The Biological Clock Protein CLOCK May Affect Resistance to Cell Damage Through the ACHL-Relish Pathway in Fruit Flies

The circadian clock plays a fundamental role in maintaining physiological homeostasis, yet its contribution to oxidative stress adaptation remains incompletely understood. To explore the effects of oxidative stress on circadian gene expression, we first exposed flies to paraquat (PQ)‐containing food to induce oxidative stress under 12:12 h light–dark (LD) conditions, and collected fly head samples every 6 h for 48 h (Figure1A). Indeed, PQ treatment markedly altered both the amplitude and overall expression levels of core pacemaker genes (clk, cyc, per, and tim), with minimal impact on their circadian phase compared with flies maintained on a standard diet (Figure 1B). Among these, clk exhibited the most pronounced changes compared to other genes (Figure 1B). Cosinor‐circacompare analysis^[38^] further confirmed significant changes in mesor, amplitude, and phase parameters (Figure 1C), indicating that oxidative stress disrupts the endogenous circadian pacemaker genes expression. We next assessed the physiological relevance of these changes using circadian gene mutants (clk^out, per¹, tim¹) upon PQ exposure at ZT (zeitegeber time) 1 (1 h after light on). While functional interrogation of other circadian mutants revealed no obvious phenotype, notably, clk^out flies of both sexes exhibited significantly prolonged survival under 20 mm PQ compared to iso³¹ controls (Figure 1D,E). In contrast, clk^out flies exhibited a reduced lifespan when maintained on regular food (Figure S1A, Supporting Information). This survival advantage persisted at higher PQ concentrations and was recapitulated under another oxidation inducer‐rotenone induced stress (Figure 1F; Figure S1B,C, Supporting Information), highlighting a specific role for clk in modulating oxidative resilience. Based on these findings, clk was selected for further mechanistic investigation. Surprisingly, despite improved survival, clk^out flies accumulated elevated levels of reactive oxygen species (ROS) in the brain, as revealed by DCFH‐DA staining (Figure S1D, Supporting Information), suggesting that enhanced oxidative resistance stems from increased ROS tolerance rather than decreased ROS production.

Next, we further dissected the cellular basis of this phenotype by selectively knocking down clk using GAL4/UAS system. Pan‐neuronally downregulation of clk with different RNAi lines using Elav‐GAL4 indeed phenocopied the effects of clk^out mutants, suggesting that clk is mainly required in neurons for the oxidative resistance. Specifically, we further restricted clk downregulation in the circadian cells with TD2 (tim‐GAL4, UAS‐dicer2), and observed a similar phenotype with pan‐neuronal knockdown, suggesting that clk is mainly required in circadian neurons for this phenotype. Consistent with this, clk knockdown within master pacemaker neurons using PDF‐GAL4 also significantly extended fly survival upon PQ exposure (Figure 1G,I; Figure S1E, Supporting Information). A comparable phenotype was observed following cyc knockdown in the same neuronal populations (Figure S1F,H, Supporting Information), implicating the CLK–CYC transcriptional complex as a key mediator of oxidative stress susceptibility. Notably, restricting knockdown to PDF‐negative circadian neurons using PDF‐GAL80 partially attenuated the survival benefit (Figure 1J). In contrast with the downregulation, overexpression of clk in PDF‐positive neurons reduced fly survival (Figure 1K), supporting a detrimental role for neuronal CLK during oxidative challenge. Lastly, targeted knockdown of clk or cyc in noncircadian mushroom body neurons using MB‐GAL4 failed to confer any survival benefit (Figure S1I,J, Supporting Information), suggesting that the oxidative stress resistance phenotype is mediated by specific subsets of circadian neurons.

To exclude the possibility that the survival difference is due to altered PQ intake, feeding behavior was evaluated using the CAFE assay. In fact, clk^out flies consumed significantly more PQ food over 24 and 48 h compared to iso³¹ controls (Figure S1K,N, Supporting Information), indicating that increased survival is likely attributable to intrinsic physiological mechanisms rather than reduced intake or toxin avoidance.

Together, these findings position CLK as a critical regulator of organismal vulnerability to oxidative damage. Depletion of CLK—particularly within PDF‐positive pacemaker neurons—enhances survival under oxidative stress, highlighting a defined neural axis through which circadian transcription factors modulate stress resilience.

Building upon the observation that clk^out mutants exhibit enhanced resilience to oxidative stress, we conducted transcriptomic profiling of fly head to uncover the molecular basis of this phenotype. Differential expression analysis identified 345 upregulated and 353 downregulated genes in clk^out flies relative to iso³¹ controls (Figure2A). Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment of total differentially expressed genes (DEGs) identified innate immune pathways, including Toll and IMD signaling, as top‐ranked categories (Figure 2B). Stratifying DEGs into up‐ and down‐regulated subsets further refined this landscape: genes upregulated in clk^out were significantly enriched for immune‐related pathways, whereas downregulated genes were primarily associated with the MAPK signaling pathway (Figure 2C). These findings suggest a global transcriptomic shift favoring immune activation–related signaling pathways in clk^out mutants. This transcriptional reprogramming was further supported by Gene Ontology (GO) enrichment, which highlighted biological processes such as “defense response” and “response to bacterium” (Figure 2D). GO biological process (BP) enrichment of upregulated genes further emphasized pathways involved in innate immunity, including “response to bacterium” and “defense response to Gram‐positive bacterium” (Figure 2E). In line with these findings, heatmap analysis demonstrated increased basal expression of AMP‐related genes, including Mtk, Drs, and AttA, in clk^out flies (Figure 2F), indicating that CLK functions to suppress immune gene expression. Together, these results collectively indicate that CLK plays a critical repressive role in regulating innate immune gene expression under normal conditions.

To determine whether this immune activation persists during prolonged oxidative challenge, another RNA‐seq analysis was conducted after 24 h of continuous PQ exposure. Similarly, volcano plot analysis again revealed significant upregulation of AMP‐related genes in clk^out flies (Figure S2A, Supporting Information). Consistently, GO and KEGG enrichment analyses identified persistent activation of innate immune pathways, including Toll and IMD signaling (Figure S2B, Supporting Information), suggesting a prolonged immuno‐transcriptomic reprogramming in the absence of clk and upon PQ exposure.

To evaluate the functional consequences of immune priming, we first subjected flies to microbial challenges to activate the immune response and then put them under oxidative stress. Interestingly, heat‐inactivated E. coli injection prior to PQ exposure significantly improved survival in both wild‐type and clk mutants compared to PBS‐injected controls, while clk^out flies consistently exhibiting superior resistance relative to iso³¹ (Figure S2C, Supporting Information). These findings suggest that clk deficiency enhances the ability to tolerate combinatorial stress, consistent with the notion that a prior stimulation of immune response confers protection against subsequent oxidative stress. Supporting this, qRT‐PCR analysis revealed E. coli‐induced expression of AMP gene AttA in both genotypes, albeit with differing magnitudes and temporal patterns (Figure S2D, Supporting Information), pointing to altered immune dynamics in the absence of clk.

In addition to immune modulation, clk loss influenced redox homeostasis. Time‐course qPCR analyses showed that transcript levels of key antioxidant enzymes (sod1, cat, gstd1) were comparable between clk^out and control, suggesting that CLK has a minimal effect on the antioxidant gene expression under oxidative stress treatment (Figure S3A, Supporting Information). However, catalase (CAT) enzymatic activity was consistently elevated in clk^out flies across multiple time points (Figure S3B, Supporting Information), reflecting enhanced functional oxidative defense. Intriguingly, baseline hydrogen peroxide (H₂O₂) concentrations were significantly higher in clk^out flies than iso³¹ (Figure S3C, Supporting Information), suggesting an elevated oxidative burden that may act as a compensatory cue to activate enzymatic antioxidant responses. Together, these results indicate that CLK coordinates both immune and redox networks to determine systemic stress susceptibility.

Collectively, these findings demonstrate that clk deficiency induces constitutive activation of innate immune pathways—particularly AMP production—thereby reshaping the immune landscape and enhancing resilience to oxidative and immunological insults.

CLK regularly functions as a transcriptional activator; given the repressive role of CLK in basal immune gene expression, we next sought to identify downstream effectors of CLK. Transcriptomic profiling revealed Achilles (Achl), a previously identified circadian output gene, as one of the top downregulated genes in clk^out flies (Figure 2F). Consistent with this, reanalysis of the published transcriptome data^[37^] from pan‐neuronal Achl knockdown flies (Elav‐GAL4;; UAS‐dicer2/Achl^RNAi) showed robust upregulation of AMP genes (Figure S4A,B, Supporting Information), resembling the immune activation profile observed in clk^out flies (Figure S4C, Supporting Information). Next, we performed comparative transcriptomic analysis between clk^out and Achl‐deficient flies, and identified 65 overlapping DEGs (Figure3A; Figure S4D, Supporting Information), many of which are enriched in immune and stress‐response pathways. Pathway enrichment analyses revealed shared activation of the Toll, IMD, and humoral immune signaling pathways (Figure 3B,C). Chord diagram visualization demonstrated convergence on functional categories such as Gram‐positive bacterial defense and humoral antimicrobial responses, with central nodes including AttA, Mtk, and Drs (Figure 3D). These results support a shared transcriptional program linking CLK and ACHL in the regulation of innate immune responses. Temporal profiling of AMP gene expression revealed that clk^out flies exhibit both elevated and dysregulated AMP induction compared to controls (Figure 3E,F; Figure S4E, Supporting Information), reinforcing the notion that CLK suppresses innate immune activation under basal conditions. Notably, pan‐neuronal knockdown of Achl also resulted in robust upregulation of multiple AMPs—including Mtk, Drs, Dro, and AttA—mimicking the transcriptional phenotype of clk^out flies (Figure 3G). These parallel outcomes suggest that Achl may act downstream of CLK in restraining immune gene expression. Supporting this, Achl transcript levels were markedly reduced in clk^out flies (Figure S4F, Supporting Information), while clk expression remained unchanged upon Achl knockdown (Figure S4G, Supporting Information), indicating a unidirectional regulatory relationship whereby CLK positively regulates Achl to maintain immune homeostasis.

Next, we examined whether Achl plays a similar role in oxidative stress resistance. We used Elav‐GAL4 to knock down Achl pan‐neuronally and checked the survival upon PQ exposure. Indeed, loss of Achl significantly improved survival under oxidative stress (Figure 3H,I), phenocopying clk^out flies. To further investigate the regulation of Achl by the circadian clock, we examined its temporal expression pattern under LD conditions. Consistent with the previous report, Achl mRNA exhibited robust circadian oscillation, peaking at ZT21 in wild‐type flies (Figure 3J). In contrast, both the rhythmic expression pattern and amplitude of Achl were markedly diminished in clk^out mutants (Figure 3J), indicating that CLK regulates the circadian transcription of Achl. CLK activates downstream clock‐controlled gene expression via binding of E‐box motifs (CACGTG). So, we examined the Achl promoter for potential CLK‐binding sites. Promoter Sequence analysis of the Achl promoter revealed the presence of multiple canonical E‐box elements (Figure 3K; Figure S4H, Supporting Information), suggesting a potential direct regulation. To functionally validate this interaction, we performed dual‐luciferase reporter assays. Overexpression of CLK led to dose‐dependent activation of the Achl promoter in vitro (Figure 3L; Figure S4I, Supporting Information). Moreover, mutation of E‐box motifs significantly reduced luciferase activity, demonstrating that CLK activates Achl transcription via these regulatory elements (Figure 3M). Collectively, these findings establish Achl as a pivotal rhythmic factor acting downstream of CLK to orchestrate innate immune restraint and promote organismal resilience under oxidative stress.

The observation of immune hyperactivation in both Achl‐deficient and clk^out flies prompted us to explore the downstream effectors linking the CLK–ACHL pathway to immune responses. Relish, the Drosophila NF‐κB homolog and terminal effector of the IMD pathway, emerged as a likely candidate. In flies with Achl knockdown, both Relish mRNA and protein levels were significantly elevated (Figure4A–C), indicating that ACHL acts as a negative regulator of Relish abundance.

Given that Achl encodes an RNA‐binding protein and its deficiency leads to elevated Relish expression, we hypothesized that ACHL may post‐transcriptionally regulate Relish by directly binding its mRNA. To explore this possibility, we first conducted predictive RNA–protein interaction analyses by RNA‐Protein Interaction Prediction (RPISeq),^[39, 40^] which revealed a putative interaction between ACHL and Relish mRNA (Figure S5A, Supporting Information). Consistently, AlphaFold3‐based structural modeling identified multiple potential ACHL‐binding sites along the Relish transcript (Figure S5B, Supporting Information), suggesting a direct post‐transcriptional regulatory interface. These predictions support a model in which ACHL binds to Relish mRNA to modulate its stability. Next, we performed RNA immunoprecipitation followed by qPCR (RIP–qPCR) by overexpression of a tagged version of ACHL and validated direct binding of ACHL to Relish mRNA at multiple predicted sites, with region 2 (1800–2000 bp) showing the highest enrichment (Figure 4D,E). Furthermore, RNA stability analysis demonstrated that overexpression of ACHL led to reduced Relish mRNA levels and decreased transcript stability (Figure 4F; Figure S5C, Supporting Information), consistent with the prediction that ACHL acts as an RNA destabilizer. Indeed, qPCR profiling of Relish expression in the fly head confirmed increased transcript abundance in the clk^out (Figure S5D, Supporting Information). Lastly, western blot analysis further revealed elevated Relish protein levels under both basal and PQ‐treated conditions in clk^out flies (Figure 4G–J; Figure S5E,F, Supporting Information). These findings establish Relish as a key downstream effector of the CLK–ACHL axis, whereby loss of Achl‐mediated post‐transcriptional repression leads to sustained Relish activation and immune gene overexpression.

Relish, which functions as a transcription factor, needs to translocate into the nucleus to exert transcriptional control over AMP genes. We next assessed N‐terminal Relish (N‐Relish, active) protein levels. Nuclear N‐Relish accumulation was significantly higher in clk^out flies than in controls (Figure 4K), indicating enhanced NF‐κB signaling. Interestingly, full‐length cytoplasmic Relish was also elevated (Figure 4L), suggesting dysregulation of both activation and processing dynamics. Functionally, circadian neuronal knockdown of Relish using tim‐GAL4, UAS‐dicer2 sensitized flies to PQ, leading to reduced survival compared to controls (Figure 4M). Importantly, simultaneous depletion of clk and Relish produced an intermediate survival phenotype—significantly lower than clk^out but improved relative to Relish knockdown alone (Figure 4N). This genetic interaction supports Relish as a downstream effector mediating the protective phenotype observed in clk^out flies. In parallel, AMP gene expression was substantially reduced in Relish‐depleted flies (Figure 4O), further confirming its essential role in AMP gene induction. Consistent with the proposed regulatory hierarchy, RNAi‐mediated knockdown of Relish significantly reduced its mRNA levels in both iso³¹ and clk^out backgrounds (Figure S5G,H, Supporting Information). In contrast, clk expression remained unaffected, and Achl expression exhibited only a modest increase upon Relish depletion (Figure S5I, Supporting Information), further supporting a unidirectional regulatory model in which CLK and ACHL function upstream of Relish.

Together, these findings define a circadian‐regulated post‐transcriptional mechanism in which CLK drives Achl expression, and ACHL suppresses Relish via direct mRNA destabilization. Disruption of this regulatory axis leads to Relish accumulation, overactivation of AMP genes, and enhanced stress resistance, positioning ACHL as a critical RNA‐binding repressor at the intersection of circadian and immune pathways.

To consolidate the mechanistic insights uncovered in this study, we propose a model in which the circadian transcription factor CLK orchestrates immune regulation and oxidative stress adaptation through a post‐transcriptional axis involving ACHL and the NF‐κB homolog Relish (Figure5). In wild‐type flies, CLK binds to E‐box motifs in the Achl promoter, thereby driving its rhythmic transcription. The resulting ACHL protein interacts with Relish mRNA to promote its degradation. This post‐transcriptional repression limits Relish protein accumulation and consequently suppresses the transcription of downstream AMP genes such as Mtk, Drs, Dro, and AttA. This multilayered circadian regulation serves to constrain immune activation under oxidative challenge, thus preserving immune homeostasis.

Disruption of clk abrogates this regulatory axis by downregulating Achl, resulting in the stabilization of Relish transcripts, accumulation of both full‐length and cleaved nuclear Relish, and subsequent overactivation of AMP gene expression. Paradoxically, this immune hyperactivation is associated with enhanced survival in clk^out flies under oxidative stress conditions. These results suggest that in the absence of circadian gating, derepression of immune effectors may act as a compensatory mechanism that augments host tolerance to oxidative damage.

Collectively, our findings define a circadian checkpoint wherein CLK modulates immune tone and redox adaptation via ACHL‐mediated post‐transcriptional repression of Relish. This CLK–ACHL–Rel regulatory cascade exemplifies a temporal interface between the circadian clock and innate immune signaling, revealing that disruption of clock components can reprogram stress responsiveness by reshaping immune homeostasis.

To evaluate whether the regulatory function of CLK in oxidative stress adaptation is evolutionarily conserved, we turned to a mammalian cell model. NIH3T3 cells were exposed to increasing concentrations of H₂O₂, which elicited a dose‐dependent accumulation of intracellular ROS as detected by DCFH‐DA fluorescence staining (Figure S6A, Supporting Information). This ROS elevation was accompanied by a marked reduction in cell viability, assessed by MTT assay (Figure S6A, Supporting Information), confirming successful induction of oxidative stress. Consistently, flow cytometric analysis using a FITC‐based ROS probe further confirmed a graded increase in fluorescence intensity with rising H₂O₂ doses (Figure S6C, Supporting Information), reinforcing the reliability of the oxidative challenge model (Figure S6C, Supporting Information). Using this model, we next examined whether CLOCK contributes to oxidative stress sensitivity. Comparative analysis between wild‐type and Clock‐knockout (Clock^−/−) NIH3T3 cells revealed that Clock^−/− cells retained significantly higher viability following exposure to 600 µm H₂O₂ (Figure S6D, Supporting Information). These results suggest that loss of Clock confers a cytoprotective advantage under oxidative insult. The enhanced oxidative resilience observed in Clock^−/− cells mirrors the phenotype of clk‐deficient Drosophila, where loss of circadian transcriptional input improves survival under oxidative stress. We also investigated the regulation of Larp6 and Larp7 (mammalian orthologs of Drosophila Achl) in mammalian cells. Quantitative PCR revealed that the mRNA levels of both Larp6 and Larp7 were significantly reduced in Clock^−/− versus WT control cells (Figure S6E, Supporting Information), suggesting CLOCK regulates these genes in mammalian cells. Furthermore, to examine whether CLOCK regulates LARP6 and LARP7 transcription, we assessed BMAL1/CLOCK transactivation on their promoters using dual‐luciferase reporter assays in HEK293T cells. Co‐expressing BMAL1/CLOCK strongly increased luciferase activity for both promoters (Figure S6F, Supporting Information), indicating BMAL1/CLOCK regulates LARP6 and LARP7 in a promoter‐dependent manner.

Collectively, these findings support an evolutionarily conserved role for CLOCK/CLK in sensitizing cells to redox stress and highlight circadian transcription factors as critical modulators of oxidative homeostasis across species.

All fly strains were maintained on standard cornmeal‐agar medium at 25 °C under a 12:12 h light‐dark (LD) cycle. The following GAL4 driver lines were used for tissue‐specific gene manipulation: TG4 (tim‐GAL4), PG4 (PDF‐GAL4), TD2 (tim‐GAL4; UAS‐dicer2), PD2 (6), TD2; PDF‐GAL80, TD2; tubP‐GAL80^ts, ED2 (Elav‐GAL4;; UAS‐dicer2). UAS‐RNAi and effector lines were obtained from the Vienna Drosophila Resource Center (VDRC), Bloomington Drosophila Stock Center, and Tsinghua Fly Center.

For behavioral assays, male flies (3–5 days old) were individually loaded into glass activity tubes and monitored using the Drosophila Activity Monitoring (DAM) system (TriKinetics, USA). Flies were entrained under LD for 4 days, followed by 6 days of constant darkness (DD). Light intensity during LD cycles was maintained at ≈500 lux. Raw activity data were processed using FaasX software,^[48^] and actograms were generated and analyzed using the MATLAB.^[49^]

Adult flies (3–5 days old) were pre‐acclimated on standard food for 3 days and then transferred to food containing 20 mm paraquat (PQ, Aladdin; M106761‐1 g). Flies were maintained at 25 °C under a 12:12 h LD cycle, and survival was scored every 12 h. Flies unresponsive to gentle tapping were recorded as dead. Each condition included ≥60 flies across at least three independent replicates. In the survival experiments, the number of flies, along with their sex and genotype, is detailed in Table S1 (Supporting Information).

To isolate RNA from fly heads, whole flies were flash‐frozen in liquid nitrogen and vortexed to separate heads. Head tissues were manually sorted on a prechilled aluminum plate and transferred to 1.5 mL microcentrifuge tubes containing 100 µL TRIzol reagent (Thermo Fisher Scientific). Total RNA was extracted following the manufacturer's protocol. Homogenization was performed on ice using a handheld homogenizer (Beijing Lanjieke, L‐C119‐0974). Phase separation was achieved with 200 µL chloroform, followed by centrifugation at 12 000 rpm for 15 min at 4 °C. The aqueous phase was precipitated with an equal volume of isopropanol and incubated at −20 °C, followed by centrifugation. The RNA pellet was washed twice with 75% ethanol, air‐dried, and resuspended in DEPC‐treated water. RNA purity and concentration were measured using a NanoDrop spectrophotometer (Thermo Fisher Scientific).

Fly heads from iso³¹ and clk^out flies were collected at 0 and 24 h postparaquat treatment for RNA sequencing. Total RNA was extracted and subjected to library preparation and Illumina sequencing (BMKCloud, Qingdao, China). Sequencing reads were mapped to the Drosophila melanogaster reference genome (Release 6 + IS01 MT). Differential gene expression was analyzed using DESeq2, with significance defined as p < 0.05 and |log₂FC| > 1. Functional enrichment analyses (KEGG and GO) were performed via SangerBox^[50^] and SRplot.^[51^] Three biological replicates were included per condition.

The CDS region of the target gene was amplified from adult fly head cDNA using high‐fidelity polymerase and gene‐specific primers. PCR products were gel‐purified and cloned into linearized pAc‐FLAG or pGL3 vectors via homologous recombination (Exnase MultiS, Vazyme). Vector linearization was achieved by double digestion with restriction enzymes, followed by gel purification. Recombinant plasmids were transformed into competent E. coli and screened by colony PCR. Positive clones were verified by Sanger sequencing. Verified constructs were cultured, and plasmids were extracted using commercial kits according to the manufacturer's instructions.

Drosophila S2 cells were cultured in Schneider's Insect Medium (Sigma‐Aldrich) supplemented with 10% fetal bovine serum (FBS) at 25 °C. Cells were transfected using Cellfectin reagent (Thermo Fisher Scientific) according to the manufacturer's instructions. The plasmids used included pAc‐EGFP, pAc‐Achl‐FLAG, pAc‐clk‐V5, pGL3‐per‐promoter, pGL3‐Achl‐promoter, and pGL3‐Achl‐promoter‐mut.

For the mRNA stability assay, cells were transfected with 2.0 µg of pAc‐Achl‐FLAG, incubated for 48 h, and then treated with 10 µg mL⁻¹ actinomycin D (ATCD; Sigma‐Aldrich) for 0, 1, 2, or 4 h prior to RNA extraction and qPCR analysis.

For luciferase reporter assays, S2 cells were co‐transfected with 50 ng of pGL3‐per‐promoter or pGL3‐Achl‐promoter constructs, pAc‐Renilla (internal control), and varying amounts of pAc‐clk‐V5 or pGL3‐Achl‐promoter‐mut plasmids. Luciferase activity was measured at 48 h post‐transfection using the Dual‐Luciferase Reporter Assay System (Promega), and firefly luciferase signals were normalized to Renilla.

Following LD entrainment for three days, adult fly brains were dissected in 1× PBS on ice. Brains were incubated with 10 µm DCFH‐DA (Beyotime, S0033S) at room temperature for 1 h, followed by three PBS washes. Fluorescence imaging was performed using a laser‐scanning confocal microscope (Leica SP8 LIGHTING).

Protein lysates were prepared from dissected fly heads (20–30 per sample) or cultured S2 cells using RIPA buffer supplemented with protease and phosphatase inhibitors. Brain tissues were homogenized on ice, incubated for 30 min at 4 °C, and centrifuged at 12 000 rpm for 15 min. S2 cells were pelleted at 5000 rpm for 3 min prior to lysis. Protein samples were denatured at 99 °C for 10 min, resolved by SDS‐PAGE, and transferred to PVDF membranes (Millipore) at 200 mA for 2 h. Membranes were blocked in 5% nonfat milk in TBST for 1 h, incubated overnight at 4 °C with primary antibodies, followed by 1 h incubation with HRP‐conjugated secondary antibodies. Signals were visualized using enhanced chemiluminescence (ECL). TUBULIN (Proteintech;11224‐1‐AP); HSP90 (Proteintech;13171‐1‐AP); Relish (DSHB; AB‐1553772); N‐Relish (Raybiotech; RB‐14‐00040).

One microgram of total RNA was treated with gDNA wiper mix (Vazyme, R123‐01) at 42 °C for 2 min and reverse‐transcribed using HiScript Q RT SuperMix (Vazyme). qPCR was performed using ChamQ Universal SYBR qPCR Master Mix (Vazyme, Q711‐02) under the following cycling conditions: 95 °C for 3 min; 40 cycles of 95 °C for 10 s and 60 °C for 30 s. Relative mRNA expression was calculated using the 2^‐ΔΔCt method, with rpl32 as the internal reference. The primers used in the experiments are listed in Table S2 (Supporting Information).

To isolate nuclear and cytoplasmic proteins, ≈20 mg of fly heads were homogenized in Cytoplasmic Extraction Buffer A (Beyotime, E101‐01) containing protease and phosphatase inhibitors. After adding Buffer B, lysates were vortexed and centrifuged to collect the cytoplasmic fraction. The pellet was further washed with Buffer A and resuspended in Nuclear Extraction Buffer. After repeated vortexing and incubation on ice, samples were centrifuged to isolate nuclear proteins. Extracted proteins were mixed with SDS loading buffer, denatured at 99 °C, and stored at −80 °C until use.

S2 cells were transfected for 48 h and then lysed in RIP lysis buffer (3 mL per sample). Following centrifugation at 2500 × g for 15 min at 4 °C, the supernatant was incubated overnight with either anti‐FLAG (Sigma; F18/04) antibody or IgG (Proteintech; 10283‐1‐AP) control, followed by incubation with protein A/G magnetic beads. Beads were extensively washed with RIP buffer and PBS, then resuspended in proteinase K digestion buffer for 30 min at 55 °C. RNA was extracted using TRIzol, reverse‐transcribed, and subjected to qPCR using relish‐specific primers. The primers used in the experiments are listed in Table(Supporting Information). S2

Catalase (CAT) activity was measured in fly heads using a commercial hydrogen peroxide assay kit (Beyotime Biotechnology, Cat. No. S0051) following the manufacturer's instructions. Adult flies were collected at 0, 12, and 24 h after paraquat (PQ) feeding. Heads were separated on dry ice and homogenized in the kit lysis buffer. H₂O₂ standards were prepared from the supplied 1 m stock solution, and a standard curve was generated. For each sample, 50 µL of homogenate was loaded in triplicate wells and incubated at 25 °C for 30 min, after which absorbance was measured at 560 nm. Total protein concentration was determined by the BCA assay (Beyotime), and CAT activity was calculated according to the manufacturer's formula and expressed as units per mg protein.

To explore the structural basis of RNA‐binding protein (RBP)–mRNA interactions, we employed AlphaFold 3 for predictive modeling of the protein–RNA complex. The amino acid sequence of the target protein was retrieved from the UniProt database (https://www.uniprot.org/↗) and submitted to the AlphaFold Protein Structure Database (https://alphafold.ebi.ac.uk/↗) for structure prediction.^[52^] For mRNA modeling, the full‐length nucleotide sequence or specific binding region of the target transcript (e.g., CDS or 3′UTR fragment) was input along with the protein sequence into the AlphaFold 3 multimer framework.

The resulting predicted protein–RNA complex structure was exported in PDB format and visualized using PyMOL (The PyMOL Molecular Graphics System, Version 2.6, Schrödinger, LLC). Key contact residues and RNA‐binding interfaces were identified based on spatial proximity and surface accessibility. Interaction sites were highlighted using cartoon and stick representations. High‐resolution structural images were generated using ray tracing and exported as PNG files for subsequent figure preparation.

Rhythmicity analysis was performed using the JTK_CYCLE^[53^] algorithm in R, with the circadian period set to 24 h. Phase and amplitude estimates were derived as previously described. Survival analysis was conducted using the log‐rank (Mantel–Cox) test. Gene expression was analyzed using the ΔΔCt method, normalized to rpl32. Additional normalization strategies—using either control group values or the median across time points—were applied where appropriate, as detailed in figure legends. Western blot quantification was conducted using ImageJ (FIJI), and data were normalized to loading controls for statistical comparison. All data are presented as mean ± SEM. Statistical analyses were performed using GraphPad Prism 10.0 software. For comparisons between two groups, a two‐tailed unpaired t‐test was used. For multiple group comparisons, one‐way ANOVA was applied, followed by Tukey's post‐hoc test. A p‐value of less than 0.05 was considered statistically significant.