The C. elegans strains used in this study were N2 Bristol (wild-type) and acox-1.1(ok2257) mutant worms. All worms were cultured as previously described [27].

RNAi was performed using the feeding method. The ctl-3(RNAi) clone was constructed by amplifying a specific genomic region (approximately 800 bp) of the ctl-3 gene from C. elegans genomic DNA using the following primers containing XbaI and HindIII restriction sites at their 5′ ends: forward, 5′-ACT​TCT​AGA​CCG​CCG​TGC​TCA​CCG​CC-3′ and reverse, 5′-TCT​AAG​CTT​GTT​CGA​ATG​TCA​TCA​CTT​G-3’. The resulting PCR product was digested with XbaI and HindIII and ligated into the L4440 vector at the corresponding restriction sites. The resulting construct was then transformed into E. coli HT115 (DE3) RNAi bacteria. For feeding RNAi experiments, ctl-3 RNAi bacteria clones were cultured overnight in LB medium at 37 °C. The cultured bacteria were then seeded onto NGM agar plates containing 1 mM IPTG and 100 μg/mL ampicillin to induce double-stranded RNA expression.

Total RNA was extracted from age-synchronized worms using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA; Thermo Fisher Scientific, Waltham, MA, USA) and purified with the RNeasy Kit (QIAGEN, Hilden, Germany). Reverse transcription was performed using the Transcriptor First Strand cDNA Synthesis Kit (Roche, Basel, Switzerland) with oligo(dT) primers. Quantitative real-time PCR (qRT-PCR) was conducted using iQ SYBR Green Supermix (Bio-Rad Laboratories, Hercules, CA, USA) according to the manufacturer’s instructions. Amplification products were analyzed on a CFX Connect™ Real-Time PCR Detection System (Bio-Rad Laboratories, Hercules, CA, USA). Relative mRNA expression levels were calculated using the ΔΔCT method, with ama-1 or act-1 serving as reference genes for normalization. It minimizes potential errors caused by subtle fluctuations in a single housekeeping gene, thereby ensuring a more reliable normalization base. All biological replicates within each experiment were consistently normalized using ama-1 as the reference gene. The primers used for qRT-PCR in this study are listed in Supplementary Table S1.

Measurement of ROS was performed as described [28]. Worms were transferred to 100 µL PBS-T (0.1% Tween 20) using L4-stage worms. The worms were frozen in liquid nitrogen and then sonicated for 20 s and 90 s, a process repeated five times. All samples were analyzed with five technical replicates in 96-well black-bottom, clear plates. Samples were incubated with 50 µM DCF-DA at 37 °C. Fluorescence was measured using the Victor 5 multilabel plate reader (Perkin Elmer, USA) with excitation at 485 nm and emission at 640 nm for 2 h, with readings taken every 10 min. The experiments were normalized to PBS-only samples, with five technical replicates for each condition.

L4-stage wild-type and acox-1.1(ok2257) worms were exposed to 1 mM PQ and basal conditions, with Day 0 corresponding to the L4 stage. For each experiment, 36 to 60 worms were used per biological repeat. All assays were performed with 3 or 4 independent replicates at 20 °C. Statistical analyses of lifespan curves were performed using the log-rank (Mantel-Cox) test in Prism 10. Hazard ratios were calculated with the Cox proportional-hazard model.

Wild-type and transgenic worms were harvested following PQ treatment and lysed in Laemmli sample buffer (Bio-Rad Laboratories, Inc., Hercules, CA, USA) by boiling for 10 min. Equal amounts of protein were separated by 10% SDS-PAGE (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Protein-transferred membranes were blocked with 5% (w/v) non-fat milk (Biotium, Inc., Fremont, CA, USA) in Tris-buffered saline (TBS) containing 0.1% Tween 20 (TBST) for 30 min at 25 °C. Primary antibodies (anti-total Ub, Santa Cruz 1:200, anti-Tub, 1:1000, DSHB AA4.3) were incubated overnight at 4 °C in 5% non-fat dry milk in TBST, followed by incubation with secondary antibodies (HRP-goat anti-mouse 1:5000, Invitrogen 31430) for 2 h at 25 °C. Western blot images were obtained using the LI-COR C-DiGit Chemiluminescence Western Blot Scanner and Image Studio software (LI-COR, Lincoln, Nebraska, USA).

We performed as previously described [29]. Worms were washed with distilled water and resuspended in sample buffer A (50 mM Tris, 5 mM EDTA, 7 M urea, 2 M thiourea, 4% CHAPS, and protease inhibitor). The suspensions were sonicated on ice for 30 s, and the soluble fractions were obtained by centrifugation at 36,000 × g for 40 min at 4 °C. Protein concentrations were determined using the Bradford method. Aliquots were stored at −70 °C until use. Protein samples (100 µg for analytical gels and 1 mg for preparative gels) were diluted in sample buffer B (7 M urea, 2 M thiourea, 2% v/v IPG buffer, pH 3–10 nonlinear (Amersham Biosciences), 2% CHAPS, 15 mM dithiothreitol, and a trace of bromophenol blue). The protein mixtures were applied onto IPG strips (Immobiline DryStrip, pH 3–10 nonlinear, 18 cm; Amersham Biosciences) prehydrated with the protein solution at 20 °C for 14 h. Isoelectric focusing (IEF) was performed at 20 °C with a current limit of 50 mA/strip using the following protocol [29]. After focusing, IPG strips were equilibrated for 20 min with gentle agitation in buffer containing 375 mM Tris-HCl (pH 8.8), 6 M urea, 2% SDS, 5 mM tributyl phosphine, 2.5% acrylamide solution, and 20% glycerol. For the second dimension, vertical SDS-PAGE was performed using 9–16% gradient slab gels (180 × 200 × 1.5 mm). Equilibrated strips were trimmed and sealed onto the gels with 0.5% agarose in running buffer (24.8 mM Tris (pH 8.3), 192 mM glycine, 0.1% SDS, and a trace of bromophenol blue). Electrophoresis was performed at a constant current of 15 mA per gel. Gels were fixed in 40% methanol and 5% phosphoric acid for at least 1 h and stained overnight with Coomassie Brilliant Blue G-250, destained, scanned using a GS-710 image scanner (Bio-Rad), and analyzed with Melanie 3 software (GeneBio).

We performed as previously described [29]. Protein spots were excised from the gels using pipette tips with the ends removed to accommodate varying diameters. Excised gel pieces that were transferred to microtubes were destained and dehydrated in 50 µL acetonitrile for 5 min at room temperature. The dried gels were rehydrated on ice for 45 min with 10 µL of trypsin solution (10 mg/mL in 25 mM ammonium bicarbonate, pH 8.0). Excess solution was removed, and proteins were digested at 37 °C for 24 h. The resulting peptides were extracted with POROS R2 beads [30] and analyzed using a Voyager DE Pro MALDI-TOF mass spectrometer (Applied Biosystems, Foster City, CA, USA). For sample preparation, 0.5 µL of α-cyano-4-hydroxycinnamic acid matrix solution was mixed with the peptide solution. Time-of-flight (TOF) measurements were performed under the following conditions: accelerating voltage of 20 kV, grid voltage of 75%, guide wire voltage of 0%, a delay time of 120 ns, and a low-mass gate of 500 Da. Internal calibration was performed using the autodigestion peaks of porcine trypsin (M + H⁺, m/z 842.5090 and 2211.1064). Peptide mass profiles were searched against the National Center for Biotechnology Information (NCBI) database using MS-Fit 3.2 (University of California, San Francisco;¹) and ProFound (version 4.7.0; The Rockefeller University;²), applying a mass tolerance of 20 ppm for measurements acquired in reflector mode.

To examine the effects of acox-1.1 deficiency on lifespan under oxidative stress, we compared the survival of wild-type and acox-1.1(ok2257) mutant worms under basal conditions and in the presence of 1 mM paraquat (PQ). Under normal growth conditions, acox-1.1(ok2257) mutants exhibited a significantly shorter mean lifespan than wild-type worms (Figures 1A,B), consistent with impaired metabolic homeostasis. Upon PQ treatment, wild-type worms showed a significant reduction in mean lifespan relative to untreated conditions (Figures 1A,B). In contrast, acox-1.1(ok2257) mutants did not exhibit PQ-induced lifespan reduction and instead maintained a comparable or slightly increased mean lifespan under oxidative stress (Figures 1A,B). Consistently, Log-rank (Mantel-Cox) analysis revealed that wild-type worms showed a significant difference in survival distributions with and without PQ treatment (p < 0.001), whereas acox-1.1(ok2257) mutants exhibited no significant difference under the same conditions (Figure 1A). Notably, acox-1.1(ok2257) mutants maintained high viability during early and mid-adult stages under PQ treatment, with a delayed onset of mortality compared to wild-type worms. These results indicate that loss of acox-1.1 confers enhanced resistance to oxidative stress, potentially through metabolic adaptations or activation of stress-response pathways. This interpretation is consistent with previous studies linking peroxisomal function to lifespan regulation in C. elegans [31] and supports the notion that metabolic rewiring in response to oxidative stress contributes to stress resistance and survival outcomes [31, 32].

Since acox-1.1 plays a role in peroxisomal β-oxidation, we next investigated whether its loss alters ROS homeostasis. Intracellular hydrogen peroxide (H₂O₂) accumulation was measured using DCF fluorescence intensity over a 2-h period (Figure 1C; Supplementary Figure S1). The results showed that acox-1.1(ok2257) mutants consistently exhibited higher DCF fluorescence intensity than wild-type controls at all time points, indicating a sustained increase in endogenous ROS levels. At 120 min, acox-1.1(ok2257) mutants displayed a marked elevation in fluorescence signal, suggesting a significant accumulation of ROS (p < 0.001) (Supplementary Figure S1). The continuous rise in fluorescence indicates that the loss of acox-1.1 disrupts peroxisomal function, leading to impaired H₂O₂ detoxification and an oxidative stress-prone cellular state. In previous studies, acox-1.1 (RNAi) did not significantly alter H₂O₂ levels compared with wild-type worms [31], whereas acox-1.1(ok2257) mutants accumulate internal ROS. Despite increased ROS levels, the prolonged survival of acox-1.1(ok2257) mutants under oxidative stress conditions implies the presence of adaptive mechanisms that mitigate ROS-induced damage, possibly through enhanced antioxidant defense or stress tolerance pathways.

Given that the acox-1.1(ok2257) mutants lead to elevated endogenous levels of ROS, we examined the expression of C. elegans catalase genes involved in ROS detoxification to assess potential compensatory responses. In C. elegans, ctl-1 encodes a cytosolic catalase, whereas ctl-2 and ctl-3 encode peroxisomal catalases, highlighting distinct subcellular localization among the family members [33]. Elevated catalase activity can mitigate tissue metabolic damage by reducing toxic H₂O₂ levels [34]. qRT-PCR analysis revealed that the expression of ctl-1 and ctl-2 genes remained unchanged (p > 0.05) compared to wild-type worms, whereas ctl-3 was significantly upregulated (p = 0.0006) (Figure 1D). These results suggest that loss of acox-1.1 alters the expression of ROS scavenging genes, with ctl-3 exhibiting the most pronounced induction, potentially compensating for increased ROS levels. Peroxisomes are increasingly recognized as key organelles in cellular adaptation to oxidative stress, given their capacity to both generate and detoxify ROS. Both ctl-1 and ctl-2 have been implicated in lifespan regulation [35], while the role of ctl-3 remains uncharacterized. A recent study demonstrated that peroxisomes actively participate in the cell-wide response to transient heat shock, revealing a critical role for organelle communication in coordinating organismal stress responses [36]. Under such extreme oxidative environments, the cell appears to shift its energetic priorities toward a survival-centric mode, prioritizing the clearance of oxidized proteins via specific antioxidant defenses such as CTL-3. We observed that the oxidative stress resistance of acox-1.1(ok2257) mutants is heavily dependent on this axis; indeed, ctl-3 RNAi completely abolishes this resistance and leads to a short lifespan (Figure 1E). This suggests that, in the face of severe oxidative stress, the failure of the Ub-proteasome system (UPS) complex assembly is superseded by a CTL-3-mediated stress clearance pathway to maintain cellular proteostasis and survival.

To assess the effect of oxidative stress on acox-1.1 transcription, we quantified acox-1.1 mRNA levels using qRT-PCR in wild-type worms with and without PQ treatment. As shown in Figure 1F, acox-1.1 expression was significantly downregulated in PQ-treated wild-type worms compared to untreated controls (p = 0.034). This result suggests that oxidative stress negatively regulates acox-1.1 transcription, possibly as part of a broader metabolic adaptation to ROS accumulation. We also analyzed the mRNA levels of gcs-1, ptps-1, gst-4, and mtl-1 to determine whether oxidative stress modulates the expression of key antioxidant defense genes (Figure 1G). Consistent with previous reports, SKN-1 target detoxification genes were suppressed under mild chronic oxidative stress, likely conserving energy to support adaptive responses rather than acute detoxification [20, 37]. Notably, only one of the daf-16 downstream genes, mtl-1, was significantly upregulated following PQ treatment. This finding suggests that acox-1.1(ok2257) mutants may adapt through preferential activation of DAF-16 rather than SKN-1 under mild chronic oxidative stress. Together, these findings indicate that oxidative stress triggers transcriptional alterations that downregulate acox-1.1 expression and induce a distinct ROS-response pathway from the canonical SKN-1-mediated oxidative stress defense, thereby enhancing stress resilience.

To identify proteins differentially expressed in acox-1.1(ok2257) mutants that may contribute to their oxidative stress-resistant lifespan extension, we performed two-dimensional (2D) gel electrophoresis to compare global protein expression profiles with wild-type worms (Figure 2A). Several protein spots displayed altered intensity in acox-1.1(ok2257) mutants, indicating widespread changes in protein abundance. Among the significantly altered proteins, PAS-5, a key regulatory subunit of the 20S proteasome, exhibited an approximately one-third reduction in acox-1.1(ok2257) mutants compared to wild-type worms. PAS-5 was detected in the wild-type proteome (white circle, left panel), whereas its levels were markedly diminished in acox-1.1(ok2257) mutants (right panel). In addition, CCO-2 (cytochrome c oxidase subunit 2, COX2) showed a notable increase in acox-1.1(ok2257) mutants. These changes were further validated by a zoomed-in analysis (Figure 2B). Given the essential role of PAS-5 in proteasome assembly and function, these findings suggest that peroxisomal dysfunction may indirectly impact proteostasis by modulating proteasomal activity, possibly through metabolic stress-induced alterations in protein turnover. Moreover, acox-1.1(ok2257) mutants manage ROS through SKN-1-independent pathways, potentially involving DAF-16, with increased mitochondrial CCO-2, supporting oxidative stress adaptation. Under basal conditions, reduced PAS-5 levels in acox-1.1(ok2257) mutants are associated with impaired proteostasis, as reflected by the accumulation of ubiquitinated proteins, consistent with incomplete assembly or reduced activity of the 20S proteasome core. Surprisingly, upon exposure to low-dose oxidative stress, this phenotype is reversed, with a marked reduction in ubiquitin-conjugated protein smearing (Figures 2C,D). We suggested that this decrease in ubiquitinated substrate under oxidative stress does not reflect restored proteasome activity, but rather inhibition of the ubiquitination machinery itself. Elevated reactive oxygen species (ROS) are known to oxidize critical cysteine residues within the catalytic sites of E1, E2, and E3 enzymes, thereby impairing ubiquitin activation and conjugation steps [38, 39]. Together, these findings suggested that PAS-5 downregulation and oxidative stress-mediated suppression of ubiquitination converge to reshape proteostasis in acox-1.1(ok2257) mutants.