Male Sprague-Dawley (SD) rats aged 6-7 weeks (weighing 250-280 g; total number: n=315 rats) were purchased from the experimental animal center of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology. All rats were housed in the specific pathogen-free animal facility, under controlled environmental conditions: Temperature maintained at 23±1°C, relative humidity at 50±10%, and a 12/12-h light/dark cycle. Rats were group-housed (2 per cage) in polypropylene cages with autoclaved wood chip bedding and provided with standard rodent chow and sterile water ad libitum. All experimental procedures were approved by the Experimental Animal Care and Use Committee of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology (approval no. TJH-202106615; Wuhan, China). The study was conducted in accordance with the ARRIVE guidelines to ensure transparency and reproducibility. All experiments were carried out under blinded conditions. Rats were randomly assigned to distinct groups, with syringes holding different drugs randomly coded by an independent researcher; injections were then administered strictly according to these codes. Behavioral assessments were conducted without the assessors knowing which treatment each group received.

Neuropathic pain was induced in rats using the SNI model, as previously described (15). Rats were anesthetized with isoflurane (induction, 5%; maintenance, 2-3%; cat. no. R510-22-10, RWD) and placed on a 37°C heating pad. The left hindlimb was shaved and disinfected, followed by a 1.5-2-cm incision along the thigh to expose the sciatic nerve and its three branches (common peroneal, tibial and sural). The common peroneal and tibial nerves were ligated with 6-0 silk sutures (5 mm apart), transected, and 2 mm of distal stumps were removed; the sural nerve was spared. Muscles and skin were sutured with 4-0 absorbable and 5-0 nylon sutures, respectively. After surgery, rats were allowed to recover in individual cages with free access to food and water. Sham-operated rats underwent the same exposure but without nerve ligation/transections.

Mechanical pain sensitivity was evaluated using the von Frey test (36). All tests were performed by a blinded experimenter in a quiet room. Rats were acclimatized to Plexiglas chambers on a wire mesh floor for 30 min before testing. Von Frey filaments (1.0-15.0 g) were applied vertically to the ipsilateral hindpaw plantar surface for 5 sec. A positive response (paw withdrawal, flinching, or licking) prompted using a weaker filament; no response prompted a stronger one. Using the up-down method, paw withdrawal threshold (PWT, 50% withdrawal threshold) was calculated after 6 measurements. Each paw was tested 3 times (5-min intervals), and the mean was recorded. Cold hyperalgesia was assessed using the acetone drop method, as previously described (37). Prior to testing, rats were individually housed in transparent plastic observation chambers with a wire-mesh floor and allowed a 30-min habituation period to minimize environmental stress. Following acclimatization, a 0.1-ml syringe was used to form an acetone bubble, which was gently applied to the plantar surface of the left hind paw. This procedure was repeated three times per paw, with a 5-min interval between consecutive applications to avoid sensory adaptation. During each trial, the cumulative duration of paw-related nocifensive behaviors, including licking, biting, and lifting the paw off the mesh floor, was recorded. The paw withdrawal cold duration (PWCD) for each rat was determined as the mean value derived from six independent trials.

Intrathecal catheterization was performed as previously described (38). Rats were anesthetized with isoflurane (induction, 5%; maintenance, 2-3%) and placed in a stereotaxic frame in a prone position. The atlanto-occipital membrane was exposed via a midline incision at the base of the skull. A PE-10 polyethylene catheter filled with sterile saline was inserted through the membrane into the subarachnoid space at the lumbar enlargement (L5-L6 level). The catheter was secured to the surrounding tissues with 4-0 silk sutures, and the free end was exteriorized at the back of the neck, sealed with a stainless-steel obturator, and protected by a plastic cap. Successful catheter placement was verified by injecting 10 μl of sterile saline containing 2% lidocaine. A transient (30-60 sec) bilateral hindlimb paralysis indicated correct positioning. Rats were allowed a 7-day recovery period before experimental use.

Recombinant CTRP3 (rCTRP3; cat. no. 9398-TN; Novus Biologicals, LLC) and EX-527 (a selective SIRT1 inhibitor; cat. no. HY-15452; MedChemExpress) were administered via intrathecal injection. rCTRP3 was dissolved in sterile physiological saline to prepare working solutions at concentrations corresponding to three doses: 10 μg/rat, 30 μg/rat and 90 μg/rat. Intrathecal injections were performed daily for 7 consecutive days starting from 7 day after SNI surgery. The injection volume was adjusted to 5 μl per rat, followed by 5 μl PBS to ensure complete delivery. To investigate the role of SIRT1 in CTRP3-mediated effects, the selective SIRT1 inhibitor EX-527 was used. EX-527 was dissolved in 10% dimethyl sulfoxide and administered at a fixed dose of 1.5 μg per rat. The dose of EX-527 was chosen according to a previous study (15). Intrathecal injection of EX-527 was performed 30 min prior to rCTRP3 administration in the respective experimental groups. At the end point of experiment, rats were euthanized by intraperitoneal injection of an overdose of pentobarbital sodium at a dose of 150 mg/kg body weight (39). Death was confirmed by respiratory/cardiac arrest, dilated pupils, and absent corneal reflex.

siRNAs targeting PGC-1α (PGC-1α siRNA) and ATF5 (ATF5 siRNA), along with non-targeting scrambled siRNA (negative control), were used to silence the expression of these genes in the spinal cord. All siRNAs were purchased from TsingKe Biological Technology. The sequences of siRNAs are listed in Table SI. To verify the knockdown efficiency and specificity of PGC-1α siRNA and ATF5 siRNA, independent in vitro validation was performed in PC12 cells. Cells were transfected with PGC-1α siRNA or ATF5 siRNA using Lipofectamine 3000 transfection reagent (cat. no. L3000008; Invitrogen; Thermo Fisher Scientific, Inc.), as previously described (40). The knockdown efficiency was verified by reverse transcription-quantitative PCR (RT-qPCR). For in vivo, the siRNA was dissolved in RNase-free water at 2.5 μg/μl and mixed with the transfection reagent branched polyethyleneimine (PEI; cat. no. 408727; MilliporeSigma). Intrathecal injection of siRNA in a total volume of 5 μl was initiated 5-11 days post SNI, and the knockdown efficiency of PGC-1α and ATF5 was verified by RT-qPCR (38).

Total RNA was extracted from the L4-L6 segments of the spinal cord using RNA Extraction kit (cat. no. R701-01; Vazyme Biotech Co., Ltd.) following the manufacturer's protocol (41). The complementary DNA (cDNA) was synthesized using a reverse transcription reagent kit (cat. no. R212-01; Vazyme Biotech Co., Ltd.), following the manufacturer's protocols. Amplifications were run on a Real-Time PCR System (Applied Biosystems QuantStudio 6; Thermo Fisher Scientific, Inc.). The reaction mixture (20 μl total volume) contained 10 μl of SYBR Green Supermix (cat. no. 172515; Bio-Rad Laboratories, Inc.), 0.4 μl of forward primer, 0.4 μl of reverse primer, 2 μl of cDNA template, and 7.2 μl of nuclease-free water. The thermal cycling conditions were as follows: Initial denaturation at 95°C for 30 sec, followed by 40 cycles of denaturation at 95°C for 5 sec and annealing/extension at 60°C for 30 sec. Primers for target genes and the reference gene (β-actin) were synthesized by TsingKe Biological Technology. The primer sequences are listed in Table SII. Relative mRNA expression levels were calculated using the 2^−ΔΔCq method, with β-actin serving as the internal reference to normalize target gene expression (42).

IF staining was conducted according to a previous study (43). Spinal cord segments (L4-L6) were harvested and immediately fixed in 4% paraformaldehyde (PFA) in 0.01 M phosphate-buffered saline (PBS) at 4°C for 24 h. After fixation, tissues were transferred to 30% sucrose solution (in 0.01 M PBS) for cryoprotection at 4°C, with solution changes every 24 h until tissues fully sank. Transverse frozen sections (20-μm thickness) were cut using a cryostat microtome (Leica Microsystems GmbH). To permeabilize cell membranes and block non-specific binding, sections were incubated in blocking buffer (5% bovine serum albumin (Wuhan Boster Biological Technology, Ltd.) + 0.3% Triton X-100 in 0.01 M PBS) at 25°C for 1 h. Primary antibodies were diluted in blocking buffer and applied to sections, which were then incubated overnight at 4°C in a humidified chamber. The following day, sections were washed with 0.01 M PBS and incubated with a mixture of second antibodies at 25°C for 1 h in the dark. The antibodies used in IF staining are listed in Table SIII. Images were acquired using a fluorescence microscope (Olympus Corporation). The results were expressed as the number of double-positive cells per square millimeter and quantified using ImageJ 1.8 software (National Institutes of Health), as previously described (38).

WB analysis was performed as previously described (15). Spinal cord tissues (L4-L6 segments) were homogenized in RIPA (Wuhan Boster Biological Technology, Ltd.; main components: 1% TritonX-100, 1% Sodium deoxycholate, 0.1% SDS) lysis buffer (containing 1% protease and phosphatase inhibitor cocktail) on ice, at a ratio of 1 g tissue to 10 ml buffer. After centrifugation at 12,000 × g for 15 min at 4°C, the supernatant was collected, and protein concentration was determined using a BCA protein assay kit (Thermo Fisher Scientific, Inc.). Equal amounts of protein (50 μg per sample) were separated by 10% SDS-PAGE and transferred to polyvinylidene difluoride membranes. Membranes were blocked with 5% non-fat milk in Tris-buffered saline with 0.1% Tween-20 (TBST) for 1 h at 25°C, then incubated overnight at 4°C with primary antibodies. After washing with 10% TBST, the membranes were incubated with second antibodies for 1 h at 25°C. The antibodies used in WB are listed in Table SIV. Protein bands were visualized using an enhanced chemiluminescence (ECL) detection kit (Millipore) and imaged with a ChemiDoc XRS+ system (Bio-Rad Laboratories, Inc.). Band intensities were quantified using ImageLab software 5.2 (Bio-Rad Laboratories, Inc.), with target protein expression normalized to β-actin.

The mtDNA copy number quantification was performed as previously described (15). Total DNA was extracted from L4-L6 spinal cord segments using a Tissue Genomic DNA Extraction Kit [Elk (Wuhan) Biotechnology Co., Ltd.] following the manufacturer's protocol, and mtDNA copy number was determined by RT-qPCR using primers specific for mitochondrial genes and a nuclear reference gene. Relative mtDNA copy number was assessed by the mitochondrial encoded NADH dehydrogenase 1 (ND1) and was normalized to nuclear-encoded β-actin. The primer sequences are listed in Table SII.

Superoxide anions were detected using DHE staining, following the previously described method (36,44). Frozen sections (20-μm thickness) of the L4-L6 spinal cord segments were prepared. These sections were incubated with 1.5 μmol/ml DHE (cat. no. HY-D0079; MedChemExpress) at 25°C for 50 min in a dark environment. Subsequently, the sections were stained with 4',6-diamidino-2-phenylindole (DAPI) at a concentration of 0.001 mg/ml at 25°C for 10 min in the dark. Images were acquired using a fluorescence microscope (Olympus Corporation). For DHE staining analysis, the number of DHE-positive cells per square millimeter was counted, and the data were processed using ImageJ 1.8 software (National Institutes of Health).

Oxidative stress parameters and antioxidant indices were measured using commercial kits, following procedures similar to those previously described (36). Tissue samples were homogenized and centrifuged at 16,000 × g for 30 min at 4°C, with the supernatants collected for subsequent biochemical assays. The resulting supernatant was collected for subsequent analyses, with protein concentrations determined via BCA assay to normalize the results. Levels of oxidative damage markers, including malondialdehyde (MDA) and protein carbonyl (PCO), were quantified using the corresponding assay kits respectively, in strict accordance with the manufacturers' protocols. For antioxidant indices, glutathione (GSH) levels, superoxide dismutase (SOD) activity and glutathione peroxidase (GSH-PX) activity were assessed using the corresponding assay kits respectively, following the manufacturer's instructions. The commercial assay kits used in biochemical assays are listed in Table SV.

Mitochondria were isolated from rat spinal cord tissues using a mitochondrial isolation kit (cat. no. ab110168; Abcam) according to the manufacturer's instructions. Briefly, tissue samples were homogenized in ice-cold isolation buffer, and the homogenate was centrifuged at low speed to remove debris. The supernatant was then centrifuged at high speed to obtain the mitochondrial pellet, which was washed and resuspended in assay buffer. For MMP detection, the isolated mitochondria were incubated with 5 μM JC-1 dye (cat. no. C2003S; Beyotime Institute of Biotechnology) at 37°C for 30 min in the dark. Fluorescence intensities were measured by flow cytometry (BD FACSCanto II; BD Biosciences), and MMP was calculated as the ratio of red fluorescence (aggregated JC-1) to green fluorescence (monomeric JC-1) using FlowJo software (version 10.8.1; Tree Star, Inc.).

ATP levels were measured using an ATP Assay Kit (cat. no. S0026; Beyotime Institute of Biotechnology), as previously described (45). Briefly, spinal cord tissues were homogenized in lysis buffer, and the homogenate was centrifuged at 12,000 × g for 5 min at 4°C. The supernatant was collected and mixed with ATP detection working solution, and the luminescence intensity was determined using a microplate reader. The ATP concentration was calculated according to a standard curve. Protein concentration was measured using a BCA Protein Assay Kit, and ATP levels were normalized to total protein content and expressed as nmol/mg protein.

All data are presented as the mean ± standard error of the mean (SEM). Statistical analyses were performed using GraphPad Prism 9.0 software (Dotmatics). Normality distribution was evaluated using the Shapiro-Wilk test. For comparisons between two groups, unpaired Student's t-test was used. For multiple group comparisons, one-way analysis of variance (ANOVA) was applied, followed by Bonferroni's post hoc test for multiple comparison correction. Behavioral data were analyzed using two-way repeated-measures ANOVA, with group (row factor) and time (column factor) as the main factors, and interaction between group and time was also tested. P<0.05 was considered to indicate a statistically significant difference. Effect sizes were reported as eta squared (η², representing partial eta squared) for all main effects, interactions, and two-sample t-tests. Relevant statistical results have been incorporated into the following section.

A neuropathic pain model was established using the SNI procedure. As depicted in Fig. 1A [Interaction: F (3, 72)=17.45, P<0.0001, η²=0.4211; Time factor: F (3, 72)=20.39, P<0.0001, η²=0.4593; Group factor: F (1, 72)=207.9, P<0.0001, η²=0.7428] and Fig. 1B [Interaction: F (3, 72)=24.09, P<0.0001, η²=0.5010; Time factor: F (3, 72)=25.93, P<0.0001, η²=0.5193; Group factor: F (1, 72)=194.6, P<0.0001, η²=0.7299], behavioral pain testing results indicated that, relative to sham-operated controls, rats receiving SNI displayed a statistically significant decrease in PWT and a concurrent increase in PWCD over the time window of post-operative day 3 to day 14, confirming the successful induction of neuropathic pain hypersensitivity. To assess changes in CTRP3 expression, RT-qPCR and WB analyses were performed on spinal cord tissues. The results revealed that SNI led to a downregulation of both CTRP3 mRNA [Fig. 1C: F (3, 16)=47.17, P<0.0001, η²=0.8984] and protein [Fig. 1D: F (3, 16)=51.68, P<0.0001, η²=0.9065] levels in the spinal cord compared with the sham-operated group. Furthermore, IF staining was employed to determine the cellular localization of CTRP3 in the spinal dorsal horn. As shown in Fig. 1E-I [Fig. 1G: t (8)=14.17, P<0.0001, η²=0.9617; Fig. 1H: t (8)=2.194, P=0.0595, η²=0.3757; Fig. 1I: t (8)=1.965, P=0.0850, η²=0.3255], CTRP3 was predominantly localized in neurons (identified by NeuN) within the spinal dorsal horn. Notably, the number of CTRP3-positive neurons was significantly reduced post-SNI relative to the sham group. These findings suggest that neuronal CTRP3 in the spinal cord may play a critical role in the development of SNI-induced neuropathic pain in male rats.

To further investigate the role of CTRP3 in neuropathic pain, intrathecal injections of rCTRP3 at doses of 10, 30, or 90 μg were administered from day 7 to day 11 post-nerve injury. As shown in Fig. 2A [Interaction: F (20, 270)=3.781, P<0.0001, η²=0.2187; Time factor: F (5, 270)=45.32, P<0.0001, η²=0.4562; Group factor: F (4, 270)=115.9, P<0.0001, η²=0.6319] and Fig. 2B [Interaction: F (20, 270)=4.618, P<0.0001, η²=0.2548; Time factor: F (5, 270)=41.57, P<0.0001; η²=0.4350; Group factor: F (4, 270)=124.8, P<0.0001, η²=0.6489], repeated intrathecal administration of rCTRP3 at 30 and 90 μg significantly restored the reduced PWT and abrogated the elevated PWCD relative to the SNI + vehicle group. These results demonstrated that repeated rCTRP3 administration could mitigate established mechanical allodynia and cold hyperalgesia in SNI rats. Additionally, IF staining and WB analyses were used to assess spinal SIRT1 expression. IF staining results revealed that the number of SIRT1-positive neurons was decreased after SNI compared with the sham group, while treatment with rCTRP3 (30 and 90 μg) reversed this reduction in SIRT1 expression in spinal dorsal horn neurons of SNI rats [Fig. 2C and D: F (4, 20)=13.83, P<0.0001, η²=0.7344]. Consistently, WB results showed that rCTRP3 administration (30 and 90 μg) blocked the nerve injury-induced downregulation of SIRT1 in the spinal cord, relative to the SNI + vehicle group [Fig. 2E: F (4, 20)=10.86, P<0.0001, η²=0.6847]. These findings confirm that CTRP3 alleviates SNI-induced mechanical allodynia and cold hyperalgesia and suggest that neuronal SIRT1 in the spinal cord may mediate this analgesic effect in SNI rats.

Mitochondrial biogenesis is a key process for sustaining mitochondrial homeostasis. It was hypothesized that the analgesic effect of rCTRP3 may involve the regulation of mitochondrial biogenesis. To test this, targets associated with mitochondrial biogenesis were examined, including PGC-1α (a critical inducer of mitochondrial biogenesis), NRF1, TFAM and mtDNA copy number. As demonstrated in Fig. 3A and B [F (4, 20)=12.33, P<0.0001, η²=0.7115], IF staining revealed that the number of PGC-1α-positive neurons in the spinal dorsal horn was reduced after SNI compared with the sham group, whereas treatment with rCTRP3 (30 and 90 μg) reversed this decrease in PGC-1α expression in SNI rats. Consistently, WB analysis demonstrated that rCTRP3 administration (30 and 90 μg) blocked the nerve injury-induced downregulation of PGC-1α [Fig. 3C: F (4, 20)=16.94, P<0.0001, η²=0.7721], NRF1 [Fig. 3D: F (4, 20)=14.79, P<0.0001, η²=0.7473] and TFAM [Fig. 3E: F (4, 20)=12.68, P<0.0001, η²=0.7172] in the spinal cord, relative to the SNI + vehicle group. Additionally, mtDNA copy number was measured to assess mitochondrial biogenesis activity. As revealed in Fig. 3F [F (4, 20)=18.68, P<0.0001, η²=0.7888], rCTRP3 treatment (30 and 90 μg) reversed the reduction in spinal mtDNA copy number observed in SNI rats. These findings confirm that CTRP3 enhances PGC-1α-dependent mitochondrial biogenesis in the spinal cord of SNI-induced rats.

The UPR^mt is a vital mechanism for maintaining mitochondrial protein homeostasis and redox balance. It was hypothesized that CTRP3 may exert its analgesic effects partly through the regulation of UPR^mt. To explore this, UPR^mt-related targets were analyzed, including ATF5 (a key inducer of UPR^mt), chaperone proteins (Hsp60) and proteases (ClpP, LonP1). As shown in Fig. 4A and B [F (4, 20)=10.88, P<0.0001, η²=0.6851], IF staining indicated that the number of ATF5-positive neurons in the spinal dorsal horn was decreased after SNI compared with the sham group, while rCTRP3 treatment (30 and 90 μg) reversed this reduction in ATF5 expression in SNI rats. Consistently, RT-qPCR results demonstrated that rCTRP3 administration (30 and 90 μg) blocked the nerve injury-induced downregulation of ATF5 [Fig. 4C: F (4, 20)=19.39, P<0.0001, η²=0.7950], ClpP [Fig. 4D: F (4, 20)=15.12, P<0.0001, η²=0.7515], Hsp60 [Fig. 4E: F (4, 20)=13.06, P<0.0001, η²=0.7231] and LonP1 [Fig. 4F: F (4, 20)=11.67, P<0.0001, η²=0.7001] in the spinal cord, relative to the SNI + vehicle group. These results confirm that CTRP3 promotes ATF5-mediated UPR^mt in the spinal cord of SNI-induced rats.

To further verify the protective effects of rCTRP3 on mitochondrial function, MMP and intracellular ATP production were directly measured. As illustrated in Fig. S1A-C, SNI induced a marked reduction in MMP [F (4, 20)=31.33, P<0.0001, η²=0.8624] and ATP levels [F (4, 20)=14.26, P<0.0001, η²=0.7404] in the spinal cord, both of which were significantly restored by intrathecal administration of rCTRP3 at 30 and 90 μg. These data provide direct functional evidence that rCTRP3 preserves mitochondrial bioenergetic capacity in the setting of neuropathic pain. Redox balance in the spinal cord was evaluated using DHE staining and biochemical assays. As revealed in Fig. 5A and B [F (4, 20)=69.93, P<0.0001, η²=0.9333], DHE staining results revealed that administration of rCTRP3 (30 and 90 μg) reduced the number of DHE-positive cells in the spinal cord of SNI rats, which was elevated in the SNI + vehicle group. Consistent with this, biochemical assay data demonstrated that rCTRP3 treatment (30 and 90 μg) reversed the SNI-induced increases in oxidative damage markers, including MDA [Fig. 5C: F (4, 20)=16.82, P<0.0001, η²=0.7709] and PCO [Fig. 5D: F (4, 20)=16.88, P<0.0001, η²=0.7715] in the spinal cord. Additionally, rCTRP3 intervention (30 and 90 μg) significantly increased the level of reduced GSH [Fig. 5E: F (4, 20)=11.99, P<0.0001, η²=0.7056] and enhanced the activity of antioxidant enzymes, including GSH-PX [Fig. 5F: F (4, 20)=13.09, P<0.0001, η²=0.7236] and SOD [Fig. 5G: F (4, 20)=10.52, P<0.0001, η²=0.6779] in the spinal cord of SNI rats. These findings confirm that CTRP3 can alleviate SNI-induced oxidative stress in the spinal cord of rats.

To investigate whether SIRT1 mediates the analgesic effect of rCTRP3 in SNI rats, the SIRT1 antagonist EX-527 was employed. Results demonstrated that rCTRP3 treatment restored the reduced PWT [Fig. 6A: Interaction: F (15, 216)=6.652, P<0.0001, η²=0.3160; Time factor: F (5, 216)=44.13, P<0.0001, η²=0.5053; Group factor: F (3, 216)=148.5, P<0.0001, η²=0.6734] and attenuated the elevated PWCD [Fig. 6B: Interaction: F (15, 216)=7.944, P<0.0001, η²=0.3556; Time factor: F (5, 216)=51.85, P<0.0001, η²=0.5455; Group factor: F (3, 216)=175.3, P<0.0001, η²=0.7088] in SNI rats; however, co-administration of EX-527 abolished this analgesic effect of rCTRP3. These data indicated that CTRP3 alleviates mechanical allodynia and cold hyperalgesia by activating spinal SIRT1 in SNI rats. Furthermore, to explore the mechanism by which CTRP3 regulates mitochondrial biogenesis, EX-527 was used in subsequent experiments. IF staining revealed that EX-527 blocked the rCTRP3-induced increase in PGC-1α expression in spinal dorsal horn neurons of SNI rats [Fig. 6C and D: F (3, 16)=16.84, P<0.0001, η²=0.7594]. Consistently, WB analysis identified that EX-527 abolished the rCTRP3-mediated upregulation of PGC-1α [Fig. 6E: F (3, 16)=27.96, P<0.0001, η²=0.8398], NRF1 [Fig. 6F: F (3, 16)=17.25, P<0.0001, η²=0.7639] and TFAM [Fig. 6G: F (3, 16)=9.562, P=0.0007, η²=0.6419] in the spinal cord, compared with the SNI + rCTRP3 group. Additionally, rCTRP3 treatment reversed the reduction in mtDNA copy number in SNI rats, but this effect was blocked by EX-527 [Fig. 6H: F (3, 16)=44.02, P<0.0001, η²=0.8919]. These results confirm that CTRP3 alleviates mechanical allodynia and cold hyperalgesia and promotes PGC-1α-mediated mitochondrial biogenesis in the spinal cord of SNI rats through SIRT1 activation.

To investigate the mechanism by which CTRP3 regulates the UPR^mt, the SIRT1 antagonist EX-527 was utilized in the present study. IF staining results revealed that EX-527 inhibited the rCTRP3-induced increase in ATF5 expression in spinal dorsal horn neurons of SNI rats [Fig. 7A and B: F (3, 16)=15.36, P<0.0001, η²=0.7423]. Consistently, RT-qPCR analysis demonstrated that EX-527 abolished the rCTRP3-mediated upregulation of ATF5 [Fig. 7C: F (3, 16)=24.98, P<0.0001, η²=0.8240], ClpP [Fig. 7D: F (3, 16)=14.60, P<0.0001; η²=0.7324], Hsp60 [Fig. 7E: F (3, 16)=17.52, P=0.0007, η²=0.7666] and LonP1 [Fig. 7F: F (3, 16)=15.18, P<0.0001, η²=0.7400] in the spinal cord, relative to the SNI + rCTRP3 group. These findings demonstrate that CTRP3 promotes ATF5-mediated UPR^mt in the spinal cord of SNI rats through SIRT1 activation.

Moreover, the beneficial effects of rCTRP3 on restoring MMP [F (3, 16)=20.21, P<0.0001, η²=0.7912] and ATP production [F (3, 16)=12.61, P=0.0002, η²=0.7028] were completely abolished by pretreatment with the SIRT1 inhibitor EX-527 (Fig. S2A-C), indicating that SIRT1 activity is essential for rCTRP3 to maintain mitochondrial membrane integrity and energy production. As demonstrated in Fig. 8A and B [F (3, 16)=58.50, P<0.0001, η²=0.9164], DHE staining revealed that compared with the SNI + rCTRP3 group, treatment with EX-527 reversed the rCTRP3-induced reduction in the number of DHE-positive cells in the spinal cord of SNI rats. Furthermore, biochemical assay results demonstrated that EX-527 abolished the rCTRP3-mediated downregulation of oxidative damage markers, including MDA [Fig. 8C: F (3, 16)=18.93, P<0.0001, η²=0.7802] and PCO [Fig. 8D: F (3, 16)=17.45, P<0.0001, η²=0.7659] in the spinal cord of SNI rats. Notably, while rCTRP3 increased the level of GSH [Fig. 8E: F (3, 16)=9.383, P=0.0008, η²=0.6376] and enhanced the activity of antioxidant enzymes such as GSH-PX [Fig. 8F: F (3, 16)=12.27, P=0.0002, η²=0.6970] and SOD [Fig. 8G: F (3, 16)=12.34, P=0.0002, η²=0.6983] in the spinal cord of SNI rats, these beneficial effects were reversed by the SIRT1 antagonist. These findings confirm that CTRP3 mitigates spinal oxidative stress in SNI rats through SIRT1 activation.

To establish whether PGC-1α serves as a critical downstream mediator of rCTRP3-induced analgesia in SNI rats, PGC-1α expression in the spinal cord was specifically silenced using siRNA. RT-qPCR analysis confirmed that PGC-1α siRNA transfection induced significant downregulation of PGC-1α mRNA in PC12 cells in vitro without additional stimulation [Fig. 9A: t (8)=6.717, P=0.0002, η²=0.8494]. Consistently, intrathecal injection of PGC-1α siRNA also significantly inhibited the endogenous transcription of spinal PGC-1α in SNI rats compared with the negative control siRNA group [Fig. 9B: t (8)=6.654, P=0.0002, η²=0.8470]. Subsequent behavioral testing demonstrated that intrathecal administration of rCTRP3 significantly elevated the reduced PWT and reduced the prolonged PWCD in SNI-operated rats; however, these analgesic effects were completely abrogated by co-delivery of PGC-1α siRNA [Fig. 9B and C; Fig. 9B: Interaction: F (20, 270)=6.262, P<0.0001, η²=0.3169; Time factor: F (5, 270)=62.06, P<0.0001, η²=0.5347; Group factor: F (4, 270)=143.9, P<0.0001, η²=0.6807; Fig. 9C: Interaction: F (20, 270)=6.418, P<0.0001, η²=0.3222; Time factor: F (5, 270)=80.32, P<0.0001, η²=0.5980; Group factor: F (4, 270)=168.2, P<0.0001, η²=0.7137]. These results indicate that PGC-1α activation is essential for rCTRP3 to alleviate mechanical allodynia and cold hyperalgesia in the context of neuropathic pain.

To further clarify the regulatory role of PGC-1α in mitochondrial biogenesis triggered by rCTRP3, IF staining and WB analysis were performed. IF staining revealed that PGC-1α siRNA significantly blocked the rCTRP3-mediated upregulation of PGC-1α in neurons of the spinal dorsal horn [Fig. 10A and B: F (4, 20)=14.12, P<0.0001, η²=0.7385]. In line with this observation, WB analysis demonstrated that PGC-1α silencing abolished the rCTRP3-induced increases in PGC-1α [F (4, 20)=7.979, P=0.0005, η²=0.6148], NRF1 [F (4, 20)=38.50, P<0.0001, η²=0.8850] and TFAM [F (4, 20)=19.60, P<0.0001, η²=0.7967] protein levels in the spinal cord, as compared with the SNI + rCTRP3 group (Fig. 10C-E). Moreover, rCTRP3 treatment effectively reversed the SNI-induced reduction in mtDNA copy number, and this beneficial effect was also negated by PGC-1α siRNA intervention [Fig. 10F: F (4, 20)=16.59, P<0.0001, η²=0.7684].

Notably, it was further explored whether PGC-1α could modulate ATF5-mediated UPR^mt during rCTRP3 treatment. IF staining showed that PGC-1α siRNA significantly inhibited the rCTRP3-evoked upregulation of ATF5 in spinal dorsal horn neurons [Fig. S3A and B: F (4, 20)=15.48, P<0.0001, η²=0.7558]. Consistently, RT-qPCR analysis revealed that silencing PGC-1α completely reversed the rCTRP3-mediated enhancement of ATF5 [F (4, 20)=13.22, P<0.0001, η²=0.7256], ClpP [F (4, 20)=14.29, P<0.0001, η²=0.7408], Hsp60 [F (4, 20)=13.33, P<0.0001, η²=0.7273] and LonP1 [F (4, 20)=14.17, P<0.0001, η²=0.7391] mRNA expression in the spinal cord (Fig. S3C-F). Collectively, these findings demonstrate that PGC-1α functions as a key molecular hub that enables rCTRP3 to simultaneously promote mitochondrial biogenesis and ATF5-dependent UPR^mt, ultimately mitigating neuropathic pain hypersensitivity.

Similarly, silencing PGC-1α via siRNA also blocked the rCTRP3-induced restoration of MMP [F (4, 20)=24.04, P<0.0001, η²=0.8278] and ATP content [F (4, 20)=15.84, P<0.0001, η²=0.7601] in the spinal cord of SNI rats (Fig. S4A-C), supporting that PGC-1α-dependent mitochondrial biogenesis is required for rCTRP3 to sustain normal mitochondrial function. As depicted in Fig. 11A and B [F (4, 20)=52.53, P<0.0001, η²=0.9131], DHE staining analysis showed that compared with the SNI + rCTRP3 group, administration of PGC-1α siRNA counteracted the rCTRP3-induced decrease in the number of DHE-positive cells in the spinal cord of SNI rats. In addition, biochemical assay data revealed that PGC-1α siRNA eliminated the rCTRP3-driven downregulation of oxidative damage indicators, specifically MDA [Fig. 11C: F (4, 20)=18.60, P<0.0001, η²=0.7881] and PCO [Fig. 11D: F (4, 20)=18.20, P<0.0001, η²=0.7845] in the spinal cord of SNI rats. Of note, although rCTRP3 elevated the level of GSH [Fig. 11E: F (4, 20)=14.86, P<0.0001, η²=0.7483] and improved the activity of antioxidant enzymes including GSH-PX [Fig. 11F: F (4, 20)=11.54, P<0.0001, η²=0.6977] and SOD [Fig. 11G: F (4, 20)=22.86, P<0.0001, η²=0.8205] in the spinal cord of SNI rats, these favorable changes were reversed by PGC-1α siRNA intervention. These results indicate that CTRP3 alleviates spinal oxidative stress in SNI rats through the activation of PGC-1α.

To define the functional contribution of ATF5 to rCTRP3-evoked analgesic effects, siRNA was used to specifically knock down ATF5 expression in the spinal cord of SNI rats. RT-qPCR analysis confirmed that ATF5 siRNA transfection induced significant downregulation of ATF5 mRNA in PC12 cells in vitro without additional stimulation [Fig. 12A, t (8)=9.359, P<0.0001, η²=0.9163]. Consistently, intrathecal injection of ATF5 siRNA also significantly inhibited the endogenous transcription of spinal ATF5 in SNI rats compared with the negative control siRNA group [Fig. 12B, t (8)=6.954, P=0.0001, η²=0.8581]. Behavioral assessments revealed that rCTRP3 treatment significantly restored PWT and shortened PWCD in SNI rats, whereas combined administration of ATF5 siRNA fully abolished these pain-relieving effects [Fig. 12B and C; Fig. 12B: Interaction: F (20, 270)=4.874, P<0.0001, η²=0.2652; Time factor: F (5, 270)=49.97, P<0.0001, η² =0.4807; Group factor: F (4, 270)=121.0, P<0.0001, η²=0.6419; Fig. 12C: Interaction: F (20, 270)=7.286, P<0.0001, η²=0.3505; Time factor: F (5, 270)=91.13, P<0.0001, η²=0.6279; Group factor: F (4, 270)=196.1, P<0.0001, η²=0.7439]. These data confirm that ATF5 activation is required for rCTRP3 to exert its anti-nociceptive actions in neuropathic pain.

The impact of ATF5 silencing on rCTRP3-mediated regulation of UPR^mt was next examined. IF staining showed that ATF5 siRNA significantly suppressed the rCTRP3-induced elevation of ATF5 in neurons of the spinal dorsal horn [Fig. 13A and B: F (4, 20)=11.67, P<0.0001, η²=0.7001]. Correspondingly, RT-qPCR analysis demonstrated that ATF5 knockdown abolished the rCTRP3-mediated upregulation of ATF5 [F (4, 20)=14.71, P<0.0001, η²=0.7464], ClpP [F (4, 20)=14.10, P<0.0001, η²=0.7383], Hsp60 [F (4, 20)=8.704, P=0.0003, η²=0.6351] and LonP1 [F (4, 20)=18.67, P<0.0001, η²=0.7888] transcripts in the spinal cord, relative to the SNI + rCTRP3 group (Fig. 13C-F). These results confirm that ATF5 is indispensable for rCTRP3 to trigger the UPR^mt in the spinal cord after nerve injury.

To test whether ATF5 also modulates mitochondrial biogenesis downstream of rCTRP3, key molecules involved in this process were assessed. IF staining revealed that ATF5 siRNA significantly inhibited the rCTRP3-mediated increase in PGC-1α expression in spinal dorsal horn neurons [Fig. S5A and B: F (4, 20)=14.20, P<0.0001, η²=0.7395]. In agreement, WB analysis showed that ATF5 silencing significantly reversed the rCTRP3-induced upregulation of PGC-1α [F (4, 20)=15.32, P<0.0001, η²=0.7539], NRF1 [F (4, 20)=11.57, P<0.0001, η²=0.6982] and TFAM [F (4, 20)=15.69, P<0.0001, η²=0.7583] proteins in the spinal cord (Fig. S5C-E). Taken together, these results demonstrate that ATF5 acts as a critical downstream effector that supports rCTRP3-induced UPR^mt and cooperatively enhances PGC-1α-dependent mitochondrial biogenesis, thereby contributing to the alleviation of neuropathic pain.

In parallel, ATF5 siRNA eliminated the rCTRP3-induced restoration of MMP [F (4, 20)=12.34, P<0.0001, η²=0.7116] and ATP production [F (4, 20)=16.38, P<0.0001, η²=0.7661] (Fig. S6A-C), demonstrating that ATF5-mediated UPR^mt also contributes to the improved mitochondrial bioenergetics induced by rCTRP3. DHE staining analysis indicated that compared with the SNI + rCTRP3 group, administration of ATF5 siRNA reversed the rCTRP3-induced decrease in the number of DHE-positive cells in the spinal cord of SNI rats [Fig. 14A and B: F (4, 20)=65.07, P<0.0001, η²=0.9286]. Moreover, biochemical assay results revealed that ATF5 siRNA eliminated the rCTRP3-mediated downregulation of oxidative damage markers, including MDA [Fig. 14C: F (4, 20)=17.76, P<0.0001, η²=0.7803] and PCO [Fig. 14D: F (4, 20)=9.924, P=0.0001, η²=0.6650] in the spinal cord of SNI rats. Notably, while rCTRP3 increased the level of GSH [Fig. 14E: F (4, 20)=10.71, P<0.0001, η²=0.6818] and enhanced the activity of antioxidant enzymes such as GSH-PX [Fig. 14F: F (4, 20)=13.68, P<0.0001, η²=0.7323] and SOD [Fig. 14G: F (4, 20)=18.68, P<0.0001, η²=0.7888] in the spinal cord of SNI rats, these beneficial changes were counteracted by ATF5 siRNA intervention. These results indicate that CTRP3 mitigates spinal oxidative stress in SNI rats through ATF5 activation.