implicated in ROS-triggered retrograde signaling (Dogra et al. 2017, 2018, 2019a, 2019b, 2022; Kim 2019).

In summary, the intricate interplay between ROS, lipid oxidation, and protease activity highlights the complexity of chloroplast proteostasis, particularly under different thermal conditions. Future research could focus on targeted strategies to modulate these processes, with the potential to enhance plant resilience in the face of climate change.

Materials and methods

Plant materials and growth condition

All Arabidopsis (A. thaliana) seeds used in this study were derived from the Columbia-0 ecotype and were harvested on the same day from plants grown under continuous light (CL; 100 µmol photons m⁻²s⁻¹ of light from cool-white fluorescent bulbs) at 22 ± 2 °C. The seeds for var2 (SAIL_253_A03), ftsh5 (SAIL_875_E04), and ftsh11 (SALK_033047) mutants were obtained from the Nottingham Arabidopsis Stock Centre (NASC). Seeds for the ftsh1, ftsh8, var2-1, and var2-9 mutants, as well as transgenic lines (FtsH2^G267D var2-1, FtsH2^H488L var2-1, and FtsH2 var2-1), were the same as those described by Sakamoto et al. (2003) and Zhang et al. (2010). The seeds were surface-sterilized using a 1.6% (v/v) hypochlorite solution and subsequently plated on half-strength Murashige and Skoog (MS) medium (Duchefa Biochemie, Haarlem, Netherlands) with 0.75% (w/v) Plant Agar (Duchefa Biochemie). Following 3 d of stratification at 4 °C in darkness, the seeds were transferred to a growth chamber (CU-41L4; Percival Scientific, Perry, IA, USA) under CL (100 μmol m⁻²s⁻¹ from cool-white fluorescent bulbs) at 22 °C, unless stated otherwise.

Generating Arabidopsis transgenic plants

The stop-codon-less coding sequence (CDS) of FtsH2 was amplified by polymerase chain reaction (PCR). Overlap extension PCR was used to generate the CDS of FtsH2^H488L, which contains a point mutation in the protease domain. Both CDSs were cloned into the pDONR221/Zeo Gateway vector (Thermo Fisher Scientific, Waltham, MA, USA) and subsequently recombined into the Gateway-compatible plant binary vector pGWB517 (Nakagawa et al. 2007) for C-terminal fusion with 4×Myc. The resulting binary vectors were transformed into Agrobacterium tumefaciens strain GV3101 using a MicroPulser Electroporation system (Bio-Rad, Hercules, CA, USA). Stable Arabidopsis transgenic plants in the var2 mutant background, including FtsH2^H488L-4×Myc var2 and FtsH2-4×Myc var2 were generated using the Agrobacterium-mediated floral dip transformation method (Clough and Bent, 1998). Homozygous T3 transgenic plants were selected on MS medium containing 35 mg L⁻¹ hygromycin (Thermo Fisher Scientific). The primer sequences used for vector constructions are listed in Supplementary Table S1.

Determining photochemical efficiency

Measurements of the maximum photochemical efficiency of PSII (Fv/Fm) were conducted using a FluorCam system (FC800-C/1010GFP; Photon Systems Instruments) equipped with a CCD camera and an irradiation system, following the manufacturer's instructions.

RNA extraction and RT-qPCR

Total RNA was extracted using the Universal Plant Total RNA Extraction Kit (BioTeke, Beijing, China) and quantified spectrophotometrically at 260 nm with the NanoDrop 2000 (Thermo Fisher Scientific). One microgram of total RNA was reverse-transcribed using the HiScript II Q RT SuperMix (Vazyme Biotech, Nanjing, China), following the manufacturer's recommendations. RT-qPCR was performed with the ChamQ Universal qPCR Master Mix (Vazyme Biotech) on a QuantStudio 6 Flex Real-Time PCR System (Applied Biosystems, Waltham, MA, USA). Relative transcript levels were calculated using the comparative delta-Ct method (Livak and Schmittgen 2001) and normalized to the transcript levels of ACTIN2 (At3g18780). The primers used in this study are listed in Supplementary Table S1.

Protein extraction and immunoblot analyses

Total proteins were extracted from 5-d-old seedlings treated with or without lincomycin using a homogenization buffer containing 0.0625 m Tris-HCl (pH 6.8), 1% (w/v) sodium dodecyl sulfate (SDS), 10% (v/v) glycerin, and 0.01% (v/v) 2-mercaptoethanol and quantified with the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific). For lincomycin treatment, seedlings were submerged in a solution with 5 mm lincomycin (Sigma-Aldrich, St. Louis, MO, USA) and 0.002% (v/v) Silwet L-77 (GE Healthcare, Chicago, IL, USA) in the dark for 30 min. The seedlings were then incubated under CL at different temperatures and collected at the indicated time points. Total proteins were separated on 10% SDS-PAGE gels and blotted onto Immun-Blot PVDF membranes (Bio-Rad). D1, D2, FtsH1/FtsH5, FtsH2/FtsH8, and histone H3 proteins were immunochemically detected using rabbit anti-D1 (1:10,000; catalog no. AS05 084; Agrisera, Vännäs, Sweden), rabbit anti-D2 (1:10,000; catalog no. AS06 146; Agrisera), rabbit anti-VAR1 (1:5,000; catalog no. PAB13001; Orizymes, Shanghai, China), and rabbit-VAR2 (1:5,000; catalog no. PAB13002; Orizymes), and rabbit anti-histone H3 (1:3,000; catalog no. AS10 710; Agrisera) antibodies, respectively.

Statistical analyses

Statistical analyses were performed using GraphPad Prism (version 8.0.2, La Jolla, CA, USA), with detailed results provided in Supplementary Data Set 1. Information on the statistical methods and the number of biological replicates used is included in the figure legends.

Accession numbers

Sequences of the genes studied in this article can be found in the Arabidopsis TAIR database (https://www.arabidopsis.org) under the following accession numbers: FtsH1 (AT1G50250), FtsH2 (AT2G30950), FtsH5 (AT5G42270), FtsH8 (AT1G06430), FtsH11 (AT5G53170). Actin2 (AT3G18780), psbA (ATCG00020), psbD (ATCG00270), KIN1 (AT5G15960), RD29A (AT5G52310), HSP101 (AT1G74310), and HSFA2 (AT2G26150).

Acknowledgments

We thank Prof. Wataru Sakamoto (Okayama University, Japan) for kindly providing seeds of the mutants ftsh1, ftsh8, and var2-1, as well as the transgenic lines FtsH2^G267D var2-1, FtsH2^H488L var2-1, and FtsH2 var2-1.

Author contributions

C.K. conceived the presented ideas and designed the research. J.Z. and Y.L. conducted the experiments. J.Z., K.P.L., and C.K. analyzed the data. C.K. wrote the manuscript with input from J.Z.

Supplementary data

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

Supplementary Figure S1. The Fv/Fm values of ftsh mutants under normal, cold, and heat conditions.

Supplementary Figure S2. Transcript and protein abundances of FtsH2 (and/or FtsH8) in var2 mutants and var2 transgenic lines.

Supplementary Figure S3. The Fv/Fm values of var2 mutants and var2 transgenic lines under normal, cold, and heat conditions.

Supplementary Figure S4. Quantification of D1 and D2 turnover rates in mock-treated samples.

Supplementary Table S1. List of primers used in this study.

Supplementary Data Set 1. Table for statistical analysis.

Funding

This research was supported by the National Natural Science Foundation of China (NSFC) (grant nos. 32370249 and 32350710188) to C.K.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Dive Curated Terms

The following phenotypic, genotypic, and functional terms are of significance to the work described in this paper:

• FtsH2 Gramene: AT2G30950

• FtsH2 Araport: AT2G30950

• FtsH5 Gramene: AT5G42270

• FtsH5 Araport: AT5G42270

• FtsH8 Gramene: AT1G06430

• FtsH8 Araport: AT1G06430

• FtsH11 Gramene: AT5G53170

• FtsH11 Araport: AT5G53170

References

Author notes