Adult male Sprague-Dawley rats weighing 180–220 g were purchased from Liaoning Changsheng Biological Center, Shenyang, China. The rats were housed in separated cages at 23–25°C and 50–60% humidity under a 12/12-h light/dark cycle with free access to water and food. The animal experiments were performed according to the Guiding Principles for Research Involving Animal and Human Beings, and the experimental procedures were approved by the Animal Care and Use Committee of China Medical University (IACUC no. 2019028).

The rat model of neuropathic pain was induced by unilateral SNI as previously described (Decosterd and Woolf, 2000; Guo et al., 2019). Briefly, rats were anesthetized by inhalation of 3% isoflurane, and a section was made directly through the biceps femoris muscle to expose the sciatic nerve and its three terminal branches: the common peroneal, tibial and sural nerves. The common peroneal and the tibial nerves were ligated tightly with 5.0 silk and transected distal to the ligation. Approximately 2–3 mm of the distal nerve stump was then excised. Muscle layers were closed using 4-0 chromic gut, and the skin incision was closed with wound staples. In sham surgery, the sciatic nerves were exposed but not ligated. Rats with sham surgery were used as control.

Lumbar intrathecal catheter implantation was conducted as previously described (Størkson et al., 1996; Guo et al., 2019). Briefly, under 3% isoflurane anesthesia, an incision lateral to the midline was made and the polyethylene catheter was inserted into the subarachnoid space. The correct intrathecal localization was confirmed by a tail-flicking action and hind limb paralysis after administration of 2% lidocaine (10 µL) through the catheter in wakened animals.

Spinal cord (L4-6) was homogenized in ice-cold lysis buffer (Sigma-Aldrich, St. Louis, MO, United States) containing protease inhibitor cocktail. Supernatants were collected by centrifugation at 12,000 × g for 20 min at 4°C. The nuclear protein from spinal cord was prepared using a NE-PER Nuclear and Cytoplasmic Extraction Reagents (Thermo Fisher Scientific, Waltham, MA, United States) according to the manufacturer’s instructions. Protein concentrations were determined with the BCA protein assay (Thermo Fisher Scientific, Waltham, MA, United States). Equal amounts of protein were separated by 8% SDS-electrophoresis and transferred onto polyvinylidene difluoride membranes. The membranes were placed in blocking buffer (5% milk in Tris-buffered saline with Tween-20) for one hour and then incubated over night at 4°C with primary antibodies to Sirt2 (1:500, Abcam, Cambridge, United Kingdom), Nrf2 (1:1,000, Abcam, Cambridge, United Kingdom), NQO1 (1:10,000, Abcam, Cambridge, United Kingdom), lamin B1 (1:1,000, Cell Signal Technology, MA, United Kingdom) and β-actin (1:1,000, Cell Signal Technology, Beverly, MA, United States). After incubation with horseradish peroxidase-conjugated secondary antibodies (1:10,000, Santa Cruz Biotechnology, Santa Cruz, CA, United States) for 1 h at room temperature, the labeled proteins were visualized by enhanced chemiluminescence detection system (GE Healthcare, Waukesha, WI, United States) and analyzed with ImageJ software (NIH, Bethesda, Maryland, United States). Results were normalized to β-actin or lamin B.

Spinal cord samples were collected and lyzed with RIPA buffer. The enzymatic activity of superoxide dismutase (SOD) was measured using the T-SOD assay kit (Jiancheng Bioengineering Institute, Nanjing, China) and levels of 8-hydroxy-2′-deoxyguanosine (8-OHdG) were determined using ELISA kits (Thermo Scientific, Rockford, IL, United States) according to the manufacturer’s instructions.

Mechanical allodynia was assessed by measuring the paw withdrawal threshold (PWT) in response to the stimulation of Von Frey filaments, as previously described (Chaplan et al., 1994; Guo et al., 2019). Briefly, rats were acclimatized in a plastic cage with a mesh bottom for 20 min prior to testing. PWT was assessed using a dynamic plantar esthesiometer (Ugo Basile, 37450, Italy), which consists of a force transduction fitted with a 0.5-mm diameter polypropylene rigid tip. A probe was applied perpendicularly to the mid-plantar surface of the hind paw with an increasing pressure. The cutoff pressure was set to be 50 g and the force that induced the withdrawal response was automatically recorded by the esthesiometer. For each animal, at least three measurements were performed with an interval of 5 min for the stimulation of each hind paw. Thermal hyperalgesia was determined by measuring the paw withdrawal latencies (PWL) in response to heat plate test. A hot plate (Ugo Basile Srl 7280, Gemonio, Italy) with a pre-set plate temperature of 52.5 °C as recommended for rats (Hestehave et al., 2019) was used. As soon as the rat was placed onto the hot plate, the time between placement and licking, shaking or stepping of the hindpaws was recorded (Bannon and Malmberg, 2007). A cut-off-time was set at 30s to avoid tissue damage.

The immunofluorescent study was performed as previously described (Wang et al., 2014), Briefly, rats were transcardially perfused with saline containing heparin (1 unit/ml) followed by 4% paraformaldehyde in 0.1 M PBS. Spinal cords were removed and fixed overnight in 4% paraformaldehyde at 4°C and cryoprotected with 30% sucrose in 0.1 M PBS for 2 days. The spinal cords were sliced into 18-µm coronal sections. After being blocked with 5% goat serum in 0.3% Triton for 1 h at room temperature, the sections were incubated with primary to Sirt2 (1:200, Abcam, Cambridge, United Kingdom or Nrf2 (1:200, Abcam, Cambridge, United Kingdom) overnight at 4 °C, followed by Alex Fluor 568 second antibody (1:200, Thermo Fisher Scientific, Waltham, MA, United States). DAPI was used for nuclear staining. Images were captured using a confocal laser-scanning microscope (Zeiss LSM 510, Carl Zeiss, Inc.).

All data are presented as the mean ± SE. Statistical analyses were performed using GraphPad Prism 7 (GraphPad Software, Inc.). The differences between groups were analyzed by a one-way or two-way analysis of variance (ANOVA) followed by Bonferroni post hoc tests for multiple comparisons. Statistical significance was reached with P values below 0.05.

Baseline measures of mechanical allodynia and thermal hyperalgesia were recorded 24 h prior to SNI to determine preinjury thresholds. NP behaviors were examined on 1, 3, 7, 10 and 14 days following SNI. As shown in Figure 2, there were no differences in mechanical allodynia (Figure 2A) and thermal hyperalgesia (Figure 2B) between groups at baseline. SNI induced significant mechanical allodynia as indicated by decreased PWT and thermal hyperalgesia as evidenced by reduced PWL within 1 day and lasting up to 14 days compared to sham controls and baseline. The maximal mechanical allodynia and thermal hyperalgesia in SNI group were observed at day 10 and day 7, respectively.

To explore the potential role of Nrf2 pathway in regulation of NP, we first examined the expression of Nrf2 and NQO1, a downstream target of Nrf2 that plays a protective role in various cells against oxidative stress, in the spinal cord of SNI rats. Western blot analysis revealed that the expression of Nrf2 (Figures 2C,E) and NQO1 (Figures 2D,E) in the whole tissue lysates of SNI group tended to be higher at day 1, but gradually decreased from day 7 following SNI, compared with baseline or sham group. Expression of Nrf2 in the nuclear fractions (Figures 2F,G) in SNI group was significantly increased at day 1, but then started to gradually decrease in the following day. Significant reduction in expression of Nrf2 in the nuclear fractions of SNI group was observed from day 7 as compared to baseline or sham group.

To examine whether activation of Nrf2 pathway in the spinal cord of SNI rats would ameliorate NP, SNI rats were treated with intrathecal injection of DMSO (vehicle) or Nrf2 agonist tBHQ at different doses. The mechanical allodynia (Figure 3A) and thermal hyperalgesia (Figure 3B) in SNI rats, compared with SNI rats without treatment, were attenuated by intrathecal tBHQ at both doses from day 4 for PWT and day 2 for PWL, respectively. Intrathecal injection of DMSO had no effects on mechanical allodynia and thermal hyperalgesia in SNI rats.

Western blots confirmed that expression of Nrf2 (Figures 3C,E) and NQO1 (Figures 3D,E) in the whole tissue lysates and expression of Nrf2 in the nuclear fractions (Figures 3F,G) in SNI rats without treatment were significantly decreased when compared with sham rats. Intrathecal tBHQ at either dose, but not DMSO, increased expression of Nrf2 and NQO1 in the whole tissue lysates and expression of Nrf2 in the nuclear fractions.

Immunofluorescent study showed that SNI rats treated with intrathecal tBHQ at a dose of 10 μM exhibited abundant Nrf2 immunoreactivity in the spinal cord, particularly in the nucleus, compared with SNI rats treated with intrathecal DMSO (Figure 4).

Because Nrf2 pathway mediates oxidative stress that has been implicated in the pathogenesis of NP, we also measured the levels of antioxidant enzyme SOD and oxidative stress marker 8-OHdG in the spinal cord. Consistent with expression of Nrf2 and NQO1, the levels of SOD were markedly decreased (Figure 5A), whereas the levels of 8-OHdG (Figure 5B) were increased in SNI rats without treatment. Intrathecal tBHQ at both doses completely reversed SNI-induced changes in SOD and 8-OHdG. Of note, intrathecal DMSO did not alter SNI-induced changes in SOD and 8-OHdG.

New evidence reveals that Sirt2 regulates Nrf2 pathway in different tissues. We next examined the expression of Sirt2 in the spinal cord following SNI. As shown in Figure 6, the expression of Sirt2 in SNI rats was not altered at day 1 and day 3, but it was significantly decreased from day 7 and remained lower till the end of the experiment compared to baseline or sham rats.

To further determine whether upregulation of Sirt2 in the spinal cord could alleviate NP, we treated SNI rats with intrathecal injection of a recombinant adenovirus expressing Sirt2 and GFP (Ad-Sirt2) or control and GFP (Ad-control). As presented in Figure 7, SNI rats that received intrathecal Ad-Sirt2, but not Ad-control, had significantly reduced mechanical allodynia (Figure 7A) and thermal hyperalgesia (Figure 7B) as compared to SNI rats without treatment.

The expression of Sirt2 was significantly lower in the spinal cord of SNI rats without treatment compared with sham rats. Intrathecal Ad-Sirt2 restored expression of Sirt2 in the spinal cord of SNI rats to similar level as in sham group (Figures 7C,D).

Immunofluorescent study demonstrated GFP expression in the spinal cord in both SNI rats treated with intrathecal Ad-Sirt2 and SNI rats treated with intrathecal Ad-control. However, SNI rats treated with intrathecal Ad-Sirt2 had higher Sirt2 immunoreactivity than SNI rats treated with intrathecal Ad-control (Figure 8).

Importantly, we found that expression of Nrf2 (Figures 9A,C) and NQO1 (Figures 9B,C) in the whole tissue lysates and expression of Nrf2 in the nuclear fractions (Figures 9D,E) in SNI rats treated with intrathecal Ad-Sirt2 were normalized to those observed in sham rats. Moreover, intrathecal Ad-Sirt2 restored SNI-induced changes in SOD (Figure 9F) and 8-OHdG (Figure 9G) in the spinal cord. Intrathecal Ad-control had no effect on any of these measured parameters.