Introduction

Protein kinase A (PKA), also known as cyclic AMP (cAMP)-dependent protein kinase, plays a vital role in numerous cellular processes, such as regulation of metabolism1,2, signal transduction3,4 and cell death and differentiation5,6. Given its pivotal roles, dysregulation of PKA activity has been implicated in various diseases, including metabolic disorders7, neurodegenerative diseases8, cardiovascular disease9 and cancers10. Therefore, elucidating the regulatory mechanisms of PKA activity has significant implications for human health.

PKA predominantly functions in response to signals mediated by the second messenger, cAMP. It serves as an effector within the cAMP signaling pathway, particularly when activated by ligand-activated G-protein-coupled receptors (GPCRs) such as hormones and neuropeptides. The activation of GPCRs subsequently triggers adenylate cyclase (AC), which in turn converts ATP to cAMP. The cAMP then binds to the regulatory subunits of protein kinase A (PKA-R), resulting in the dissociation of the catalytic subunit (PKA-C) from the holoenzyme. Consequently, the downstream substrates of PKA are phosphorylated by PKA-C11,12,13. However, PKA has also been reported to be activated in cAMP-independent manners. Under oxidative conditions, such as in the presence of reactive oxygen species (ROS), hydrogen peroxide, and the strongly oxidizing peroxynitrite, PKA can be activated by the formation of disulfide bonds in PKA-R14,15,16. In signal transduction induced by NF-κB inducers like lipopolysaccharide and interleukin-1, IκB-α and IκB-β can bind to the N-terminal sequence of PKA, thereby inhibiting its catalytic activity17. Transforming growth factor beta (TGF-β) activates PKA by interacting with the PKA-R through the Smad3/Smad4 complex, resulting in the release of PKA-C18. Type II PKA can be specifically activated by sphingosine and its analog dimethyl sphingosine19. However, the comprehensive network of these upstream signals and their interactions remains unclear.

In our previous work, we developed a dynamic sensing system called SPARK (Separation of Phases-Based Activity Reporter of Kinase). This system enables real-time detection of kinase activity in live cells20. Leveraging the same principle, here we created a live animal sensor for monitoring PKA activity in C. elegans. C. elegans is a suitable model organism for live animal screening in a high-throughput manner due to its transparent body, short life cycle and cost-effective cultivation conditions. The signaling pathway related to PKA is highly conserved between human and C. elegans. In C. elegans, PKA consists of catalytic subunits (KIN-1) and regulatory subunits (KIN-2)21. KIN-1 has been reported to share a high degree of conservation with the catalytic subunit of murine C alpha subunit, exhibiting 82% similarity22. The research on C. elegans has contributed significantly to our understanding of the role of PKA in various biological processes, such as lipid metabolism23, oocyte maturation24, mitochondrial function25, lifespan26, immunity27, and neuronal behaviors and physiology28,29. Collectively, these features make C. elegans an ideal model for a genetic screen for PKA regulators.

The intestine of C. elegans, despite its structural simplicity, performs a range of intricate functions that are comparable to those of the mammalian digestive tract, liver, and adipose tissues. Furthermore, it contributes to pathogen defense, immunity, and the determination of longevity30. Thus, we chose to engineer the PKA sensor in the intestine of C. elegans and carried out a whole-genome RNA interference (RNAi) screen to identify regulators of intestinal PKA activity. These regulators were then classified as intestine-autonomous or intestine-non-autonomous in an intestine-specific RNAi screen. Among the regulators, we found that inhibition of mitochondrial oxidative phosphorylation (OXPHOS), either by RNAi or chemical treatment, resulted in a marked increase in intestinal PKA activity. Moreover, our findings revealed that the pyrimidine synthesis pathway plays a critical and conserved role in regulating PKA activity in both C. elegans and human cells. Together, our work provides valuable insights into the maintenance of PKA activity at the tissue level and offers potential targets for the modulation of PKA activity under physiological conditions.

Results

Establishment of a live intestinal PKA sensor in C. elegans

To create a screenable in vivo system for identifying regulators of PKA activity, we introduced the previously developed PKA-SPARK biosensor20 into the transparent nematode C. elegans. Briefly, the biosensor operates via phase separation and comprises three essential elements: (1) a canonical PKA phosphorylation motif LRRATLVD31, (2) the phosphothreonine-binding FHA1 domain31, and 3) an enhanced green fluorescent protein (EGFP) fused to the multimerizing HOTag3/6 coiled-coils (Fig. 1A and B). Intestine-specific expression of the biosensor was achieved using the ges-1 promoter32,33. The system functions through a sequence of molecular events: active PKA phosphorylates the threonine residue in LRRATLVD, allowing the recruitment of the FHA1 domain to bind the phosphorylated peptides. This kinase-dependent interaction facilitates HOTag-mediated multivalent assembly into EGFP condensates through phase separation, thereby amplifying fluorescence intensity.

Fig. 1: Establishment of the intestinal PKA sensor in C. elegans.
figure 1

A The schematic diagram displays the PKA-SPARK elements. It consists of several components, including LRRATLVD, a PKA-specific phosphorylation substrate sequence; HOTag3 and EGFP stand for homo-oligomeric coiled coils and enhanced green fluorescent protein, respectively; 2A, a self-cleaving sequence; FHA1 represents forkhead-associated domain 1; and HOTag6, another set of homo-oligomeric coiled coils. B The working model of the intestinal PKA sensor in C. elegans. C and D Representative images (left) and relative fluorescence intensity quantification (right) of the PKA sensor following RNAi treatment from L1 to adult stages (C) targeting kin-1 and kin-2 genes or a 1-hour incubation of adult worms with indicated chemicals (D), including the PKA inhibitor H89 (20 μM) and its activator FSK (50 μM). The scale bar represents 100 μm. Veh, vehicle (DMSO). N = 3 independent experiments containing at least 30 worms per group. The statistical significance values were determined by one-way ANOVA analysis followed by Dunnett’s multiple comparisons test. Error bars denote the SEM.

Full size image

We first validated the sensing capability of the wildtype PKA sensor (hereafter referred to as the PKA sensor) using genetic and chemical approaches. In C. elegans, the catalytic (KIN-1/PKA-C) and regulatory (KIN-2/PKA-R) subunits of PKA are encoded by kin-1 and kin-2, respectively. Western blot analysis confirmed their canonical roles in PKA regulation in C. elegans: kin-1 RNAi reduced PKA activity (albeit slightly), while kin-2 RNAi strongly activated it (Supplementary Fig. 1A). Consistent with these results, kin-1 RNAi significantly decreased the PKA sensor’s fluorescence intensity, whereas kin-2 RNAi caused a marked increase (Fig. 1C). The more pronounced changes observed with kin-1 RNAi in the PKA sensor suggest that the sensor is more sensitive to reduced PKA activity compared to Western blot analysis. Additionally, neither kin-1 RNAi nor kin-2 RNAi altered eGFP mRNA expression levels in the PKA sensor (Supplementary Fig. 1B), indicating that these RNAi treatments directly regulate the sensor’s PKA-SPARK element.

Effective chemical modulations using the PKA inhibitor H89 and activator forskolin (FSK) (Supplementary Fig. 1C) further validated the sensor’s specificity: H89 treatment suppressed its intensity, while FSK markedly enhanced it (Fig. 1D). To confirm phosphorylation dependency of the PKA sensor, we generated its non-phosphorylatable version by substituting the critical threonine with alanine (T → A) in the recognition motif (Supplementary Fig. 2A). Unlike the wildtype sensor, this mutant exhibited no significant fluorescence changes in response to kin-1 or kin-2 RNAi (Supplementary Fig. 2B).

Although the PKA sensor did not exhibit clear canonical puncta formation under either untreated or treated conditions (Supplementary Fig. 3A), we detected a dynamic fluorescence pulse phenomenon in the sensor worms, with approximately six peak flashes per minute (Supplementary Movies 1 and 2; Supplementary Fig. 3B–G). These rapid flashes likely reflect transient aggregation and dissociation of EGFP molecules, potentially explaining the absence of stable puncta. Notably, the peak intensities of these pulses correlated closely with PKA activity levels under genetic (kin-1 or kin-2 RNAi) and chemical (H89 or FSK) perturbations (Supplementary Fig. 3C and F), further supporting this PKA sensor as a robust readout of PKA activity. Analysis of pulse frequencies demonstrated that PKA activation via kin-2 RNAi or FSK did not affect flash rates, whereas PKA inhibition through kin-1 RNAi or H89 treatment reduced them (Supplementary Fig. 3D and G). This reduction may be attributable to suppressed PKA activity below detectable thresholds. Additionally, analysis of the PKA sensor across the C. elegans life cycle revealed that sensor intensity remained relatively stable throughout development, with only a modest adulthood increase (Supplementary Fig. 4), suggesting developmentally consistent PKA activity. Together, these results demonstrate the PKA sensor’s specificity and utility for monitoring PKA activity in live C. elegans.

High-throughput screens for intestinal PKA regulators in C. elegans

To identify regulators of intestinal PKA in C. elegans, we performed a genome-wide RNAi screen using commercially available libraries—the Ahringer and ORFeome C. elegans RNAi libraries. These libraries collectively contain 32,691 dsRNA-expressing RNAi strains, targeting ~90% of annotated C. elegans genes34,35. For efficient screening, we implemented a semi-quantitative scoring system (scale: 1–6), where 1 represented the lowest fluorescence intensity, 6 the brightest, and 3 corresponded to the empty vector control (Supplementary Fig. 5A). In the primary screen, we identified 1,093 genes whose knockdown either positively or negatively regulated PKA activity (Fig. 2A; Supplementary Data 1). To eliminate RNAi clones that might directly influence the PKA sensor’s promoter activity, we conducted parallel screening using the SJ4143 (zcIs17 [ges-1::GFP (mit)]) reporter strain, which shares the same ges-1 promoter as our PKA sensor. This validation screen identified 722 genes whose knockdown enhanced PKA activity (referred to as PKA suppressors) and 79 that suppressed it (referred to as PKA activators) (Fig. 2A and Supplementary Fig. 5B; Supplementary Data 1). Gene Ontology (GO) and KEGG analyses of the identified PKA regulators implicated multiple pathways in modulating PKA activity, including mitochondrial oxidative phosphorylation, TGF-β signaling pathway, pyrimidine metabolism, and purine metabolism (Fig. 2B and E). Notably, in addition to the canonical PKA regulators kin-1 and kin-2, our unbiased screen identified several genes and pathways previously implicated in PKA regulation (Supplementary Data 1), including Ras signaling pathways36 in the suppressor group (Fig. 2B) and components of Wnt signaling37 in the activator group (Fig. 2C and E). These results further validated the utility of our sensor system for comprehensive identification of PKA regulators. To investigate whether the identified PKA regulators act in an intestine-autonomous or non-autonomous manner, we introduced the PKA sensor into the VP303 strain. This strain carries a loss-of-function mutation in the endogenous rde-1 gene but expresses wild-type rde-1 specifically in the intestine, enabling RNAi solely in intestinal cells38. Using this integrated strain, we performed an intestine-specific RNAi screen (Fig. 2F). Out of the screen, we identified 289 genes whose RNAi induced changes in PKA sensor intensity that were equal to or stronger than those observed in the whole-genome RNAi screen; these were classified as autonomous regulators (Fig. 2F; Supplementary Data 1). Conversely, the 294 genes whose RNAi did not alter PKA activity in the intestine-specific screen were considered non-autonomous candidates. The remaining 110 genes, whose RNAi elicited weaker effects in the intestine-specific screen compared to whole-organism knockdown, were classified as dual- or multi-tissue regulators (Fig. 2F; Supplementary Data 1). Collectively, our complementary high-throughput screens provide a valuable resource of intestinal PKA regulators of C. elegans, in an intestine-autonomous and non-autonomous manner.

Fig. 2: Genetic screens for regulators of the intestinal PKA activity in C. elegans.
figure 2

A The screening design and outcomes of a genome-wide screen using the PKA sensor. B and C GO analysis of PKA suppressor (B, RNAi of which resulted in elevated intensity of the sensor) and activator (C, RNAi of which led to reduced intensity of the sensor) genes. BP, Biological Process. CC, Cellular Component. MF, Molecular Function. Pathways reported to regulate PKA are highlighted in red (suppressors) or green (activators). D and E KEGG pathway analysis of PKA suppressor (D) and activator (E) genes. Mitochondrial metabolism-related pathways are highlighted in red in the suppressor group (D), whereas pathways reported to positively regulate PKA are highlighted in green in the activator group (E). F The screening design and outcomes of the intestine-specific RNAi screen. N = 2 independent experiments containing at least 30 worms per well, and each round.

Full size image

Inhibition of mitochondrial OXPHOS triggers the activation of intestinal PKA in C. elegans

While PKA signaling is a well-established modulator of mitochondrial functions in both cytosolic and mitochondrial compartments39, whether mitochondria regulate PKA activity remains poorly understood. Intriguingly, GO and KEGG pathway analyses revealed significant enrichment of mitochondrial-associated genes among both activators and suppressors (Fig. 2B–E). Further validations demonstrated that inhibition of OXPHOS activity induced PKA activation in C. elegans (Fig. 3A–D). Such perturbations were achieved through RNAi targeting nuo-1 (mitochondrial complex I), cyc-1 (cytochrome C), and R53.4 (ATP synthase) genes (Fig. 3B), or treatment with rotenone (mitochondrial complex I inhibitor) (Fig. 3D). Importantly, Western blot analysis, independent of the PKA sensor, confirmed that OXPHOS dysfunction markedly elevated PKA activity in wildtype animals (Fig. 3C and E). These findings consistently support the crucial role of OXPHOS in the regulation of PKA activity.

Fig. 3: Disruption of mitochondrial OXPHOS leads to activation of intestinal PKA in C. elegans.
figure 3

A The schematic illustrating mitochondrial OXPHOS, including the loci of NUO-1, CYC-1, and R53.4, as well as the target of rotenone. B Representative images (left) and quantitative results (right) of measurements of the intestinal PKA activity in response to RNAi against mitochondrial OXPHOS genes. The scale bar represents 100 μm. C Western blot analysis of nuo-1, cyc-1 and R53.4 RNAi worms. D Representative images (left) and quantitative results (right) for the PKA sensor in response to rotenone treatment. Worms were treated with 4 μM Rotenone from the L4 stage for 12 h. The scale bar is 100 μm. E Western blot analysis of worms treated with rotenone.For assays using the PKA sensor, N = 3 independent experiments containing at least 30 worms per condition. The statistical significance values were determined by t-test and one-way ANOVA analysis, followed by Dunnett’s multiple comparisons test. Error bars denote the SEM.

Full size image

The pyrimidine synthesis pathway represents a novel PKA regulator in C. elegans

Notably, our screens implicated the pyrimidine synthesis pathway in PKA regulation, as they identified multiple components of the pyrimidine synthesis pathway, including the mitochondrial enzyme gene dhod-1 and its downstream genes umps-1, rnr-1, and rnr-2 (Figs. 2D, 4A and Supplementary Fig. 5B; Supplementary Data 1). In C. elegans, dhod-1 encodes dihydroorotate dehydrogenase (DHODH) that catalyzes dihydroorotate-to-orotic acid conversion in de novo pyrimidine synthesis40. To further validate these findings, we employed the specific inhibitor of DHODH, teriflunomide and found that this treatment resulted in significant activation of PKA (Fig. 4B and C). Consistently, knockdown of genes encoding proteins involved in the de novo pyrimidine synthesis, including pyr-1, dhod-1, umps-1, C29F7.6, rnr-1, and rnr-2, led to a significant increase in PKA activity (Fig. 4D). Western blot analysis, independent of the PKA sensor, also confirmed the effects of dhod-1 and umps-1 RNAi on the activation of PKA (Fig. 4E). Crucially, supplementation of the DHOD-1 product, orotic acid, rescued PKA activation in dhod-1 RNAi but not umps-1 RNAi animals (Fig. 4F), suggesting metabolite-dependent regulation of PKA activity.

Fig. 4: The pyrimidine synthesis pathway regulates the activity of intestinal PKA in a cAMP-independent manner in C. elegans.
figure 4

A A schematic diagram illustrating the de novo pyrimidine synthesis signaling pathway. B Representative images (left) and quantification results (right) of the PKA sensor worms after treatment with 100 μM teriflunomide, initiated from the L1 larval stage through adulthood. The scale bar indicates a length of 100 μm. C Western blot analysis of worms treated with Teriflunomide. D The quantification of PKA activities resulting from the knockdown of genes involved in the pyrimidine synthesis pathway. N = 3 independent experiments. The statistical significance values were determined by one-way ANOVA analysis followed by Dunnett’s multiple comparisons test. Error bars denote the SEM. E Western blot analysis of dhod-1 and umps-1 RNAi worms. F Supplementation of orotic acid suppressed the effect of dhod-1 RNAi on the activation of intestinal PKA but not the umps-1 RNAi. N = 3 independent experiments. The statistical significance values were determined by two-way ANOVA. Error bars denote the SEM. G Measurement of cAMP levels in C. elegans following treatment with indicated RNAi. N = 3 independent experiments. The statistical significance values were determined by one-way ANOVA analysis followed by Dunnett’s multiple comparisons test. Error bars denote the SEM.

Full size image

Given cAMP’s established role upstream of PKA, we tested whether pyrimidine synthesis acts through cAMP using LC–MS/MS. The assay system was validated by adding the RNAi targeting pde-4, which encodes the cAMP phosphodiesterase PDE-4 with known roles in modulating cAMP levels in C. elegans41, and the administration of the cAMP booster FSK as positive controls (Fig. 4G and Supplementary Fig. 6). The results showed that neither dhod-1 nor umps-1 RNAi altered cAMP levels (Fig. 4G), indicating cAMP-independent regulation of PKA by pyrimidine synthesis pathway.

The pyrimidine synthesis pathway plays a conserved role in modulating PKA activity in human 293T cells

To determine if pyrimidine synthesis regulates PKA activity evolutionarily conserved in humans, we used CRISPR-Cas9 to knock down DHODH in 293T cells (Supplementary Fig. 7). Western blot analysis revealed elevated phosphorylated CREB, a canonical PKA substrate, in DHODH-deficient cells, indicating increased PKA activity (Fig. 5A). Notably, orotic acid supplementation suppressed this PKA activation (Fig. 5A). Furthermore, teriflunomide treatment induced dose- and time-dependent PKA activation in 293T cells (Fig. 5B and C). Collectively, these results demonstrate an evolutionarily conserved role for the pyrimidine synthesis pathway in regulating PKA activity, from C. elegans to human cells.

Fig. 5: Inhibition of the pyrimidine synthesis pathway activates PKA in HEK293T cells.
figure 5

A Representative (left) and quantitative (right) Western blot results of DHODH knockdown cells treated with 500 μM orotic acid. B and C Western blot analysis on HEK293T cells treated with teriflunomide in a dose-dependent (B) and time-dependent (C) manner. N = 3 independent experiments, statistical significance determined by two-way ANOVA. Error bars represent SEM.

Full size image

Discussion

Multiple methods have been developed to assess protein kinase activity, including Western blot, enzyme-linked immunosorbent assay, radiometric assay, mass spectrometry42, peptide-based electrochemical sensor43, and fluorescence reporters such as fluorescence probe for cAMP44. While Förster resonance energy transfer (FRET)-based PKA sensors enable real-time monitoring in live animals, limitations include detector noise, optical interference, and photobleaching45,46,47. To overcome these constraints, we adapted the SPARK technology in the transparent C. elegans to create the PKA sensor. Using this live animal sensor, we conducted the first genome-wide genetic screen and revealed a wide range of regulators of PKA activity in an intestine-autonomous or non-autonomous manner. These regulator networks will pave the way for future studies on PKA activity under physiological or pathological conditions.

Our live imaging analysis revealed high-frequency oscillations in PKA sensor fluorescence intensity, indicating pulsatile PKA responses to upstream signals. This real-time activity capture highlights the sensor’s unique value for investigating PKA signaling dynamics, particularly signal propagation in living systems during processes like stress response48. While autofluorescence in aged animals requires further investigation and may complicate aging studies, monitoring age-related PKA dynamics could elucidate kinase–longevity relationships. Future comparisons with other established C. elegans methodologies for PKA activity testing will clarify the sensor’s advantages and limitations.

Our high-throughput screening provides the first comprehensive set of PKA regulators in C. elegans, offering new insights into the network regulation of in vivo PKA activity in both tissue-autonomous and non-autonomous contexts. Notably, our unbiased approach successfully identified canonical PKA regulators, including KIN-1 and KIN-2, as well as known modulators such as the Ras and Wnt signaling pathways36,37. However, the screen did not detect all known PKA regulators, particularly those with functional redundancy—for example, the four C. elegans adenylate cyclase isoforms49. Additionally, since the screen was performed in live animals, we likely missed RNAi candidates that cause lethality. Follow-up studies using the PKA sensor to evaluate individual genes of interest could further validate and expand these findings. Intriguingly, our findings align with reports that reduction of PKA-R induced by ROS elevates PKA catalytic activity15,50, as administration of rotenone or RNAi knockdown of mitochondrial OXPHOS genes in C. elegans consistently activated PKA. This indicates conserved OXPHOS-PKA crosstalk across species. We further identified cAMP-independent PKA regulation via pyrimidine synthesis. Pathway disruption activated intestinal PKA without altering cAMP levels, implying metabolite-driven regulation of PKA activity. Since the key enzyme DHODH localizes to mitochondria51,52, future work should explore whether mitochondrial regulation of PKA occurs through pyrimidine metabolites. Additionally, further explorations in C. elegans and other organisms on the remaining candidates uncovered in our screen would shed light on the regulation of PKA activity in vivo. In summary, we established a live-animal PKA sensor for real-time kinase monitoring and mapped spatial regulatory networks for PKA signaling in C. elegans. These resources advance understanding of the activity and roles of PKA in development, behavior, and disease. Of clinical relevance, our data suggest that teriflunomide, an FDA-approved DHODH inhibitor, may cause chronic PKA dysregulation as an off-target effect. Additionally, orotic acid supplementation could counter teriflunomide-induced PKA activation. The identification of metabolite-driven PKA regulation potentially opens avenues for drug repurposing.

Methods

Experimental materials

The experimental bacterial strains used in this study, including Escherichia coli (E. coli) strains OP50-1 (conventional nematode culture) and HT115 (RNAi nematode culture), were purchased from the Caenorhabditis Genetics Center (CGC). Forskolin (FSK, T2939, purity 99.86%), Teriflunomide (T7534, purity 99.95%), and Rotenone (T2970, purity 99.88%) were purchased from TargetMol Company. H89 (S1582, purity 99.48%) was obtained from Selleck Company, and Orotic acid (O2750, purity ≥98%) was acquired from Sigma-Aldrich Company.

C. elegans strains and cultivation

C. elegans strains were maintained at 20 °C on nematode growth medium (NGM) seeded with E. coli OP50. The PKA sensor strain WLU38 alyIs6[ges-1p::PKA-SPARK + myo-2p::RFP] served as the primary reporter. The PKA T → A mutant strain WLU453 alyEx94[myo-2p:mCherry + ges-1p::inactivated PKA-SPARK] was used to prove the reliability of the PKA sensor. To exclude nonspecific effects on the ges-1 promoter, SJ4143 zcIs17[ges-1::GFP (mit)] was used as a reference strain. All mutant strains were backcrossed five times to the wild-type N2 background prior to analysis.

Construction of transgenic C. elegans strain

The ges-1p::PKA-SPARK construct was generated by integrating a 3 kb fragment of the ges-1 promoter, fused to the PKA-SPARK sequence, into the pPD49.26 plasmid, which includes the unc-54 3’-UTR. A 1,698-bp PKA-SPARK fragment was PCR-amplified from the previously established PKA-SPARK sensor plasmid20, using primers: 5’-cag gtc gac tct aga gga tcc ATG GGC GGT GGA GGG AGC-3’ and 5’-cat ggt acc gtc gac gct agc TTA GAT TTT CTT CAG CCA CTT CTC AA-3’, which contained SphI and NheI restriction sites. The ges-1p::inactivated PKA-SPARK plasmid was generated by PCR for making a T → A mutant. The resulting construct was transformed into E. coli DH5α. Subsequently, the ges-1 promoter was inserted into the plasmid. To generate the transgenic PKA sensor strain, this construct was co-injected with the marker plasmid myo-2p::RFP into the gonads of wild-type worms using standard techniques. Finally, the extrachromosomal PKA-SPARK arrays were integrated into the C. elegans genome using UV irradiation.

RNA interference

RNAi feeding experiments were conducted on synchronized L1 larvae at a temperature of 20 °C. In these experiments, E. coli strain HT115 (DE3) expressing dsRNA was cultured for 14 h overnight in LB broth supplemented with 100 μg/mL carbenicillin at a temperature of 37 °C. The cultured bacteria were subsequently plated onto NGM plates containing 100 μg/mL carbenicillin. Most of the bacteria used for RNAi in C. elegans were derived from the Ahringer library34 and ORFeome-RNAi library35, while some RNA interference plasmids were constructed in our own laboratory.

Genome-wide RNAi screen in C. elegans

The PKA sensor strain WLU38 alyIs6[ges-1p::PKA-SPARK + myo-2p::RFP] was cultured using standard methods. For genome-wide RNAi screening, worms were maintained on IPTG-induced RNAi plates prepared in 96-well format. Three days before screening: (1) each well received 150 µL of RNAi growth medium; (2) plates were air-dried for 24 h; (3) carbenicillin-resistant RNAi bacteria (cultured in 96-well square blocks) were concentrated; (4) 30 µL of bacterial suspension was seeded per well; (5) plates were incubated 24 h for dsRNA induction prior to use. Approximately 50 synchronized L1-stage larvae were transferred to each well. Evaluators, blinded to bacterial identity except for the L4440 empty vector control, scored GFP intensity after 60 h of incubation using a Leica M205 FCA microscope. Fluorescence was assessed on a 6-point scale (1–6), with a score of 3 assigned to the control group, 6 denoting the highest fluorescence intensity, and 1 indicating the lowest fluorescence intensity of the PKA sensor.

Intestine-specific RNAi screen

The PKA sensor was introduced to the VP303 rde-1(ne219) V; kbIs7 [nhx-2p::rde-1 + rol-6(su1006)] worms53 to create the intestinal-specific RNAi PKA sensor strain WLU174 rde-1(ne219) V; kbIs7 [nhx-2p::rde-1 + rol-6(su1006)]; alyIs6[ges-1p::PKA-SPARK + myo-2p::RFP. The intestine-specific RNAi screen using this strain was conducted as in the whole-genome RNAi screen.

Chemical treatments

All drug treatments utilized synchronized C. elegans. For H89 and FSK exposure, adult worms were incubated for 1 h in 20 μM H89, 50 μM FSK, or vehicle control. In contrast, rotenone treatment began at the L4 stage (45 h post-synchronization): worms grown on standard NGM plates were transferred to plates containing 4 μM rotenone or vehicle, where they were exposed for 12 h before evaluation. Meanwhile, for teriflunomide and orotic acid treatments, compounds were supplemented directly onto NGM plates at final concentrations of 100 and 500 μM, respectively. Following the indicated treatments, animals were washed and processed for downstream assays.

cAMP extraction from C. elegans

To extract cAMP from C. elegans, approximately 5000 synchronized young adult worms were first washed with 1 mL M9 buffer in a 1.5 mL microcentrifuge tube. Subsequently, the worms underwent three additional M9 washes to remove residual bacteria, followed by one ddH2O wash to eliminate buffer salts. Immediately thereafter, samples were flash-frozen in liquid nitrogen for 30 s to halt cAMP metabolism.

Following flash-freezing, worms were sonicated in 300 μL of ice-cold extraction buffer (acetonitrile:methanol:water, 2:2:1, v/v/v) containing 25 ng/mL tenofovir (internal standard) and PDE inhibitors (EDTA and theophylline). After sonication, samples were heated at 60 °C for 10 min to denature cAMP-metabolizing enzymes, then cooled on ice for 10 min. Subsequently, samples were centrifuged at 12,000×g for 20 min at 4 °C. Upon centrifugation, supernatants were transferred to fresh tubes, while pellets underwent re-extraction with 300 μL of extraction buffer during 15-min incubation on ice. After repeating centrifugation under identical conditions, supernatants were combined with the initial extracts. Concurrently, protein pellets were solubilized in 0.1 M NaOH at 95 °C for 10 min for BCA protein quantification and normalization. Finally, combined supernatants were evaporated under nitrogen at 40 °C, and the resulting residues were redissolved in 100 μL ddH2O prior to LC–MS/MS analysis.

cAMP level determination using LC‒MS/MS

cAMP levels were quantified using liquid chromatography-tandem mass spectrometry (LC–MS/MS). LC–MS/MS analysis was conducted using an AB SCIEX QTRAP 6500+ mass spectrometer connected with an Exion LC system. All data were collected and processed using the Analyst Software. Chromatographic separation was achieved on a Waters ACQUITY UPLC BEH C18 (50 mm × 2.1 mm, 1.7 μm) column. The mobile phase consisted of 3% methanol + 20 mM ammonium acetate + 0.1% FA as solvent A and 97% methanol + 20 mM ammonium acetate + 0.1% FA as solvent B at a flow rate of 0.4 mL/min. The gradient program for mobile phase B was as follows: initially 0%, linearly increased to 50% over 5 min, then linearly decreased to 0% in 0.1 min, and finally maintained at 0% for 2.9 min. The flow rate of mobile phase was kept at 0.4 mL/min, column temperature was set at 40 °C, and the injection volume was 1 μL. Mass spectrometer operating in positive ion mode (ion spray voltage 5.5 kV) for quantification of cAMP, with the following settings: curtain gas, 35 psi, ion source temperature, 500 °C, ion source gas 1, 55 psi, ion source gas 2, 50 psi. The quantitative analysis of cAMP was measured by multiple reaction monitoring (MRM) mode. The MRM transitions used were m/z 330.2 → 136.2, 330.2 → 119.2 and 330.2 → 91.9.

Quantitative real-time PCR

Total RNA was extracted from worms or human cells using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA). Random-primed cDNA was generated by reverse transcription of the total RNA samples with SuperScript II (Invitrogen). Subsequently, qPCR analysis was performed using SYBR Premix-Ex TagTM (Takara, Dalian, China) on an Applied Biosystems Prism 7000 Sequence Detection System (Applied Biosystems, Foster City, CA, USA).

Quantification of intestinal GFP intensity

Synchronized populations of worms were cultivated on HT115(DE3) expressing either an empty vector or dsRNA at a temperature of 20 °C until they reached the young adult stage. To remove any E. coli residing on the surface of the worms, they were washed with M9 solution three times. Approximately 30 worms were then transferred to six-well slides and subjected to a 2 mg/mL levamisole solution to induce paralysis for imaging purposes. Worms were then imaged at ×2.5 and ×10 magnification using a Leica DM500 microscope with fixed exposure settings. Body area (brightfield) and GFP fluorescence were quantified using LAS X 2D Analysis software with multichannel processing. For each worm, normalized fluorescence intensity was calculated as the ratio of total GFP fluorescence intensity to its corresponding body area. Relative fluorescence intensity was then determined by dividing the normalized fluorescence intensity of experimental worms by that of the designated controls.

CRISPR-based gene knockout by dual-sgRNA system

The dual-sgRNA plasmid, gifted from Dr. Yongming Wang54, was produced using an intermediate plasmid gRNAU6 (Plasmid #135963), epiCISPRn (Plasmid #135961)—a Cas9 variant with a point mutation that enhances its efficiency compared to spCas9, and synthetic primers. The sgRNAs were designed on https://www.genscript.com/tools/gRNA-design-tool. The primers were 5’-CGC TCT TCG CCG CAA GAC CTC AGT GGA CGC CGG TTT TAG AGC TAG AAA TAG CAA-3’, 5’-CGC TCT TCT AAC TTC TGC GTA GTC CTC CGC GGC GGT GTT TCG TCC TTT CCAC-3’.

Cell transfection

A total of 1 × 106 293T cells were seeded onto a six-well plate. These cells were then cultured for 16–24 h prior to transfection. For the transfection process, lipo8000 (Beyotime, C0533-0.5 mL) was utilized. The transfection medium included culture medium containing 2 μg/mL of Puromycin Dihydrochloride (XYbio, X10020-25MG) for resistance selection. This resistance selection medium was composed of 10% FBS and PS. Moreover, it was observed that over 90% of the plasmid-transfected cells remained viable under puromycin selection for a period of two days.

Western blot analysis

For C. elegans samples, ~2000 worms were lysed in 150 μL RIPA buffer (Thermo Fisher Scientific) supplemented with protease and phosphatase inhibitor cocktails, flash-frozen in liquid nitrogen, then stored at −80 °C for 30 min. Following storage, samples were thawed on ice and sonicated using a VCX150 ultrasonicator (Sonics). Similarly, 293T cells were lysed in 150 μL ice-cold RIPA buffer. Subsequently, all lysates were centrifuged at 12,000×g for 15 min at 4 °C to remove debris, after which protein concentrations were determined by BCA assay (Thermo Fisher #23225).

Equal protein quantities were resolved on 4–12% Bis–Tris SurePAGE gels (GenScript #M00652) before electrophoretic transfer to nitrocellulose membranes. Post-transfer, membranes were blocked for 1 h at room temperature with 5% (w/v) BSA in TBST (0.1% Tween-20), then incubated overnight at 4 °C with primary antibodies diluted 1:1000: anti-α-Tubulin (Abcam ab6161), anti-β-Actin (FDbio FD0060), anti-phospho-CREB (Ser133) (CST 9198S), anti-CREB (CST 4820T), or anti-phospho-PKA substrate (CST 9624S). After three 15-min TBST washes, membranes were incubated for 1 h at room temperature with species-matched HRP-conjugated secondary antibodies diluted 1:10,000: goat anti-rat (Beyotime A0192), goat anti-rabbit (Beyotime A7016), or goat anti-mouse (Beyotime A0350). Finally, signals were developed using SuperSignal™ West Pico PLUS substrate (Thermo #34580), imaged on an Amersham Imager 680, and quantified with ImageJ software.

Statistics and reproducibility

GraphPad Prism 8.0 (GraphPad Software, Inc.) was used for statistical analyses in this study. Differences between the two groups were analyzed by an unpaired t-test. For differences among multiple groups, one-way analysis of variance (ANOVA) followed by Dunnett’s test was used. Differences between multiple groups with two variations were analyzed by two-way ANOVA followed by Sidak’s or Tukey’s multiple comparisons test. The error bars represent SEM. All the experimental results were obtained from at least three biologically replicate experiments. The sample size was provided in the figures. All schematics presented in the figures were generated using Adobe Illustrator.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.