Adult male Sprague-Dawley rats weighing 85±3 g were purchased from the Beijing Laboratory Animal Research Center (Beijing, China). They were housed in a room maintained at 23–25°C on a 12-h-light/dark cycle, and were provided with rat chow and water ad libitum. The study was approved by the ethics committee of Shandong University (Shandong, China), and all animal experiments were performed in accordance with the guidelines of the Institutional Animal Care and Use Committee of Shandong University.

Following 1 week of adaptation to their housing environment, the rats were randomly divided into the following groups (n=18 in each group): i) High-fat diet (HFD) control group, comprising HFD-fed rats without treatment; ii) HFD+UA (Sigma-Aldrich, St. Louis, MO, USA), comprising HFD-fed rats treated with UA; iii) HFD+UA+PPARα small interfering (si)RNA, comprising HFD-fed rats treated with UA and PPARα siRNA; iv) HFD+UA+scrambled siRNA, comprising HFD-fed rats treated with UA and scrambled siRNA; v) low-fat diet (LFD), comprising LFD-fed rats without treatment.

The rats, which were assigned to the HFD or LFD groups were fed a HFD (45% kcal as fat; Research Diets, New Brunswick, NJ, USA) or a LFD (10% kcal as fat) for 12 weeks, respectively. UA (250 mg/kg/day) was orally administered for 8 weeks, starting from week 4, when the HFD-fed rats had become significantly heavier, compared with the LFD-fed rats (12). The dose of UA was based on a previous study (14). PPARα siRNA (2 μg/10 μl) or scrambled siRNA was intrathecally injected at 10 weeks, as previously described (12). Body weights were measured weekly. At the end of the experiments, 12 weeks post-HFD or LFD feeding), six rats from each group were sacrificed by decapitation using a guillotine (Kent Scientific Corporation, Torrington, CT, USA) and 5-ml blood samples were collected for the measurement of metabolic parameters, and the spinal cord was removed for assessment of the expression levels of PPARα and inflammatory mediators, and the activity of nuclear factor (NF)-κB at baseline. Other rats (n=12 rats in each group) received a subcutaneous injection of 3% carrageenan (50 μl; Sigma-Aldrich) into the mid-plantar region of the right hindpaw, as previously described (12). The thermal and mechanical responses of the right hindpaw, and the paw volumes of the right and left hindpaws were then measured at baseline, and at 2, 4, 6, 8 and 24 h following carrageenan injection. At 6 h post-carrageenan injection, six rats from each group were sacrificed by decapitation using a guillotine to assess the expression levels of PPARα and inflammatory mediators, and the activity of NF-κB in the spinal cord in response to carrageenan injection. The remaining rats were sacrificed 24 h after carrageenan injection by decapitation using a guillotine.

PPARα siRNA (Shanghai GenePharma Co., Ltd., Shanghai, China) was constructed using IMG-800 vector (pSuppressorNeo; Imgenex, San Diego, CA, USA), as previously described (12). The siRNA target sequence used for the construction of the PPARα plasmid vector was: 5′-TCGAGTGTGATCGAAGCTGCAAGATGGAATTCGATCTTGCAGCTTCGATCACACTTTTT-3′. The PPARα siRNA or scrambled siRNA (Shanghai GenePharma) was suspended in serum-free media and intrathecally injected in a volume of 10 μl using a 25-μl Hamilton microsyringe (Hamilton Co., Reno, NV, USA) fitted with a 27-gauge needle. The needle was inserted into the subarachnoid space through the intervertebral foramen. The correct placement of the needle was verified by a tail or foot flick response.

Prior to the assessment of response to thermal and mechanical stimulation, the animals underwent acclimatization to the assessment environment for 1 week by placing them in the thermal and mechanical assessment boxes for 30–60 min twice daily. The hindpaw withdrawal responses to thermal stimulation were assessed using a plantar test instrument (#7370; Ugo Basile; Comerio, VA, Italy), as previously described (12). The thermal stimulation was performed using an infrared beam, which was directed onto the plantar surface of the hindpaw, and the latency to paw withdrawal was recorded. At each time point (0, 2, 4, 6, 8 and 24 h after the carrageenan injection), three measurements were performed and the average value was calculated. A cut-off value of 25 sec was applied in the absence of a response to prevent tissue damage. The hindpaw withdrawal responses to mechanical stimulation were assessed using a dynamic plantar esthesiometer (# 37400; Ugo Basile), which consisted of a filament, which touched the plantar surface of the hindpaw and began to exert an upwards force until the paw was withdrawn. A cut-off of 50 g was applied to prevent tissue damage.

Paw edema was measured using a plethysmometer (#7140; Ugo Basile), as described previously (12). The paw volume was determined based on the displacement of water when the paw was submerged into the water. The volumes of the left and the right paws were measured, and the changes in paw volume were calculated as the percentage difference in volume between the ipsilateral and contralateral paw.

Nuclear and cytoplasmic fractions were extracted from the spinal cord using a Nuclear Extract kit (cat. no. 40010; Active Motif, Carlsbad, CA, USA). For the assessment of inflammatory mediators, protein was isolated from the spinal cord using cell lysis buffer (20 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA-Na₂, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, 1 mM Na₃VO₄ and 1 μg/ml leupeptin; Cell Signaling Technology, Inc., Beverly, MA, USA). The protein concentration was determined by a Bradford assay (Bio-Rad Labs Informatics Division, Philadelphia, PA, USA). Briefly, dye reagent was prepared (dilution, 1 part dye reagent concentrate with 4 part double-distilled water). The bovine serum albumin (BSA) protein standard was prepared by making serial dilutions from a stock solution. 10 μl BSA protein standards and samples were added and a blank of double-distilled water was used. After incubation for 5 min at room temperature, the absorbance was measured at 595 nm on a microplate reader (EL311S; Bio-Tek Instruments, Winooski, VT, USA). The optical density readings were taken and a standard curve was generated. The concentrations of samples were obtained based on the standard curve.

Nuclear protein levels of PPARα and NF-κB p65, protein levels of interleukin (IL)-1β, tumor necrosis factor (TNF)-α, cyclooxygenase (COX)-2 and inducible nitric oxide synthase (iNOS), and cytoplasmic protein levels of inhibitor of NF-κBα (IκBα), were detected using western blot analysis. Briefly, 20 μg protein was loaded onto 10% SDS-polyacrylamide gel (Bio-Rad Labs Informatics Division). Following electrophoresis (200 V for 35 min), proteins were transferred to a polyvinylidene fluoride membrane (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The membrane was blocked with Tris-buffered saline and Tween-20 (Bio-Rad Labs Informatics Division) with 5% skimmed milk, and incubated overnight at 4°C with the following primary antibodies (all purchased from Santa Cruz Biotechnology, Inc., Dallas, TX, USA): Rabbit polyclonal anti-human IL-1β (cat. no. sc-7884; 1:200); goat polyclonal anti-human TNF-α (cat. no. sc-1350; 1:400); goat polyclonal anti-human COX-2 (cat. no. sc-23983; 1:400); rabbit polyclonal anti-mouse iNOS (cat. no. sc-650; 1:200); rabbit polyclonal anti-human NF-κB p65 (cat. no. sc-109; 1:200); rabbit polyclonal anti-human IκBα (cat. no. sc-847; 1:300); rabbit polyclonal anti-human PPAR-α (cat. no. sc-9000; 1:200); and goat polyclonal anti-human β-actin (cat. no. sc-1615; 1:1,000). After washing three times, the membrane was incubated for 1 h at room temperature with the following secondary antibodies (all purchased from Santa Cruz Biotechnology, Inc.): Goat anti-rabbit horseradish peroxidase conjugated secondary antibody (cat. no. sc-2030; 1:5,000) or donkey anti-goat horseradish peroxidase conjugated secondary antibody (cat. no. sc-2020; 1:5,000). After washing three times, the signals of the detected proteins were visualized using an enhanced chemiluminescence reaction system (EMD Millipore, Billerica, MA, USA). The density of the bands was analyzed using National Institutes of Health ImageJ software (version 1.48; NIH, Bethesda, MD, USA) and all data were normalized to β-actin.

The DNA binding activity of PPARα and NF-κB p65 in the spinal nuclear extraction were measured using Transcription Factor Assay kits (cat. no. ab13112; Abcam, Cambridge, MA, USA and Active Motif), according to the manufacturer's protocols.

Whole blood samples were centrifuged at 2,100 × g for 15 min at 4°C and the plasma from each blood sample was collected. The plasma levels of glucose, total cholesterol and triglyceride were measured using a Roche Hitachi 911 Chemistry Analyzer (Hitachi Inc.,Pleasanton, CA, USA). The plasma levels of insulin, leptin and adiponectin were analyzed using Invitrogen rat ELISA kits (cat. nos. ERINS, KRC2281 and KRP0041, respectively; Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to the manufacturer's protocols.

GraphPad Prism 6 (GraphPad Software, Inc., La Jolla, CA, USA) was used for data analysis. Data are presented as the mean ± standard error of the mean. The differences between groups were analyzed by two-way analysis of variance, followed by Bonferroni post-hoc tests for multiple comparisons. P<0.05 was considered to indicate a statistically significant difference.

As shown in Fig. 1, body weights were significantly higher throughout the investigation in all HFD groups, compared with those in the LFD group, beginning at week 4. No differences in body weight were observed among the four HFD groups at 4 weeks. Compared with the HFD control rats, the HFD+UA rats exhibited significantly reduced body weight gain at 12 weeks. The intrathecal injections of PPARα siRNA or scrambled siRNA had no effects on body weight gain in the HFD+UA rats.

At 12 weeks, the plasma levels of insulin, leptin and cholesterol were higher, and the plasma levels of adiponectin were lower in the HFD control rats, compared with those in the LFD rats. Compared with the HFD control rats, the HFD+UA rats exhibited significant improvements in all the above parameters. The intrathecal injections of PPARα siRNA and scrambled siRNA did not alter these metabolic parameters in the HFD+UA rats. No differences in the plasma levels of glucose or triglycerides were observed between the experimental groups.

No differences were observed in thermal hyperalgesia, mechanical allodynia or paw edema between the experimental groups at baseline (prior to carrageenan injection), as shown in Fig. 2. Following carrageenan injection, all groups of rats exhibited significant thermal hyperalgesia, mechanical allodynia and paw edema in the injected paw. The maximum decreases in thermal response latency and mechanical response threshold, and maximum increase in paw volume, were observed in the carrageenan-injected paw at 6 h in the LFD rats, but at 4 h in the HFD control rats. Compared with the LFD rats, the HFD control rats had more pronounced thermal hyperalgesia between 4 and 8 h following carrageenan injection (Fig. 2A). Compared with the HFD control rats, the HFD+UA rats exhibited significantly attenuated thermal hyperalgesia. Of note, the intrathecal injection of PPARα siRNA, but not of scrambled siRNA, completely eliminated the beneficial effect of UA on thermal hyperalgesia in the HFD+UA rats in response to carrageenan injection. No significant differences were observed in the mechanical response thresholds between the experimental groups throughout the experimental period (Fig. 2B). Compared with the LFD rats, the HFD control rats exhibited increased paw edema between 2 and 24 h following carrageenan injection (Fig. 2C). The HFD+UA rats exhibited significantly reduced paw edema, compared with the HFD control rats. The improvement in paw edema observed in the HFD+UA rats was completely abrogated by the intrathecal injection of PPARα siRNA, but not scrambled siRNA.

As the inflammatory response in the spinal cord has been shown to be associated with peripheral inflammation and pain sensitivity, the present study examined the effect of systemic administration of UA on the expression levels of inflammatory mediators and activation of NF-kB, a key mediator of inflammation, in the spinal cord at baseline and 6 h post-carrageenan injection. The time points selected were based on previous studies showing that the acute phase of the inflammatory response and hyperalgesia (0–6 h) is characterized by central sensitization, mediated predominantly by prostanoids, the products of COX-2 (11,23).

No differences were observed in the expression levels of the IL-1β, TNF-α, COX-2 or iNOS inflammatory mediators (Fig. 3), the activity of NF-κB p65 or the expression of IκBα (Fig. 4) in the spinal cord between the experimental groups at baseline. At 6 h post-carrageenan injection, all groups of rats exhibited markedly increased expression levels of inflammatory mediators, augmented expression and activity of NF-κB p65, and reduced expression of IκBα in the spinal cord, compared with respective baseline values. Notably, the HFD control rats exhibited a more pronounced inflammatory response, which was prevented in the HFD+UA rats. The decreased inflammatory response in the spinal cord of the HFD+UA rats was reversed by intrathecal injection of PPARα siRNA, but not scrambled siRNA.

To further investigate the molecular mechanisms by which UA attenuates exaggerated inflammatory response in the spinal cord, and prevents augmented peripheral inflammation and inflammatory hyperalgesia in obesity, the present study measured the expression and activity of PPARα in the spinal cord.

At 12 weeks, the HFD control rats exhibited significant decreases in the expression and activity of PPARα at baseline, compared with the LFD rats (Fig. 5). Of note, the decreases in the spinal expression and activity of PPARα were normalized in the HFD+UA rats. However, intrathecal injection of PPARα siRNA in the HFD+UA rats reduced the spinal expression and activity of PPARα to the same extent as in the HFD control rats. Intrathecal injection of scrambled siRNA in the HFD+UA rats had no effects on the expression and activity of PPARα.

Carrageenan injection caused significant reductions in the spinal expression and activity of PPARα in all groups, compared with their baseline values. Of note, the spinal expression and activity of PPARα remained significantly higher in the LFD rats, HFD+UA rats and HFD+UA rats treated with scrambled siRNA, compared with the other two groups.