Soluble Epoxide Hydrolase Inhibitor TPPU Alleviates Nab-Paclitaxel-Induced Peripheral Neuropathic Pain via Suppressing NF-κB Signalling in the Spinal Cord of a Rat

Objective Paclitaxel-induced peripheral neuropathy (PIPN) is a debilitating and difficult-to-treat side effect of paclitaxel. Soluble epoxide hydrolase (sEH) can rapidly metabolize the endogenous anti-inflammatory mediators' epoxyeicosatrienoic acids (EETs) to dihydroxyeicosatrienoic acids. This study aimed to assess whether the sEH inhibitor N-(1-(1-oxopropy)-4-piperidinyl]-N′-(trifluoromethoxy) phenyl)-urea (TPPU) plays a critical role in PIPN of rats and provides a new target for treatment. Methods A Sprague–Dawley male rat model of PIPN induced by nab-paclitaxel was established. Rats were randomly divided into a control group, nab-paclitaxel group, and nab-paclitaxel + TPPU (sEH inhibitor) group, with 36 rats in each group. The effects of the sEH inhibitor TPPU on behavioural assays, apoptosis, glial activation, axonal injury, microstructure, and permeability of the blood-spinal cord barrier were detected, and the underlying mechanisms were explored by examining the expression of NF-κB signalling pathways, inflammatory cytokines, and oxidative stress. Results The results showed that the mechanical and thermal pain thresholds of rats were decreased after nab-paclitaxel treatment, accompanied by an increased expression of axonal injury-related proteins, enhanced cell apoptosis, aggravated destruction of vascular permeability, intense glial responses, and elevated inflammatory cytokines and oxidative stress in the L4-L6 spinal cord. TPPU restored the mechanical and thermal thresholds, decreased cell apoptosis, alleviated axonal injury and glial responses, and protected vascular permeability by increasing the expression of tight junction proteins. TPPU relieved PIPN by inhibiting the activation of the sEH and NF-κB signalling pathways by decreasing the levels of inflammatory cytokines and oxidative stress. Conclusion These findings support a role for sEH in PIPN and suggest that the inhibition of sEH represents a potential new therapeutic target for PIPN.


Introduction
Paclitaxel-induced neuropathic pain (PINP) is a common and difcult-to-treat side efect that can persist for months to years after paclitaxel treatment. PINP could result in signifcant symptoms such as burning pain and allodynia and afect the quality of life of patients [1]. A previous study found that the mechanism of PINP might be related to neuroinfammation, mitochondrial dysfunction, and disordered cation channels [2]. Although several drugs, including anti-infammatory drugs, opioids, and antidepressants, are recommended to treat PINP, they cannot obtain satisfactory results in clinical trials [2]. Te concrete mechanism of PINP needs to be further studied.
Epoxyeicosatrienoic acids (EETs) are important angiogenic mediators and could be degraded by soluble epoxide hydrolase (sEH) [3]. EETs have been shown to exert benefcial efects on neuropathic pain [3]. A previous study found that EETs could prevent continued nerve damage by decreasing the levels of infammatory factors and promoting nerve regeneration [3]. EET actions also provide superior pain relief involving both the endogenous opioid and GABAergic systems [3]. Genetic deletion or pharmacologic inhibition of sEH have been demonstrated to greatly help in maintaining or elevating EET levels in vivo to achieve benefcial efects in several disease models [4]. Nonetheless, the role of sEH inhibitors in PINP is still unknown.
Te current study aimed to investigate whether sEH inhibition could alleviate PIPN by infuencing the integrity of the blood-spinal cord barrier (BSCB). To simulate the PIPN of humans, a rat model of PIPN was established by nab-paclitaxel. We detected the efects of the sEH inhibitor on behavioural assays, apoptosis, glial activation, axonal injury, microstructure, and permeability of the BSCB. Our results showed that the protective efects of the sEH inhibitor were associated with decreases in levels of infammatory factors and oxidative stress, suggesting that the sEH inhibitor may be a prospective target in the therapy of PINP.

Experimental Animals and Ethical Approval.
Male SD rats were obtained from the Experimental Animal Center of Xi'an Jiaotong University and the number of the animal license is SCXK (Shaanxi) 2006-001. All the rats were 8-10 weeks old and weighed 250-300 g. Te Biomedical Ethics Committee for Animal Experiments of Shaanxi Province (China, 2021-0145) approved this investigation. All procedures were carried out in accordance with the Guidelines for the Care and Use of Laboratory Animals. Te rats had free access to food and water and housed 5 per cage. Rats were raised at an ambient temperature of 22 ± 1°C on a 12-h light/dark cycle.

Foundation of the Rat Model of PIPN.
In this study, to simulate a paclitaxel-induced neuropathic pain model, nabpaclitaxel (Shi Yao Group Ou Yi Pharmaceutical Co., Ltd., Hebei Province, China) was dissolved in saline. Rats were intraperitoneally injected with nab-paclitaxel on four alternate days (2.47 mg/kg on days 1, 3, 5, and 7) (Figure 1(a)) [5,6]. Rats were given the nab-paclitaxel with a fnal cumulative dose of 9.9 mg/kg. Te same dose of saline was given to the control rats. Te rats were fed at a temperature of 22 ± 2°C after treatment until euthanasia. No rats died during the experiment. Rats were euthanized by intraperitoneally injecting 5% (w/v) pentobarbital sodium.

Groups and Drug
Administration. Sample size estimation was calculated by a statistical power analysis. A one-way ANOVA with three groups was used with the assumption that α � 0.05 and power � 0.80. Te suggested sample size was n � 36 per group (6 rats for pathology; 6 rats for TEM; 6 rats for western blot; 6 rats for ELISA and oxidative stress detection; 6 rats for Evans blue detection; 6 rats for spinal cord water content). By using a randomized digital table, the total 108 rats (sample size was decided according to a previous study) were randomized into three groups [7,8]: (a) 36 rats in the control group; (b) 36 rats in the nab-paclitaxel (Nab-P) group, which was treated with nab-paclitaxel for four alternate days (days 1, 3, 5, and 7); and (c) 36 rats in the nab-paclitaxel + N-(1-(1-oxopropyl)-4-piperidinyl)-N′-(4-(trifuoromethoxy)phenyl)-urea (TPPU, sEH inhibitor) (nab-P + TPPU) group. Rats in the nab-P + TPPU group were treated with nab-paclitaxel in the same manner as the nab-P group and TPPU. In this group, TPPU (Merck KGaA, Darmstadt, Germany) was initially dissolved in 0.1% dimethyl sulfoxide (DMSO) as a stock solution and diluted to 1 mM as the working solution before each study [7,8]. TPPU (3 mg/kg/day) was intraperitoneally injected on 14 consecutive days using a microinjection syringe (Figure 1(a)) [7,8]. All rats were sacrifced on day 15 after the frst nabpaclitaxel treatment. Mechanical hyperpathia and heat hypersensitivity were tested on 4, 9, and 14 days after the frst treatment of nab-paclitaxel, respectively (Figure 1(a)). Te order of treatments and measurements is randomized. Apart from the conductor, none was aware of the group allocation at the diferent stages of the experiment.

Behavioural Assays.
Te paw withdrawal threshold (PWT) and thermal withdrawal latency (TWL) were used to perform behavioural assays. PWT and TWL were tested at 0, 4, 9, and 14 days after the frst nab-paclitaxel treatment at 1 day (Figure 1(a)). Calibrated von Frey flaments (Stoelting, WoodDale, USA) were applied to measure PWT after acclimation for 15 minutes [9]. Rat feet were stimulated with von Frey flaments, ranging from 2 to 26 × g bending force, and the minimum stimulus intensity (g) that resulted in feet shrinkage was recorded. Each irritation was consecutive for 5 s, and the interval between two irritation is 20 s. A plantar tester and an infrared radiant heat stimulus generator (Ugo Basile, Varese, Italy) were used to test the TWL of the hind paws [9]. A radiant heat source was used to focus on the hindlimb plantar. A foot shrinkage occurred as soon as the rats sufered pain. Te delay time (s) of paw withdrawal was considered TWL [9]. All rats were tested 4 times and the timepiece between tests is 5 min.

Haematoxylin-Eosin
Staining. Te L4-L6 spinal cord was excised from the rat following transcardial perfusion with saline followed by 4% paraformaldehyde. Tey were sectioned longitudinally into 5 μm sections before staining. Haematoxylin was used to stain tissue sections for 2.5 min and eosin for 16 s. Ten, sections were obtained by dehydration, hyalinization, and fxation. Sections were observed under a high-power light microscope.
2.6. Immunoblot. Te L4-L6 spinal cord was homogenized in a protease inhibitor lysis bufer, and the supernatants were collected. A bicinchoninic acid protein assay kit was used to determine the protein concentration. A total of 30-50 μg of protein was separated using sodium dodecyl sulfatepolyacrylamide gel electrophoresis. Te protein was then transferred to polyvinylidene difuoride membranes.  were detected by enhanced chemiluminescence. A luminescent image analyser was used to acquire images. Quantifcation of the western blot bands was performed using Image J software. Te results were normalized to β-actin.

Terminal Deoxynucleotidyl Transferase-MediateddUTP-
Biotin Nick End Labelling (TUNEL) Assay. Cell apoptosis was confrmed by a DeadEnd Fluorometric TUNEL System (Promega, Madison, Wisconsin, USA) using parafn sections. Briefy, parafn sections of L4-L6 spinal cords were deparafnized, rehydrated, and digested with proteinase K. Sections were then covered with an equilibration bufer and incubated with TdT Labelling Reaction Mixture in a humid atmosphere. After that, the nuclei were stained with DAPI. Six felds were randomly selected for the count of apoptotic cells at 400x magnifcation.

Transmission Electron Microscopy.
Te ultrastructural changes were examined by transmission electron microscopy (TEM) in the L4-L6 spinal cords of rats. After being cut into 1 mm 3 sections, spinal cords were fxed in glutaraldehyde (2.5%) overnight. Ten, an ethanol and acetone gradient series was used to dehydrate the samples. Next, the tissue was soaked in linoleate at 60°C overnight. Ultramicrotome was used to cut slices after methylene blue staining. Finally, TEM (H-7650, Hitachi, Tokyo, Japan) was used to observe the spinal cord ultrastructure at 20,000x and 8,000x magnifcation.
2.11. Immunofuorescence Staining. Te L4-L6 spinal cord was removed, fxed in 4% paraformaldehyde, and then embedded in parafn. As with the procedure of immunohistochemistry, sections were then incubated with goat polyclonal anti-ZO-1 (1 : 200, Abcam) overnight at 4°C. Te fuorochrome-conjugated secondary antibody was used to incubate sections. A fuorescence microscope was used to observe the stained slices.

Vascular Permeability Assay.
Te vascular permeability of the spinal cord was assessed by Evans blue. Evans blue (2%, 4 mL/kg of body weight, Sigma-Aldrich, St. Louis, MO, USA) was injected intravenously via the tail vein. After 1 h, rats were perfused with normal saline. L4-L6 of the spinal cord was quickly homogenized. Te obtained proteins were precipitated with 50% trichloroacetic acid overnight. Te supernatant was obtained after centrifugation. Te wavelength is set to 610 nm in order to examine absorbance. Te content of Evans blue was presented as micrograms per Gram of spinal cord tissues.

Evaluation of Spinal Cord
Oedema. Te wet-dry method was employed to assess spinal cord water content. Briefy, the L4-L6 spinal cord wet weight was measured. An oven was used to dry the spinal cord at 105°C for 72 h. Te dry weight was measured. Spinal cord water content (%) was calculated by using the formula: (wet weight − dry weight)/wet weight × 100% [11].
2.14. Oxidative Stress Marker Measurement. Te L4-L6 spinal cords were prepared, homogenized, and centrifuged. After protein quantifcation, tissue lysates were analysed to detect the levels and activities of malondialdehyde (MDA), superoxide dismutase (SOD), glutathione peroxidase (GSH), and catalase (CAT) according to procedures. Te concentrations were calculated with reference to standard curves.

Statistical Analysis.
Te results are expressed as the mean ± standard deviation (SD). Statistical analysis was performed by SPSS 18.0 (SPSS, Chicago, IL, USA). One-way analysis of variance (ANOVA) was used to analyse data in more than 2 groups, followed by LSD (L) to conduct a posthoc test. A P value less than 0.05 was considered statistically signifcant. Shapiro-Wilk's method is recommended for the normality test. A normal distribution of all the data is shown.

Efects of the sEH Inhibitor on PINP and Pathological
Changes in the Rat's Spinal Cord. Te PWT and TWL were lower in the nab-P group at d4, d9, and d14 (P all <0.001) in comparison with the control group. Compared to the nab-P group at d9 and d14, the PWT and TWL were signifcantly increased in the Nab-P + TPPU group (P all <0.001), while there was no signifcant diference between these two groups at d4 (P > 0.05) (Figure 1(b)). Te histopathological features and pathological changes in the rat spinal cord were evaluated. Te cell morphology was normal in the control group, while neuronal pyknosis, contortion, and variant were detected in the nab-P group. Compared to the nab-P group, the histopathological lesions were alleviated in the Nab-P + TPPU group (Figure 1(c)). TEM images showed that the structure of mitochondria was integrated and that the structures around the microvessels were intact. No interstices, oedema, or distortion were observed. In the nab-P group, the structures around the microvessels were destroyed with large interstices and oedema. Mitochondria vacuoles, fracture, swelling, and disordered sparse cristae were observed. In the nab-P + TPPU group, the damage to microvessels and mitochondria was mitigated, with fewer interstices, vacuoles, fractures, and swelling than in the Nab-P group (Figure 1(c)).

sEH Inhibition Alleviated Apoptosis and Glial Responses
after Nab-Paclitaxel Treatment. Te expression of GFAP and Iba-1 is considered a marker of the glial response. Te expression of GFAP and Iba-1 was signifcantly higher in the nab-P group (P � 0.005, P � 0.004, respectively) in comparison with the control group. Compared with the nab-P group, the expression of GFAP and Iba-1 was lower in the nab-P + TPPU group (Figure 3(a)) (P � 0.005, P � 0.006, respectively). TUNEL-positive cells were barely observed in the control group. Te number of TUNEL-positive cells was higher in the nab-P group (P � 0.006, P � 0.008, respectively) in comparison with the control group. Compared with the nab-P group, the number of apoptotic cells was lower in the nab-P + TPPU group (Figure 3(b)) (P � 0.007, P � 0.007, respectively).

TPPU Ameliorated Vascular Permeability after Nab-Paclitaxel Treatment.
Te expressions of ZO-1, claudin-5, and occludin-1 were decreased in the nab-P group (P � 0.002, P � 0.004, and P � 0.005, respectively) in comparison with the control group. sEH inhibition by TPPU signifcantly upregulated the expression of ZO-1, claudin-5 and occludin-1 compared to the nab-P group (Figure 4(a)) (P � 0.003, P � 0.004, and P � 0.006, respectively). Little Evans blue difusion was detected in the control group, while the level of Evans blue was signifcantly increased in the Nab-P group, and the content of Evans blue was decreased in the nab-P + TPPU group compared to the Nab-P group (P all <0.001). Compared with the control group, the oedema of the spinal cord assessed by water content was signifcantly higher in the nab-P group. Te water content in the nab-P + TPPU group was decreased than that in the nab-P group (Figure 4(b)) (P all <0.001).

Te Neuroprotective Efects of TPPU Were Related to the sEH/NF-κB Pathway and Downstream Infammatory Cytokines and Oxidative Stress Levels.
Te expressions of sEH and NF-κB were higher in the nab-P group (P all <0.001) in comparison with the control group. Compared to the nab-P group, the expression levels of sEH and NF-κB were decreased in the Nab-P + TPPU group ( Figure 5(a)) (P � 0.008 , P � 0.009, respectively). Te levels of proinfammatory factors (TNF-α, IL-1β, and IL-6) were higher and the levels of anti-infammatory factors (IL-4 and IL-10) were lower in the spinal cords of rats in the nab-P group (P all <0.001) in comparison with the control group. Compared to the nab-P group, the levels of TNF-α, IL-1β, and IL-6 were reduced, and the levels of IL-4 and IL-10 were elevated in the nab-P + TPPU group ( Figure 5(b)) (P all <0.001). Compared to the control group, the levels of MDA were increased and the levels of SOD, GSH, and CAT were decreased in the nab-P group (P all <0.001). Compared to the nab-P group, the levels of MDA were lower, and the levels of SOD, GSH, and CAT were higher in the nab-P + TPPU group ( Figure 5(c)) (P all <0.001).

Discussion
Te bioactivity of EETs is transient in vivo principally due to the rapid metabolism by sEH. Pharmacologic inhibition of sEH has been demonstrated to greatly help in maintaining or elevating EET levels in vivo to achieve benefcial efects in neuropathic pain [3,12]. For the frst time, we demonstrated that sEH inhibition with TPPU alleviates mechanical and thermal pain in a model of nab-paclitaxel-induced neuropathic pain. Te ameliorative neuropathic pain was associated with improved vascular permeability and decreased spinal cord apoptosis, glial response, and axonal damage. Tese protective efects of TPPU treatment were related to decreased levels of infammatory cytokines and oxidative stress by inhibiting the activation of the sEH and NF-κB signalling pathways in the rat spinal cord. Tese data suggest that sEH inhibition may be a potential therapeutic target of PIPN in rats.
Neuroflaments could encode a neuronal protein and play roles in nerve conduction velocity and axonal transport [13]. NF-L has been regarded as a biomarker of neurologic damage in disease states such as amyotrophic lateral Pain Research and Management sclerosis, diabetes, and Parkinson's disease and increases with neurotoxic chemotherapy [13,14]. Serum NF-L and chemotherapy-induced peripheral neuropathy (CIPN) symptoms increased concurrently during taxane treatment [13]. It is possible that NF-L can be used to diferentiate women who will develop CIPN during taxane treatment. A previous study found that small interfering RNA-7a could ameliorate neuropathic pain by blocking the activator of the transcription signalling pathway by repressing neuroflament light polypeptide in a spinal nerve ligation rat model [15]. In this study, nab-paclitaxel treatment could increase the expression levels of NF-L, NF-M, and NF-H. sEH inhibition with TPPU could decrease the expression of NF-M, NF-H, and NF-L accompanied by improved mechanical and thermal pain, which indicated that nab-paclitaxel could induce axonal injury in the spinal cord and that neuroflaments might be biomarkers of the development of PIPN.
It has been found that chemotherapeutic agents can accumulate in peripheral nerves and cause neurotoxicity as well as signifcant nociceptive hypersensitivity [16]. Te neurotoxicity caused by chemotherapy drugs might be related to the disruption of the integrity of the BSCB. For example, vincristine treatment afects the integrity of the BSCB by infuencing Evans blue extravasation and tight junction disruption in the spinal cord [16]. Te endothelial cells that form the BSCB are tight junction-coupled proteins and include claudins, zonula occludens and occludin [17]. After nab-paclitaxel treatment, we detected the downregulation of junction-coupled proteins in the spinal cord tissue. We also found signifcant increases in EB extravasation and water content. Additionally, the ultrastructure structures surrounding the microvessels were destroyed, with large interstices and oedema. Specifcally, TPPU treatment increased the expression levels of the tight junction proteins claudin-5, occludin-1, and ZO-1 in the spinal cord and decreased EB extravasation and water content in neuropathic pain rat models. Terefore, alterations in the permeability of the BSCB after nab-paclitaxel treatment could be related to changes in expression of tight junction protein [17].
EET plays crucial roles in pain, including osteoarthritis knee pain, neuropathic pain, and central poststroke pain [3,8,18]. However, EETs are rapidly hydrolysed into dihydroxyeicosatrienoic acid when exposed to sEH.  Pain Research and Management Pharmacologic inhibition of sEH could increase EET levels. Te sEH inhibitor TPPU is a novel antihypertensive and anti-atherosclerotic pharmaceutical that attenuated vascular permeability, decreased axonal damage assessed by neuroflament proteins, lowered cell apoptosis, and alleviated glial response in this study. Te protective efect of TPPU was related to decreased levels of infammatory cytokines and oxidative stress by inhibiting activation of the sEH and NF-κB signalling pathways in the rat spinal cord. Te activation of the NF-κB signalling pathway is closely correlated with the expansion of infammatory reactions.
Te activation of the NF-κB pathway and upregulation of infammatory factors have been confrmed to aggravate neuropathic pain [19,20]. In this study, we found the  progression of the NF-κB pathway in PIPN and found that TPPU reversed the protein levels of sEH, NF-κB, and downstream infammatory cytokines and oxidative stress levels which promoted by nab-paclitaxel. TPPU decreased the levels of proinfammatory factors, including TNF-α, IL-1β, and IL-6, and increased the levels of anti-infammatory factors, including IL-4 and IL-10, in the spinal cord. Similar to our fndings, a study found that TIPE2 reduced the expression of TAK1, thereby inhibiting the activated pathway of NF-κB and further improving sciatic nerve injuryinduced neuropathic pain [19]. A growing number of studies suggest that the development and maintenance of neuropathic pain were related to mitochondrial dysfunction and abnormal oxidative stress [21,22]. In this study, the levels and activities of SOD, CAT, MDA, and GSH were tested, and the results showed that nab-paclitaxel treatment could downregulate antioxidant levels and upregulate oxidative stress. Mitochondrial vacuoles, fractures, swelling, and disordered sparse cristae were also observed after nab-paclitaxel treatment. Te sEH inhibitor alleviated mitochondrial injury by downregulating oxidative stress. A previous study found that mitoxantrone reduces neuropathic pain by alleviating mitochondrial dysfunction, inhibiting oxidative stress, and relieving apoptosis [23].
In this study, we also found that the activation of microglial cells and astrocytes was signifcantly enhanced after nab-paclitaxel treatment. TPPU decreased the activation of glial response. Similar to previous studies, our results also suggested that nab-paclitaxel may promote the activation of glial cells, thereby facilitating the release of infammatory cytokines and contributing to the aggravation of PIPN [24,25]. Infammatory cytokines could activate microglial cells and astrocytes. Microglia and astrocytes are also most afected by a reduction in the infammatory response [26]. Our previous study also found that inhibition of the Notch pathway could reduce the expression of infammatory factors accompanied by repressive glia response through the HMGB1/TLR4 signalling pathway in a rat  model of nab-paclitaxel-induced peripheral neuropathy [27]. All the results suggested that TPPU may suppress the activation of the glial response through the regulation of infammatory cytokines. 8-10 weeks-old male rats were used to conduct this research which was consistent with previous studies. Multivariable linear regression analyses which predicted factors of CIPN severity revealed that the male sex was associated with greater CIPN [28]. Tus, the male sex seems to sufer CIPN more easily. Recently, more and more studies were conducted by using both male and female animals simultaneously. However, it is still difcult to determine sex diferences from the current literature about CIPN. Te 8-10 weeks-old rats correspond to young adults of humans. In several previous studies about CIPN, young adult animals were used as test subjects. Moreover, in the previous study, spared nerve injury of SD rats could perform at 33 days after birth and lead to signifcant and persistent allodynia with the threshold falling to 55% of control values [29]. It means that mechanical allodynia can be evoked in very young animals with infammatory pain. However, the median age of many common solid cancers at diagnosis is between 60 to 70 years. Te age of young adult rats used in many animal experiments is contradicted by epidemiological evidence of humans [30]. Aging also could change the function and structure of the peripheral nerves and inhibit compensatory reinnervation [31]. Tus, in order to gain more generalizability of fndings from animal studies of CIPN, the inclusion of groups of older animals should be considered [31].
In conclusion, our current study shows that sEH inhibition by TPPU could alleviate PIPN by decreasing cell apoptosis, glial activation, and axonal injury and ameliorating the integrity of the blood-spinal cord barrier through the sEH/NF-κB signalling pathway and downstream infammatory cytokines and oxidative stress. Tese results suggest that the sEH inhibitor TPPU can serve as a potential drug target for the treatment of PINP.

Data Availability
Some or all data, models, or code generated or used during the study are available in a repository or online in accordance with funder data retention policies (provide full citations that include URLs or DOIs.).

Conflicts of Interest
Te authors declare that they have no conficts of interest.