Up-regulation of oxytocin receptors on peripheral sensory neurons mediates analgesia in chemotherapy-induced neuropathic pain

Key Results: Expression of OXTR in DRG neurons was enhanced significantly after PTX treatment. Activation of OXTR exhibited antinociceptive effects, by decreasing the hyperexcitability of DRG neurons in PTX-treated mice. Additionally, OXTR activation up-regulated the phosphorylation of protein kinase C (pPKC) and, in turn, impaired voltage-gated sodium currents, particularly the voltage-gated sodium channel 1.7 (Na V 1.7) current, that plays an indispensable role in PTX-induced neuropathic pain. OXT suppressed excitatory transmission in the spinal dorsal horn as well as excitatory inputs from primary afferents in PTX-treated mice. Conclusion and Implications: The OXTR in small-sized DRG neurons is up-regulated in CINP and its activation relieved CINP by inhibiting the neural excitability by impairment of Na V 1.7 currents via pPKC. Our results suggest that OXTR on peripheral sensory neurons is a potential therapeutic target to relieve CINP.


| INTRODUCTION
Chemotherapy-induced neuropathic pain (CINP) is a somatosensory dysfunction, which is characterized by a series of sensory symptoms, including neuropathic pain (Gutiérrez-Gutiérrez et al., 2010).
Paclitaxel (PTX), a commonly used chemotherapeutic agent, has toxic effects on dorsal root ganglion (DRG) neurons (Sisignano et al., 2014) and is widely used to establish neuropathic pain animal models . Most recently, researchers found dynamic changes in the expression of genes during neuropathic pain development (Kim et al., 2009;. Some changes contributed to the development of neuropathic pain, but others might relieve pain. It is well documented that central oxytocin (OXT) and the oxytocin receptor (OXTR) play roles in analgesia (Gamal-Eltrabily et al., 2020;Sun et al., 2018;Yang et al., 2007). However, the peripheral expression of OXTR in dorsal root ganglia (DRG) and the antinociceptive function of OXT remain controversial. In terms of the peripheral antinociceptive function, Yang et al. (2007) found that peripheral administration of OXT did not have antinociception in naïve animals. However, it was demonstrated that intraperitoneal administration of a OXTR agonist inhibited pain behaviour via OXTR activation in the complete Freund's adjuvant (CFA)-induced model (Hilfiger et al., 2020). It also was reported that peripheral local administration of OXT attenuated ipsilateral formalin-produced pain behaviour, whereas contralateral administration of OXT was ineffective (González-Hernández et al., 2017). These findings implied that OXT exhibits peripheral antinociceptive effects under pathological conditions, which may stem from the dynamic manner of OXTR expression in the peripheral nervous system. However, the underlying mechanisms of how OXTR modulates the activity of peripheral neurons has not yet been investigated, and the expression pattern of OXTR on DRG neurons in the CINP model also should be investigated.
The mechanism underlying paclitaxel (PTX) induced neuropathic pain is complicated, with multiple ion channels in DRG neurons involved in its mechanisms (Chang et al., 2018;Y. Li et al., 2017;Zhang & Dougherty, 2014). For example, PTX increased the expression of the voltage-gated sodium channel 1.7 (Na V 1.7) in smalldiameter (<50 μm) human DRG neurons (Chang et al., 2018). Similar results were reported by Y. Li et al. (2018), who showed that Na V 1.7 expression was increased in a CINP model as well as in human DRG neurons of patients who had experienced PTX-induced neuropathic What is already known • The central antinociceptive roles and mechanisms of oxytocin and its receptor (OXTR) are well documented.

What does this study add
• Expression of OXTR in the peripheral nervous system is up-regulated in a neuropathic pain model.
• Up-regulated OXTR in the peripheral nervous system is effective in reducing chemotherapy-induced neuropathic pain.

What is the clinical significance
• OXTR in peripheral sensory neurons is a potential antinociceptive target to relieve chemotherapy-induced neuropathic pain. pain symptoms. Although these studies have advanced our understanding of CINP, there is still no effective treatment for CINP in the clinic. Therefore, it has attracted considerable attention to find a safe and effective treatment for CINP.
In this study, we established a PTX-induced neuropathic pain model. Strikingly, we found that Oxtr mRNA and protein levels had apparently increased in DRG neurons after PTX treatment. Peripheral application of OXT relieved PTX-induced neuropathic pain via OXTR activation, but it did not change the pain thresholds in vehicle-treated mice. Using whole-cell patch-clamp methods, we found that OXTR activation inhibited the excitability of DRG neurons in PTX-treated mice, an effect which was mediated by an impairment of voltagegated sodium (Na V ) currents via the phosphorylation of protein kinase C (pPKC). Taken together, PTX up-regulated the expression of OXTR in DRG neurons, and this up-regulation can be exploited to relieve PTX-induced neuropathic pain.

| Animals
All experimental procedures were conducted by following the guidelines of the International Association for the Study of Pain, and were approved by the Institutional Animal Care and Use Committee of Southern University of Science and Technology and Shenzhen University followed the guidelines formulated by the European Community (2010/63/EU). C57Bl/6 mice (5-8 weeks of age) were purchased from Guangdong Province Laboratory Animal Center (Guangzhou, China). Animal studies are reported in compliance with the ARRIVE guidelines (Percie du Sert et al., 2020) and with the recommendations made by the British Journal of Pharmacology (Lilley et al., 2020). The animals were housed in standard cages (five per cage), in a temperature-controlled environment, and on a 12/12-h light/dark cycle. Food and water were available ad libitum.
In all experiments, animals were randomly assigned to treatment groups. Both males and females were included in each group in a sexmatched manner. The data from both sexes were combined and used equally throughout this study, because no differences due to sex were observed in preliminary experiments, as shown in Figure S1. The number of mice in each group was designated in each figure. The sample size in each group was determined according to our previous studies with similar experimental protocols Jiang et al., 2020;Muralidharan et al., 2021). All investigators were blinded to the treatment assignment and outcome assessment. Different experiments were conducted separately to avoid mutual influence.
For all the experiments with PTX treatment, such as electrophysiological or staining experiments, the establishment of neuropathic pain was confirmed by behavioural tests.
Overall, the mice were euthanized by CO 2 inhalation and death was confirmed by decapitation. For quantitative real-time polymerase chain reaction (qRT-PCR), fluorescence in situ hybridization (FISH) and staining, the mice were deeply anaesthetized with isoflurane and underwent fixation by transcardial perfusion. For primary cultures of DRG neurons or patch-clamp recording in spinal cord slices, mice were anaesthetized with pentobarbital sodium (80 mgÁkg À1 , i.p.). The above procedures are according to the ARRIVE Guidelines for the Euthanasia of Animals (2020 Edition) (Lilley et al., 2020

| PTX-induced neuropathic pain and reagents administration
Twenty-five milligrams of PTX powder was dissolved in 700 μL of dimethyl sulfoxide (DMSO). For the stock solution (6 mgÁmL À1 ), the dissolved PTX was transferred into 3.46 ml of ethanol and Cremophor EL mixed at a 1:1 ratio. The stock solution was diluted to the desired concentration (0.2 mgÁml À1 ) in saline before use. PTX was injected intraperitoneally on 4 alternate days (2.0 mgÁkg À1 on Days 1, 3, 5 and 7, with a final cumulative dose of 8.0 mgÁkg À1 ) to induce peripheral neuropathic pain, whereas the vehicle (DMSO, ethanol, Cremophor EL and saline mixed at a 1:2.5:2.5:172.4 ratio) was injected as the control.
The concentration of dOVT used in the current study refers to our previous and Pierre-Eric Juif's reports (Jiang et al., 2014;Juif & Poisbeau, 2013). The concentration of TCOT39 used refers to the EC 50 value for the OXTR given in Pitt's report (Pitt et al., 2004) and the EC 50 for the effect of OXT on neuronal OXTR (Jiang et al., 2014).

| Behaviour testing
The mice were habituated to their environment for 3 days before experiments. All testing was done with the experimenters blinded to the treatment conditions. Von Frey testing was performed to assess mechanical allodynia. Briefly, the mice were placed separately in a clear plastic box with a wire mesh floor (40 cm above the bench) and allowed to habituate to the environments for 60 min before testing.
The hind paw was stimulated with a series of von Frey hairs with logarithmically increasing stiffness (0.02-2.56 g, Stoelting), presented perpendicularly to the central plantar surface. The 50% paw-withdrawal threshold was determined by Dixon's up-down method (Sun et al., 2018;. The hot plate test (Ugo Basile, Italy) was used to examine thermal hyperalgesia. Each mouse was placed on the hot plate (53 C), and the latency of paw withdrawal was measured when the animal began to exhibit signs of pain avoidance, such as jumping or paw licking, which was measured twice, separated by a 5-min interval. The cut-off time was set at 40 s to avoid tissue damage.

| Fluorescence in situ hybridization
The mice were deeply anaesthetized with isoflurane and perfused with phosphate-buffered saline (PBS), followed by 4% paraformaldehyde (PFA). After perfusion, the DRGs were removed and fixed in the same fixative for 2 h at 4 C. Then, the DRGs were cryopreserved in a 30% sucrose/PBS solution for 2 days. DRG sections (12 μm) were cut using a cryostat (Leica CM1520). FISH was performed using the RNAscope system (Advanced Cell Diagnostics).
Pre-treatment consisted of dehydration, followed by incubation with hydrogen peroxide and protease IV at room temperature. Subsequently, the protocol for the Multiplex Fluorescent Kit v2 was followed using commercial probes for Oxtr (Mm-Oxtr, #402651). FISH images were taken using a Nikon fluorescence microscope.
For quantification, all the images were obtained using the same acquisition settings, four to five DRG sections from each animal were selected and five animals were included for data analysis. To determine the percentage of labelled cells, we quantified the number of fluorescence puncta using the bioimage analysis software QuPath (Z. Wang et al., 2020). The size distribution was analysed using the following criteria: small (<300 μm 2 ), medium (300-700 μm 2 ) and large (>700 μm 2 ) (Scherrer et al., 2010). For visualizing neurons, Nissl (1:200, Invitrogen, Cat# N21483) was used over a 2-h incubation.

| Immunocytochemistry
For OXTR staining, the mice were deeply anaesthetized with isoflurane and perfused with PBS. To observe the effects of the activation of OXTR on the pPKC in DRG neurons, fresh DRGs were removed from anaesthetized mice from different groups. Thereafter, mice were killed by decapitation. The fresh DRGs were moved to dishes immediately and incubated with either PBS or OXT for 30 min.
Subsequently, the DRGs were post-fixed in 4% PFA overnight.
Then the DRGs were cryopreserved in 30% sucrose/PBS solution overnight. DRG sections (12 μm) were cut using a cryostat (Leica

| RNA purification, cDNA synthesis and qRT-PCR
The mice were deeply anaesthetized with isoflurane and perfused with PBS. Excised DRGs were immediately snap-frozen after collection.
Total RNA from DRG was isolated using the MolPure ® Cell/Tissue Total RNA Kit (Yepsen Biotechnology) according to the manufacturer's instructions. For the RT reaction, first-strand cDNA was generated using Hifair ® III Reverse Transcriptase (Yepsen Biotechnology), oligo-(dT) primers and dNTPs (Sangon Biotech, Shanghai, Republic of China).
In brief, 5 μg of total RNA was reverse transcribed to cDNA. qRT-PCR was performed using an ABI 7500 real-time PCR thermal cycler (Applied Biosystems, Thermo Fisher Scientific) with Heff UNICON ® Universal Blue qPCR SYBR Green Master Mix (Yepsen Biotechnology) according to the manufacturer's instructions. Target mRNA levels were normalized to the mRNA level of glyceraldehyde-3-phosphate dehydrogenase (Gapdh) and calculated using the 2 ÀΔΔCT value method.

TGTAGACCATGTAGTTGAGGTCA.
Lumbar DRGs (L3-L5) were removed aseptically from mice that were deeply anaesthetized with pentobarbital sodium ( To evaluate the excitability of recorded neurons, the current clamp mode was used to record action potentials (APs) and determine the rheobase. Pipette resistance was 4-6 MΩ for whole-cell recordings, and pipettes were filled with an internal solution containing (in mM) 126 K-gluconate, 10 NaCl, 1 MgCl 2 , 10 EGTA, 10 HEPES and 2 Na-ATP (adjusted to pH 7.4 with KOH). The external solution contained (in mM) 140 NaCl, 5 KCl, 2 CaCl 2 , 1 MgCl 2 , 10 HEPES and 10 glucose, adjusted to pH 7.4 with NaOH. A series of APs was triggered by current injection steps from 0 to 130 pA, with increments of 10 pA, over 600 ms. To observe the waveform of AP, a single AP was recorded by current injection steps from 0 to 130 pA, with increments of 10 pA, over 30 ms.
Na V 1.7-mediated currents were isolated by subtraction of the ProTx II-resistant Na + currents from Na + total currents (Y. Li et al., 2018).
ProTx II was perfused for 5 min. For quantification, six to seven DRG neurons from five mice per treatment group were selected for data analysis.
In brief, the stimulation was performed using a suction electrode with a constant current source for the pulse at a frequency of 0.1 Hz (Ataka et al., 2000;Jiang et al., 2014). were included in the figure legends. The criterion for statistical significance was defined as P < 0.05. All statistical analyses were performed using Prism GraphPad 8.0. The data and statistical analysis comply with the recommendations of the British Journal of Pharmacology on experimental design and analysis in pharmacology (Curtis et al., 2015(Curtis et al., , 2022 3 | RESULTS

| The expression of Oxtr mRNA and OXTR in DRG neurons increased after PTX treatment
To examine the OXTR expression in DRG neurons, we first determined its mRNA level after intraperitoneal administrations of PTX.
The qRT-PCR analysis in Figure 1a

| Intravenous injection of OXT exhibited antinociceptive effects via activation of OXTR after PTX treatment
To examine whether OXTR expression in DRG neurons mediates peripheral analgesia of OXT, we evaluated peripheral antinociceptive effects of OXT by intravenous injection, because OXT has limited ability to cross the blood-brain barrier (BBB) (Ermisch et al., 1985). The PTX-treated mice were given a single dose of OXT (0.12 mgÁkg À1 , i.v.) or its vehicle (Figure 3a,d). OXT showed a considerable increase in both the mechanical withdrawal threshold (Figure 3b, P < 0.05, two-way ANOVA, F (4, 56) = 12.23) and thermal withdrawal latency ( Figure 3c, P < 0.05, two-way ANOVA, F (4, 56) = 87.98). The duration of analgesia was sustained for 40 min and disappeared at 60 min after injection (Figure 3b,c). There were no differences due to sex observed, as shown in Figure S1. 3.4 | Multiple PTX injections increased Na + currents in small-sized DRG neurons, and OXT reduced the Na + currents, especially Na V 1.7 currents Na V channels serve as the basis for the generation and conduction of APs. Thus, we examined the effect of OXT/OXTR on Na V channels in cultured small neurons using whole-cell patch clamp recordings. Our results indicated that the Na + current amplitude in the PTX group was greater than that in the vehicle-treated group (Figure 6a,b, P < 0.05, unpaired t test, t = 5.595). As shown in Figure 6c,d, OXT perfusion (0.5 μM, 2 min) significantly diminished the Na + current amplitude recorded from small DRG neurons in the PTX group (P < 0.05, twoway ANOVA, F (10, 120) = 6.976).
Given that PTX enhances Na V 1.7 currents in DRG neurons (Y. Li et al., 2018), we tested the effect of OXT on Na V 1.7 currents in cultured DRG neurons from PTX-treated mice. The total Na + currents were recorded firstly ( Figure 6e, the red line), and then ProTx II (10 nM, 5 min) was applied to selectively inhibit the Na V 1.7 channels in the same recorded neuron. The blue trace in Figure 6e represents Na + currents after ProTx II perfusion. Na V 1.7-mediated currents were isolated by subtraction of the ProTx II-resistant Na + currents from total Na + currents. As shown in Figure 6e,f, OXT significantly reduced the Na V 1.7 current amplitudes in cultured DRG neurons from PTX-treated mice (Figure 6f, P < 0.05, unpaired t test, t = 3.596).

| The effects of OXTR activation on Na + currents and nociceptive behaviour in PTX-treated mice were mediated by pPKC
Since OXTR is a G-protein coupled receptor, we investigated the intracellular mechanisms of how OXTR affects the Na + current in nociceptors. First, fresh DRGs were removed from PTX-or vehicletreated mice and incubated with OXT for 30 min (Figure 7a). We found that the incubation of OXT significantly elevated pPKC in DRG neurons from PTX-injected mice (Figure 7b,c,f, P < 0.05, one-way performed an electrophysiological test to observe whether or not pPKC mediated the effect of OXT on Na + currents. We found that the effect of OXT on the Na + currents disappeared in the presence of the PKC antagonist chelerythrine (10 μM; Figure 7g, P > 0.05, unpaired t test, t = 1.149). In line with these findings, the effect of F I G U R E 4 The activation of oxytocin receptor (OXTR) mediated the antinociceptive effects of intravenous injection of oxytocin (OXT) on paclitaxel (PTX)-induced neuropathic pain. (a) The schematics of the experimental design for the PTX, OXTR agonist and behaviour experiments. On Day 0, the pain behaviours were tested, followed by PTX injection on Days 1, 3, 5 and 7. On Day 10, the pain behaviours were monitored before intravenous injection. Subsequently, the pain behaviours were measured at 20, 40 and 60 min after intravenous injection. (b, c) The antinociceptive effects of intravenous injection of OXTR agonist ([Thr 4 , Gly 7 ]oxytocin [TGOT]) or saline on the pawwithdrawal threshold (b) and latency (c) in PTX-treated mice (*P < 0.05, TGOT vs. saline, two-way analysis of variance [ANOVA], F (4, 40) = 6.750 and F (4, 40) = 39.76, respectively, n = 6 mice per group). (d) The schematics of the experimental design for the PTX, OXTR or vasopressin receptor 1a antagonist, and OXT injection and behaviour experiments. On Day 0, the pain behaviours were tested, followed by PTX injection on Days 1, 3, 5 and 7. On Day 10, the pain behaviours were monitored before intravenous injection. Subsequently, the pain behaviours were measured at 20, 40 and 60 min after intravenous injection. (e, f) The antinociceptive effects of intravenous injection of OXT or saline on the paw-withdrawal threshold (e) and latency (f) after the pre-treatment of OXTR antagonist ([d(CH2)5 1 , Tyr(Me) 2 , Thr 4 , Orn 8 , des-Gly-NH2 9 ]-vasotocin [dOVT]) in PTX-treated mice (not significantly different [n.s.], two-way ANOVA, F (4, 56) = 1.406 and F (4, 56) = 1.016, respectively, n = 8 mice per group). (g, h) The antinociceptive effects of intravenous injection of OXT or saline on the pawwithdrawal threshold (g) and latency (h) after the pre-treatment of vasopressin receptor 1a antagonist (SR49059) in PTX-treated mice (n.s., two-way ANOVA, F (4, 40) = 0.2584 and F (4, 40) = 0.3140, respectively, n = 6 mice per group). Data are presented as mean ± SEM; *P < 0.05, significantly different as indicated.
OXT on alleviating mechanical allodynia induced by PTX was inhibited by chelerythrine pre-treatment as well (Figure 7h, P < 0.05, two-way ANOVA, F (4, 40) = 11.13). These data suggest that the pPKC plays a critical role in OXTR peripheral analgesia.
3.6 | OXT decreased synaptic excitatory transmission in spinal lamina II from PTX-treated mice Given that OXT impaired DRG neurons excitability in the PTX-treated mice, we investigated the excitatory transmission in spinal lamina II, which is a key area for nociceptive transmission involving primary afferents. As shown in Figure 8a, the perfusion of OXT (0.5 μM, 2 min) decreased spontaneous excitatory transmission in spinal lamina II. The frequency of the sEPSC decreased gradually over time after OXT perfusion, peaking at 1 min (Figure 8a,b, P < 0.05, paired t test, t = 8.576), and this reduction was accompanied by an impairment in the sEPSC amplitude (Figure 8a,b, P < 0.05, paired t test, t = 7.986).
Next, we examined the effects of OXT on monosynaptic Aδ-fibre and C-fibre EPSCs in spinal lamina II. As shown in Figure 8c

| DISCUSSION
Accumulating evidence suggests that OXT and OXTR exert an antinociceptive action in the brain (Dumais & Veenema, 2016;Etehadi Moghadam et al., 2018). Although central nervous system (CNS)specific delivery routes are available, these options are invasive for human subjects. Recent evidence indicates that OXT exerts an antinociceptive action not only in the CNS but also in the periphery (González-Hernández et al., 2014). Because OXT has limited ability to cross the BBB (Ermisch et al., 1985), the peripheral and central mecha-   F (26, 195) = 4.937, respectively, n = 6 recordings per group, from 5 mice). (c, d) Representative current-evoked AP traces in a small DRG neuron from a vehicle-treated mouse on Day 10 before (c, black) and after OXT (0.5 μM, 2 min, c, orange) perfusion, and the quantification of the firing rate of AP (d, not significantly different [n.s.], two-way ANOVA, F (13, 130) = 0.3925, n = 6 recordings per group, from 5 mice).
First, we found using FISH technology that Oxtr mRNA was rarely detected in the DRG neurons of mice in the vehicle-treated group.
This finding is consistent with Lopes's finding that Oxtr mRNA was faintly present in mouse DRG neurons when using the conventional RNA-sequence method (Lopes et al., 2017). Interestingly, we found that the Oxtr expression level in DRG was gradually elevated after In this study, OXT was administrated intravenously and only exhibited antinociceptive effects in PTX-treated mice. We did not find sexual dimorphism in this study, even though it was reported that the antinociceptive effects of OXT exhibit a sex difference in either lipopolysaccharide (LPS)-or carrageenan-induced pain models (Chow et al., 2016;Salinas-Abarca et al., 2022). It should be noted that the effects of OXT were observed over a short time after LPS or carrageenan treatment, and the difference may be due to the activation of spinal glia or a higher expression of insulin-regulated aminopeptidase (IRAP) in the spinal cord of female rats. In this study, we observed that OXT produced antinociception when the neuropathic pain was well established, and the peripheral neuronal mechanisms were found. This might be the reason why we did not observe sex difference in the current study. In agreement with our findings that OXT (intravenously) did not change the pain threshold in vehicle-treated mice, Hilfiger's group and Yang's group did not observe peripheral OXTR agonist or OXT-mediated analgesia in the physiological status (Hilfiger et al., 2020;Yang et al., 2007). Huang et al. (2021) used local microinjection of OXT at the ganglion and found that it inhibited nociceptive activity after nerve injury. The above results imply that the peripheral OXT antinociceptive effect may be mediated by the rise and activation of OXTR after PTX treatment. Therefore, we investigated the F I G U R E 6 Effect of oxytocin (OXT) on Na + currents in small-sized dorsal root ganglion (DRG) neurons in paclitaxel (PTX)-treated mice.
(a) Representative traces of total Na + currents in small DRG neurons from a vehicle-(black) or PTX-treated (red) mouse on Day 10 after the first injection. (b) The quantification of the amplitude of Na + currents from the saline and PTX-treated mice (*P < 0.05, unpaired t test, t = 5.595, n = 7, 6 recordings, respectively, from 5 mice). (c) Representative traces of total Na + currents before OXT application (red), in the presence of OXT (blue) and washout (grey) in a small-sized DRG neuron from a PTX-treated mouse on Day 10 after the first injection. (d) Time-course of normalized Na + currents with or without OXT perfusion (0.5 μM, 2 min, *P < 0.05, $ P < 0.05, two-way analysis of variance, F (10, 120) = 6.976, n = 7 recordings per group, from 5 mice). (e) Representative traces of Na + current with (blue)/without (red) ProToxin II (ProTx II) (10 nM, 5 min) in the absence of OXT (left) or in the presence of OXT (0.5 μM, 2 min, right) in small-sized DRG neurons from PTX-treated mice on Day 10 after the first injection. Voltage-gated sodium channel 1.7 (Na v 1.7) currents were obtained by subtraction of recordings of blue from that of red in (e).
(f) Analysed data of the amplitudes of Na v 1.7 currents in OXTÀ and OXT+ groups from (e) (*P < 0.05, unpaired t test, t = 3.596, n = 6 recordings per group, from 5 mice). Data are presented as mean ± SEM; *P < 0.05 and $ P < 0.05, significantly different as indicated.
analgesia of OXT against its antagonist. We found that dOVT blocked OXT-induced analgesia, and its agonists mimicked its effects in PTXtreated mice. However, there is a report that indicated that OXT- Although it is reported that intranasal administration or intraperitoneal injection of OXT (12 μg) caused an increase in OXT concentration in cerebrospinal fluid (CSF) (Kou et al., 2021;Neumann et al., 2013), the ability of OXT to penetrate BBB is thought to be limited. For example, it is reported that only 0.002% of the peripherally applied OXT reaches the CNS (Mens et al., 1983). Additionally, the influence of systemic application of OXT on the concentration of OXT in CSF or plasma occurs in a time-dependent manner. Yamamoto et al. (2019) found that steady-state levels in blood increased rapidly, peaking within 10 min after subcutaneous injection of OXT and returning to baseline by 1-2 h, but OXT in CSF increased gradually to maxima at 1-2 h before returning to baseline at 4 h. In the current study, OXT was applied at 0.12 mgÁkg À1 , which suggests that only 0.06-ng OXT reached the CSF according to Greidanus's report (Mens et al., 1983 Subsequently, we tried to reveal the underlying mechanisms of OXT antinociception. We found that the excitability of small-sized DRG neurons increased after PTX treatment, which was consistent with previous findings (H. Luo et al., 2019), and that OXT decreased the firing frequency of APs. However, OXT did not alter the excitability of small DRG neurons in vehicle-treated mice. It is known that sodium channels are essential in the production of electrical activity in neurons. Therefore, sodium currents were recorded in small DRG neurons from different groups. We found that sodium currents in PTX-treated mice were greater in amplitude than that in control mice, and OXT reduced these sodium currents. Recently, different groups have proven that the expression and function of Na V 1.7 are upregulated in PTX-treated rodents or in humans during neuropathic pain (Chang et al., 2018;Y. Li et al., 2018). We therefore tested the effect of OXT on Na V 1.7 currents. It is interesting that OXT reduced Na V 1.7 currents in small DRG neurons from PTX-treated mice.
Because Na V .7 defines the AP threshold and contributes to its upstroke (Meents et al., 2019), we compared the waveform of AP before and after OXT perfusion. We found that OXT extended the duration to AP threshold. Because mutations of Na V 1.7 led to congenital insensitivity to pain (Cox et al., 2006;Goldberg et al., 2007), our findings reveal a promising potential therapeutic strategy for CINP.
Indeed, that OXT exerts its effect on pain may have multiple mechanisms. For example, in rats with spinal nerve ligation, OXT produces analgesia by an inhibition of the increase in intracellular calcium during membrane depolarization (Hobo et al., 2012). In the spinal dorsal horn, OXT induces an increase in the release of γ-aminobutyric acid (GABA) and glycine (Breton et al., 2008;Jiang et al., 2014), which is due to a change in membrane permeabilities to K + and/or Na + in the partial interneurons (Jiang et al., 2014). Nevertheless, further studies are needed to elucidate the peripheral antinociceptive mechanism of OXT.
Next, we investigated the intracellular mechanisms of how OXTR affects the Na + currents after PTX treatments. It is reported that PKC is required for OXTR-mediated neuronal responses in the central amygdala (Hu et al., 2020), and the activity of PKC is dependent on its phosphorylation (Aslam & Alvi, 2022). Accordingly, our F I G U R E 7 The phosphorylation of protein kinase C (pPKC) in dorsal root ganglion (DRG) neurons after oxytocin receptor activation and the effect of PKC inhibitor on the action of oxytocin (OXT) on Na + currents and OXT-induced analgesia in paclitaxel (PTX)-treated mice. ]-vasotocin (dOVT) (d) or co-incubation of OXT and dOVT (e) in PTX-injected mice. (f) Quantitative analysis shows the percentage of pPKCpositive neurons in DRGs following PBS or OXT incubation in the vehicle or PTX-injected mice (*P < 0.05, one-way analysis of variance [ANOVA], F (5, 27) = 10.93, n = 5, 6, 6, 6, 5, 5 mice, respectively) and following co-incubation of PBS and dOVT or co-incubation of OXT and dOVT in PTX-injected mice (*P < 0.05, one-way ANOVA, F (5, 27) = 10.93, n = 5, 6, 6, 6, 5, 5 mice, respectively). (g) Representative traces (left) and the quantifications (right) of the amplitude of Na + currents without OXT perfusion (black) or with OXT perfusion (0.5 μM, 2 min, red) in the presence of the PKC inhibitor (chelerythrine, 10 μM) in small-sized DRG neurons from PTX-treated mice on Day 10 after the first injection (not significantly different [n.s.], paired two-tailed t test, t = 1.149, n = 6 per group, from 5 mice). (h) The antinociceptive effects of intravenous injection of OXT on the paw-withdrawal threshold in PTX-treated mice under the pre-treatment of either chelerythrine or saline (*P < 0.05, twoway ANOVA, F (4, 40) = 11.13, n = 6 mice per group). Scale bar: 20 μm. Data are presented as mean ± SEM; *P < 0.05, significantly different as indicated.
results showed that OXT elevated the level of pPKC in DRG neurons in PTX-injected mice, via the activation of OXTR. Additionally, the inhibitory effect of OXT on Na + currents was blocked by the presence of a PKC inhibitor. These findings indicated that the activation of OXTR is induced the pPKC, which in turn inhibits the Na + currents in DRG neurons. To verify whether pPKC participates in OXT/OXTR-mediated peripheral analgesia, we tested the analgesia of OXT by pre-treatment of the PKC inhibitor chelerythrine in PTXtreated mice. As expected, we found that the pPKC inhibitor blocked OXT/OXTR analgesia.
In the spinal dorsal horn, it is impressive that OXT attenuated not only the glutamatergic spontaneous transmission but also evoked excitatory transmission in the PTX-treated mice. However, OXT did not change the excitatory transmission in vehicle-treated mice, which is consistent with reports in naïve rats (Jiang et al., 2014). This suggests that OXT attenuates nociceptive transmission only in PTXtreated mice.

| CONCLUSIONS
In this study, we found that the chemotherapeutic agent PTX upregulated OXTR expression in peripheral somatosensory neurons without gender differences. In addition, activation of OXTR relieved F I G U R E 8 Effect of oxytocin (OXT) on spontaneous and evoked excitatory postsynaptic currents (EPSCs) in spinal dorsal horn from paclitaxel (PTX)-treated mice. (a, g) Chart recording showing spontaneous EPSCs in the absence and presence of OXT (0.5 μM, 2 min) obtained from a PTXtreated mouse (a) and vehicle-treated mouse (g) on Day 10 after the first injection, respectively. The duration of drug perfusion is shown by a horizontal bar above the chart recording and three consecutive traces of spontaneous events for a period indicated by a short bar located below the chart recording are shown in an expanded scale in time. (b, h) The average of spontaneous EPSC frequency (right) and amplitude (right) before (control) and 2 min after the beginning of OXT perfusion respectively in the PTX-treated mice (b, *P < 0.05, paired two-tailed t test, t = 8.576 and t = 7.986, respectively, n = 6 recordings per group, from 5 mice) and the vehicle-treated mice (h, not significantly different [n.s.], paired twotailed t test, t = 0.1055 and t = 0.7986, respectively, n = 6 recordings per group, from 5 mice). (c, i) Recordings of monosynaptically evoked Aδfibre EPSCs before (control) and 2 min after the beginning of OXT perfusion respectively in a PTX-treated mouse (c) and vehicle-treated mouse (i) on Day 10 after the first injection. (d, j) The peak amplitudes of the monosynaptically evoked Aδ-fibre in the absence (control) and presence of OXT respectively in the PTX-treated mice (d, 0.5 μM, 2 min, *P < 0.05, paired two-tailed t test, t = 10.20, n = 5 recordings per group, from 5 mice) and the vehicle-treated mice (j, n.s., paired two-tailed t test, t = 0.7986, n = 6 recordings per group, from 5 mice). (e, k) Recordings of monosynaptically evoked C-fibre EPSCs before (control) and 2 min after the beginning of OXT perfusion respectively in a PTX-treated mouse (e) and vehicle-treated mouse (k) on Day 10 after the first injection. (f, l) The peak amplitudes of the monosynaptically evoked C-fibre EPSCs in the absence (control) and presence of OXT respectively in the PTX-treated mice (f, *P < 0.05, paired two-tailed t test, t = 9.554, n = 5 recordings per group, from 5 mice) and the vehicle-treated mice (l, n.s., paired two-tailed t test, t = 1.359, n = 6 recordings per group, from 5 mice). Data are presented as mean ± SEM; *P < 0.05, significantly different as indicated.
PTX-induced neuropathic pain by inhibiting the neural excitability, which was mediated by an impairment in Na + currents, especially Na V 1.7, via pPKC. Taken together, OXTR on peripheral sensory neurons may be a potential therapeutic target for the relief of chemotherapy-induced peripheral neuropathic pain.