2.1 Animals

Male and female C57BL/6 mice (5–6 weeks and 12 months, weighing 18–30 g) were purchased from the Guangdong Medical Laboratory Animal Centre in Guangzhou, China. All animal experiments were approved by The Sun Yat-sen University Cancer Centre Animal Care and Use Committee (Sun Yat-sen University No. L025501202204003). Before performing any experiments, all the mice were quarantined for 7 days. All mice were housed under controlled environmental conditions (temperature: 24 ± 1°C, humidity: 50%–60%) and had a 12-h light/dark cycle with water and food available ad libitum.

2.2 Treatment of animals

To induce PIPN pain, male and female C57BL/6 mice were i.p. administered PTX diluted in saline (NaCl 0.9%) at a dose of 2 mg/kg every other day for a total of 4 injections (days 0, 2, 4, and 6).^34-36 The control group was i.p. administered the same volume of physiological saline. Von Frey and hot plate tests were performed on days 1, 3, 5, 7, 14 and 21 after injection to confirm the successful establishment of the PIPN models.

For i.t. injection of the PD-L1 protein (diluted in saline, 10 ng or sterile water) on day 14 after PTX injection, spinal cord puncture between the L5 and L6 levels was performed with a Hamilton syringe equipped with a 30-gauge needle to deliver reagents.³⁷ The accuracy of the i.t. injection was confirmed by observing the tail flick response in the mice.

2.3 Intrathecal injection of AAV virus

To functionally determine the role of TRPV1 in the sex-dependent analgesic effect of PD-L1 in PIPN, we used RNA interference driven by AAV9-Trpv1-RNAi to downregulate TRPV1 levels in the spinal cord. To knock down TRPV1, we selected the siRNA sequences GGTACTGTACTTCAGCCATCG and GCAACTGTGAGGGC GTCAAGC to construct AAV viral vectors (Gene, China). After the mice were anesthetized with pentobarbital, a 30-gauge needle connected to a Hamilton syringe was introduced into the L5- and L6-level space. Sudden tail movement indicated successful delivery of the catheter. Subsequently, the virus solution (5 μL, 1 × 10¹³ vg/mL) was injected slowly, and the animals were restricted from movement for a minimum of 1 min following injection to prevent drug efflux. Four weeks after AAV9 delivery, the downregulation of TRPV1 expression was verified by immunofluorescence staining and Western blotting.

2.4 Measurement of mechanical hyperalgesia using the Von-Frey test

Prior to commencing the formal experiment, we acclimated individual mice under glass on an elevated wire mesh grid for 1 h per day for a total of 3 days to facilitate adaptation to the experimental environment. Subsequently, we employed a series of Von Frey hairs with logarithmically increasing stiffness (ranging from 0.04 to 2.0 g) to evaluate the paw withdrawal threshold. A positive response was defined as either hind paw licking or rapid withdrawal. Once the back paw has shown a positive response in at least 3 out of 5 measurements at 5-min intervals for both the left and right hind paws, the fiber value in grams is considered PWT.^{24, 38}

2.5 Measurement of thermal sensitivity using hot plate tests

Thermal sensitivity (paw withdrawal latency) was determined using a hot plate set (ZH-6C, ANHUI Zenna Biologenic Apprax FACILITIES, China) at 55°C 2 h after the von Frey test. Before the formal experiments, mice were subjected to an adaptive test for 3 days, as previously described.³⁹ Paw licking, paw contraction, scratching or four-foot hopping were considered positive responses. The endpoint was this reaction time, and an increase in hot plate delay was used to determine the analgesic activity. In order to prevent tissue injury, mice were promptly removed from the hot plate if they did not respond within 30 s. The thermal sensitivity test was conducted three times at 5-min intervals, and the average value was calculated.

2.6 Western blotting

After deep anesthesia, L4-L5 spinal dorsal horn tissue segments were collected from PIPN mice and control mice and subsequently frozen at −80°C until required for further use. The proteins were lysed with RIPA buffer, and the concentrations of the lysates were measured via a BCA protein assay as described previously.³⁷ Equal amounts of proteins were separated via 10% or 12.5% SDS–PAGE and transferred to polyvinylidene fluoride (PVDF) membranes. The membranes were blocked with Tris-buffered saline containing Tween-20 (TBST) and containing 5% nonfat dry milk (FUJIFILM Wako, Japan) or BSA (Epizyme Biotech, China) for 1 h at room temperature and then incubated with primary antibodies against GAPDH (1:10,000), PD-L1 (1:500-1:1000), TRPV1 (1:1000), and CGRP (1:500-1:1000) overnight at 4°C. On the second day, the membranes were washed and incubated with species-matched secondary antibodies at room temperature for 1 h. The protein bands were visualized using Millipore's enhanced chemiluminescence reagent. All Western blot experiments were conducted in triplicate, and the integrated optical density of the bands was quantified using a Bio-Rad image analysis system (Bio-Rad Laboratories, USA) and ImageJ software.

2.7 Quantitative RT–PCR

The L4–L5 segments of spinal dorsal horn tissue were collected, and RNA was extracted using a Tissue RNA Purification Kit Plus (ES Science, China) according to the manufacturer's instructions. cDNA synthesis was performed using an EZBioscience Color Reverse Transcription Kit (Roseville, USA). RT–PCR was performed with SYBR Green qPCR Super Mix (EZBioscience, Roseville, USA) on a CFX96 Touch Real-Time PCR System (Bio-Rad, USA), and the results were calculated by the comparative threshold cycle (Ct) method. Assays were conducted in technical replicates, and the results were normalized to mouse GAPDH as a reference standard.

2.8 Immunofluorescence staining

The mice were anesthetized via i.p. injection of sodium pentobarbital (100 mg/kg) and subsequently perfused through the heart with PBS followed by 4% cold paraformaldehyde. The segments were subsequently immersed in a 30% sucrose solution at a temperature of 4°C until they sunk to the bottom of the container following fixation. The samples were sectioned into 25 μm thick sections on a cryostat and fixed on gelatin-coated glass slides as described previously.⁴⁰ The sections were blocked with Immunol Staining Blocking Buffer (Beyotime, China) at room temperature for 1 h and subsequently incubated overnight with rabbit anti-PD-L1 (1:100–1:250) and mouse anti-TRPV1 (1:100–1:200) primary antibodies mixed with anti-NeuN (1:200) at 4°C. After washing, the sections were incubated with FITC-Alexa Fluor 488- and Cy3-conjugated secondary antibodies (1:500–1:1000) at room temperature for 1 h. Sections were viewed using a confocal laser scanning microscope (LSM980, Zeiss, Germany). The data were analyzed with ImageJ software.

2.9 Hematoxylin–eosin staining

The mice were euthanized at a designated time. The hearts, livers, spleens, lungs and kidneys were collected, fixed with para-formaldehyde, and then stained with hematoxylin and eosin (H&E).

Antibodies and oligos used for Western blot, immunofluorescence, RT–PCR, and AAV–RNAi assays were showed in Table 1.

2.10 Statistical analysis

For all experiments, the number of biological replicates (n), measure of central tendency (e.g. average), error bars, and statistical analysis has been explained in the Figure legends. For each experiment where statistics were computed, we used at least n = 3 or more biological replicates. All statistically significant comparisons are detailed in the Figures and corresponding legends. Data are presented as mean ± SEMs after testing for normality by Shapiro–Wilk test using GraphPad Prism software (version 9, GraphPad Software, San Diego, CA, USA). Unpaired two-tailed Student t test is used to compare the means of 2 groups. To compare multiple groups, comparisons of the means are performed using one-way or two-way ANOVA followed by the Bonferroni post hoc correction to determine statistically significant differences. A p value of <0.05 was considered to indicate statistical significance.

3.1 Greater PTX-induced nociceptive behavior was observed in female mice than in male mice

To investigate whether the effects of PTX differed based on the sex of the mice, we first established a PIPN mouse model via four i.p. injections of 2 mg/kg PTX once a day. Behavioral tests of mechanical hyperalgesia and thermal sensitivity were performed on days 1, 3, 5, 7, 14 and 21 after injection (schematic diagram shown in Figure 1A). The body weights of 5-6-week-old male and female C57/6J mice used for PIPN model establishment significantly differed; however, there was no statistical difference between the weights of control and PIPN mice within the male or female groups (Figure 1B). Prior to PTX injection, there was no statistically significant difference in mechanical hyperalgesia and thermal sensitivity in control and PIPN group mice of both sexes (n = 8–10, 5–6 weeks; Figure 1C,D). The paw withdrawal threshold (PWT) (Figure 1E) and paw withdrawal latency (PWL) (Figure 1F) decreased more in female mice than male mice and both significantly differed from those observed in the control group on days 7 and 14. Moreover, PWT and PWL recovered gradually after 21 days.

To further explore whether sex differences in PTX-induced pain-related behaviors were age-related, we next established young and elderly PIPN mouse models via the same protocol as above (schematic diagram shown in Figure 2A). The body weights of male and female C57/6J mice aged 5–6 weeks used to establish the young PIPN mouse mode significantly different, and statistical differences in the weights of young and elderly mice was observed in both the male and female groups were observed (Figure 2B). Before modeling, there was no statistically significant difference in the PWT or PWL between the young and elderly groups of either sex (n = 8–10, 5–6 weeks vs. 12 months; Figure 2C,D). The behavioral experiment result revealed that the PWT (Figure 2E) and PWL (Figure 2F) also decreased more in young or elderly female mice than in the corresponding male mice on days 7 and 14. However, there was no difference between young and elderly female or male mice. Consistent with previous results, all four groups recovered gradually after 21 days. Taken together, these findings provide strong evidence that PTX induced sex-difference pain-related behaviors in mice and we explored the underlying mechanism further.

3.2 The analgesic effect of PD-L1 in PIPN mice was sexually dimorphic

To assess the potential analgesic effect of PD-L1 in male and female PIPN mice, we injected 10 ng of PD-L1 into the spinal cord of PIPN mice on day 14. The details are shown in Figure 3A. To further evaluate the optimal time of assessing the PD-L1 effect, we measured the PWT (Figure 3B) and PWL (Figure 3E) in mice of both sexes before injection and at 1, 3, and 12 h after PD-L1 injection. As expected, we found that the analgesic effect of PD-L1 was optimal 3 h after injection. Next, we focused on calculating mechanical hyperalgesia and thermal sensitivity at 3 h after injection. The rescue effect of PD-L1 on PWT was significantly greater in female mice than in male mice (Figure 3C), and the calculated relative therapeutic effect was almost 2.717-fold higher in the female PIPN group than in the male PIPN group (Figure 3D). Simultaneously, PD-L1 had the same therapeutic effect on PWL in mice, and the effect was also significantly greater in female mice than in male mice (Figure 3F). The relative therapeutic effect on the PWL was approximately 2.303-fold higher in the female PIPN group than in the male PIPN group (Figure 3G). Notably, we observed a robust increase in the protein (Figure 3H,I) and mRNA (Figure 3J) levels of PD-L1 after injection into the spinal cords of both sexes. Consistently, immunofluorescence staining also showed that increased expression of PD-L1 in the spinal cords of PIPN mice of both sexes (Figure 3K). These results demonstrated that the involvement of PD-L1 in PIPN mice was sexually dimorphic. In order to assess the potential toxic effects on mice in each experimental group, we conducted continuous measurements of body weight from pre-experiment baseline through 14 days after PTX injection and PD-L1-treatment. Our findings indicated that there were no significant differences in body weight among the PIPN and PD-L1 groups compared to the control group within their respective genders (Figure S1A). The histological analysis of the heart, liver, spleen, lung and kidney revealed no apparent pathological changes following treatment compared to both sexes control groups respectively (Figure S1B). This further indicated that the experimental procedure did not induce acute toxicity or tissue damage and there was no discernible difference between males and females.

3.3 Role of TRPV1 and CGRP in the analgesic effect of PD-L1 in both sexes of PIPN mice

To understand how PD-L1 plays an analgesic role, we investigated whether certain proteins related to pain are affected by PD-L1 upregulation in PIPN mice. To this end, we examined the expression of TRPV1 (which plays an important role in the PIPN mouse DRG⁴¹) in the spinal cords of the mice in each group by Western blot analysis. Treatment with PTX increased the levels of TRPV1 and CGRP in the spinal cords of mice of both sexes (Figure 4A, lane 3 vs. 1, lane 4 vs. 2; Figure 4B). Importantly, we found that the basal expression levels of TRPV1 and CGRP were significantly greater in the female spinal cord than in the male spinal cord (Figure 4A, lane 2 vs. 1; Figure 4B). The addition of PD-L1 for 3 h partly blocked the PTX-induced upregulation of TRPV1 and CGRP at the protein level in the spinal cords of mice of both sexes (Figure 4A, lane 5 vs. 3; lane 6 vs. 4) as well as at the mRNA levels (TRPV1 [Figure 4E] and CGRP [Figure 4F]). Meanwhile, immunofluorescence staining showed that PD-L1, CGRP and TRPV1 were partially co-localized in the dorsal horn of PIPN mice spinal cord on day 14 (Figure 4C). We also detected the mRNA levels of the inflammatory cytokines TNF-α (Figure 4G) and IL-1β (Figure 4H), the results of which were consistent with those described above. Collectively, these results indicate that, in PIPN mice, the analgesic effect of PD-L1 is sexually dimorphic and might occur through TRPV1 and CGRP.

3.4 TRPV1 is involved in PD-L1-induced sex differences in PIPN inhibition

The above experimental results showed that the involvement of PD-L1 in PIPN mice was sexually dimorphic and that injection of PD-L1 could partially block the PTX-induced upregulation of TRPV1 and CGRP at the protein and mRNA levels in the spinal cords of both male and female mice. However, the necessity of TRPV1 for the analgesic effect of PD-L1 on PIPN has not been determined. We generated a TRPV1 knockdown mouse model by i.t. injection of vector #1 or AAV9-TRPV1-RNAi viruses #2, #3 and #4. Next, PTX and PD-L1 were administered to mouse models as described above (Figure 5A). Western blot analysis (Figure 5B,C) revealed that TRPV1 knockdown by AAV9-TRPV1-RNAi virus #3 and #4 was significant (significant difference) after 4 weeks of injection. These constructs were used for subsequent experiments. Consistent with the Western blot results, compared with those of the control group, immunofluorescence staining of spinal cord sections from TRPV1^−/−-#3 and #4 mice of both revealed a significant decrease in TRPV1 expression. Notably, TRPV1 expression in the spinal cord was markedly greater in the control female group than in the corresponding male group. Moreover, there was no significant difference in the expression of TRPV1 between female and male TRPV1^−/−#3 or TRPV1^−/−#4 mice (Figure 5D). After PD-L1 injection (10 ng) for 3 h on day 14, the analgesic effect of PD-L1 on the male TRPV1^−/−-#3 and -#4 mouse models, as measured by the PWT and PWL, was partly blocked compared with that on the control group (Figure 5E,F). Similar results were obtained for the female TRPV1^−/− model (Figure 5G,H). Moreover, there was no significant difference in the analgesic effect of PD-L1 between female and male TRPV1^−/− mice. Collectively, these findings provide strong evidence that TRPV1 is involved in the PD-L1-induced sex difference inhibition of PIPN, which is shown in the schematic diagram in Figure 6.