Journal Logo

Research Paper

Targeting interleukin-20 alleviates paclitaxel-induced peripheral neuropathy

Chen, Li-Hsiena; Yeh, Yu-Minb,c; Chen, Yi-Fand; Hsu, Yu-Hsiangb; Wang, Hsiao-Hsuand; Lin, Peng-Chanc; Chang, Lian-Yuna; Lin, Chou-Ching K.e; Chang, Ming-Shif; Shen, Meng-Rua,g,*

Author Information
doi: 10.1097/j.pain.0000000000001831

1. Introduction

Paclitaxel is an established option for taxane-based regimens to treat breast, cervical, ovarian, and lung cancer. Peripheral neuropathy is a dose-limiting side effect of paclitaxel therapy that is often managed by dose reductions or delayed treatments.16,39 Paclitaxel-induced peripheral neuropathy negatively affects the quality of life, leading to the discontinuation of chemotherapy, and thus impacting overall survival. The underlying mechanisms of paclitaxel-induced peripheral neuropathy remain unclear, and no medication is effective for the prevention or treatment of this side effect. However, increasing evidence suggests that neuroinflammation is one possible mechanism.23,29 The expression of proinflammatory cytokines, induced by activated macrophages and satellite cells, is essential for the sensitization of peripheral sensory neurons, which leads to allodynia. In animal models, inhibiting macrophage infiltration and proinflammatory cytokines in the dorsal root ganglion (DRG) suppresses the development of paclitaxel-induced neuropathic nociception.23,45 Hence, preventing the production of inflammatory cytokines is one hypothesized therapeutic target for chemotherapy-induced peripheral neuropathy (CIPN).

Interleukin-20 (IL-20), a member of the IL-10 family of cytokines, plays a vital role in regulating inflammation and tissue homeostasis. IL-20 signaling, through heterodimeric receptors composed of IL-20R1/R2 or IL-20R2/IL-22R1, acts as an inflammatory mediator in multiple cell types, including monocytes, epithelial cells, endothelial cells, and astrocytes.13,32,42 As seen in the previous studies, exposure of IL-20 leads to trigger the release of inflammatory cytokines, including IL-6 and IL-1β in murine astrocytes, and promotes the secretion of chemoattractants, including macrophage inflammatory proteins-1 (MCP-1) and IL-8 in the human astrocytic cell line.6,8 Previous studies have shown that IL-20 contributes to the pathogenesis of rheumatoid arthritis, liver fibrosis, osteoporosis, breast cancer, and stroke.8,12,14,15,26 Notably, inhibiting IL-20 signaling with monoclonal antibody ameliorated the severity of rheumatoid arthritis, prevented the development of neuropathic osteoarthropathy, and reduced the production of TNF-α, IL-6, IL-1β, and IL-20 proinflammatory cytokines, indicating that it is a potent modulator of inflammatory diseases.12,15,26

Neuroinflammation is a pathological feature of CIPN, and consequently, therapeutics targeting the suppression of immune response becomes an amenable strategy for CIPN.16 Previous studies from ours and other groups suggest that IL-20 plays a role in modulating the secretion of proinflammatory cytokines and chemokines.6,8,32 In addition, our earlier data indicated that blockade of IL-20 signaling by neutralized antibody (7E) attenuates the release of proinflammatory cytokine and chemokines in cell culture and mouse serum.8,12,15 Therefore, IL-20 might be a potential therapeutic target for CIPN. Accordingly, we aimed to investigate the role of IL-20 in the pathogenesis of paclitaxel-induced peripheral neuropathy. The results highlight the fact that serum IL-20 is closely correlated with the development of paclitaxel-induced peripheral neuropathy. Importantly, targeting IL-20 with a neutralizing antibody ameliorated paclitaxel-induced peripheral neuropathy by suppressing neuroinflammation and restoring intracellular calcium ([Ca2+]i) homeostasis.

2. Methods

2.1. Patients

We recruited 12 healthy controls and 15 gynecologic cancer patients, including 6 ovary cancer and 9 cervical patients, in a chemotherapy-induced neuropathy study conducted from 2015 to 2019 by the Department of Obstetrics and Gynecology, National Cheng Kung University Hospital. In this study, the Quality of Life Questionnaire-CIPN twenty-item scale (QLQ-CIPN20) and nerve conduction velocity were used to assess the severity of CIPN and to evaluate the function of motor and sensory nerve, respectively. Quantitative sensory testing (QST) was applied to detect the thermal pain threshold in cancer patients. Before these electrophysiological examinations began, the purpose of this study, potential hazards, and experimental procedures were fully explained to each patient. Detailed patient history of medications was surveyed, and neurological examination was performed by a staff physician in a quiet, air-conditioned room with the room temperature maintained at 23 to 25°C. There were 3 time points for the blood sample collection from cancer patients. The first time point was on the admission day for the first cycle of paclitaxel treatment. The second and third time point was 2 and 7 days after the first cycle of paclitaxel treatment (at a dose of 175 mg/m2), respectively. These collected samples were sequentially centrifuged for 15 minutes at 1000g. Serum samples were aliquoted, and IL-20 level was determined by using an ELISA kit (DL200; R&D Systems, Minneapolis, MN). These procedures were performed in accordance with the guidelines approved by the Ethics Committee of National Cheng Kung University Hospital (IRB No: A-ER-103-395).

2.1.1. Recruitment of participants

The current study was conducted to investigate the potential role of targeting IL-20 in the management of CIPN for female patients with confirmed ovarian cancer or endometrial cancer receiving chemotherapy with 6 courses of 175 mg/m2 paclitaxel. Cancer patients were recruited if they met the inclusion criteria, including confirmed adequate organ function, Eastern Cooperative Oncology Group performance status 0 to 1, age ≥20 years old, and negative pregnancy test for women of childbearing potential only. In the meanwhile, cancer patients were excluded if they received neurotoxic treatment before chemotherapy with paclitaxel and had pre-existing peripheral neuropathy of any grade, active uncontrolled infection, and poor compliance.

2.1.2. Chemotherapy-induced peripheral neuropathy evaluation

For all the recruited cancer patients, the QLQ-CIPN20, nerve conduction velocity, and QST results were determined by the same neurologist before and 6 months after paclitaxel treatments. The QLQ-CIPN20 questionnaire included 9 questions for assessing sensory function, 8 questions for motor activity, and 3 questions for autonomic symptoms.22 The nerve conduction velocity studies, including F-wave latencies of the bilateral arms and legs, were detected by skin surface electrodes and recorded by an electrogoniometer (TSD130B; Biopac System, Inc, Goleta, CA).17 Four sensory submodalities in the QST including warm threshold, cold threshold, heat pain, and cold pain were evaluated with a sensory and pain threshold evaluation system (Pathway; Medoc Advanced Medical Systems, Ramat Yishai, Israel) following the established standard protocols.18

2.2. Animals

The 6-week-old female C57BL/6, BALB/c, and NOD.CB17-Prkdcscid/NcrCr mice were purchased from the Laboratory Animal Center (National Cheng Kung University, Taiwan). IL-20R1-deficient mice were generated as previously described.12 These mice were maintained on a diet of mouse chow and water ad libitum. These animals were housed in a pathogen-free, temperature- (25 ± 2°C), and the light-control environment under a 12:12-hour light–dark cycle (lights on at 6:00 am). Animal use and animal tissue collection for this study was in accordance with Animal Care guidelines and approved by the Institutional Animal Care and Use Committee (No. 103243) of National Cheng Kung University.

2.3. Cell culture and conditions

ND7/23 DRG cells, Raw 264.7 mouse macrophage cells, and 4T-1 murine mammary carcinomas were obtained from Sigma-Aldrich, St. Louis, MO, ATCC (TIB-71), and ATCC (CRL-2539), respectively. The human cervical squamous cell carcinoma (SiHa) was authenticated by the short-tandem repeats analysis using the Promega StemElite ID System (GeneLabs Life Science Corp, Taipei, Taiwan). ND7/23 DRG cells, Raw 264.7, 4T-1, and SiHa cells were grown in Dulbecco's modified Eagle's medium with 2 mM glutamine and 10% fetal bovine serum at 37°C in an environment containing 5% CO2.

2.4. RNA isolation, cDNA synthesis, and real-time quantitative polymerase chain reaction

2.4.1. RNA isolation from the dorsal root ganglion and macrophage cell line

For assessing the role of IL-20 in regulating inflammatory cytokine and chemokine expression, ND7/23 DRG and Raw 264.7 macrophages cell lines were incubated with recombinant mouse IL-20 protein (200 ng/mL; CYT-379, ProSpec, Bengaluru, Karnataka), paclitaxel (1 μM; PHYXOL; Sinphar Pharmaceutical CO, LTD, Yilan county, Taiwan, formulated in a 1:1 mixture of Cremophor EL and dehydrated ethanol), p38 inhibitor SB202190 (20 μM; #10010399, Cayman, Ann Arbor, MI), or extracellular signal-regulated kinase (ERK) inhibitor PD98059 (25 μM; #10006726, Cayman) at 37°C. These cells were collected for extracting total RNA (74106; Qiagen, Hilden, Germany) and further quantification of cDNA by real-time polymerase chain reaction (PCR).

2.4.2. RNA extraction from mouse tissues

After paclitaxel treatment, mice were euthanized by exposure to CO2; the mouse spinal cord, dorsal root ganglion, and foot-paw tissue were isolated for total RNA extraction with the RNeasy Mini Kit (74106; Qiagen).

2.4.3. Real-time quantitative polymerase chain reaction

Reverse transcription was conducted with Moloney murine leukemia virus reverse transcriptase according to the manufacturer's protocol (M1705; Promega, Madison, WI). IL-20, RANTES, IL-8, MCP‐1, TNF-α, IL‐6, IL‐1β, IL-17, TGF‐β1, and IL-10 gene expression were amplified using SYBR Green qPCR SuperMix (11733038; Invitrogen, Carlsbad, CA) with a real-time quantitative PCR system (LightCycler 480; Roche Diagnostics, Basel, Switzerland). Quantification analysis of mRNA was normalized with mouse glyceraldehyde 3-phosphate dehydrogenase as the housekeeping gene. Real-time quantitative PCR primer sequences are shown in Supplementary Table 1 (available at http://links.lww.com/PAIN/A961). The relative expression ratio of mRNA was determined by the 2−ΔΔCT method.

2.5. Generation of IL-20 and IL-20 R1 monoclonal antibodies

The IL-20 monoclonal antibody used in this study have been validated and well described elsewhere.12,14,41 Briefly, the recombinant human IL-20 protein was purified and used as an antigen to immunize the mice. The B cells of the spleen from the immunized mice were fused with myeloma cells following standard protocols. The hybridoma was selected using HAT medium. The monoclonal cell expressing anti-IL-20 antibody (named as 7E) was isolated and identified using the limiting-dilution method and direct ELISA assay.41 In the previously published article, the specificity of 7E recognizing human IL-20 and mouse IL-20 was confirmed by ELISA.14 7E did not recognize any other members of the IL-10 family cytokines. Furthermore, 7E is very potent to neutralize IL-20 activity both in vitro and in vivo.12,14,15,41 This 7E antibody has been reported for use in inhibition, immunoblot, flow cytometry, and radioimmunoassay.8,12,14,15,41 In addition, anti-IL-20R1 monoclonal antibodies (mAbs) (51D) displayed a specific affinity for the extracellular domain of human IL‐20R1 and cross‐reacted with mouse IL‐20R1.12

2.6. Immunoblot

ND7/23 were treated with IL-20 (100 or 200 ng/mL) or paclitaxel (10, 100, 1000 nM) at 37°C. The cells were then washed twice with phosphate-buffered saline (PBS) and lysed in RIPA lysis buffer (#89901; Thermo Fisher, Waltham, MA) mixed with 1 mM dithiothreitol, protease inhibitor cocktail (#78430; Thermo Fisher) and phosphatase inhibitor cocktails (#78420, Thermo Fisher) on ice. These lysates were centrifuged at 500g for 10 minutes at 4°C, and the supernatants were collected, which concentration was determined by the Bradford assay (#5000006; Bio-Rad, Hercules, CA). These supernatants (35 μg of protein/lane) were loaded onto a 12% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and subsequently transferred onto polyvinylidene fluoride (PVDF) membranes (#162-0177; Bio-Rad). These membranes were blocked in a blocking buffer (5% fat-free milk in phosphate-buffered saline with Tween-20 detergent) for 1 hour and then incubated with the primary antibody overnight at 4°C. The following immunoblot antibodies were used, including rabbit anti-Phospho-p38 MAPK (1:1000, #4511; Cell Signaling Technology, Danvers, MA), rabbit anti-p38 MAPK (1:1000, # 8690; Cell Signaling Technology), rabbit anti-Phospho-Erk1/2 (1:1000, #4370; Cell Signaling Technology), anti-Erk1/2 (1:1000, #4695; Cell Signaling Technology), and rabbit anti-β-actin (1:5000, GTX109639; GeneTex, Irvine, CA). After washing 3 times with 0.1% PBST buffer, these membranes were incubated with peroxidase-conjugated anti-rabbit secondary antibodies and then detected by the chemiluminescence reagent (GTX400006; GeneTex).

2.7. Peripheral neuropathy induced by paclitaxel

For clarifying the alleviative effect of IL-20 mAb or IL-20R1 deficit on CIPN, 7-week-old C57BL/6J, IL-20R1+/+, and IL-20R1−/− female mice were used. The animal model of paclitaxel-induced neuropathy was conducted according to the methodology described previously.7,10

After 1-week housing adaptions, these mice were weight-matched and randomized to different subgroups to receive an intraperitoneal injection with vehicle (saline) or anti-IL-20 mAb (7E, 5 mg/kg) 6 hours before paclitaxel treatment (4.5 mg/kg per injection, cumulative dosage: 18 mg/kg) on alternate days (days 1, 3, 5, and 7). The paclitaxel (PHYXOL) was obtained from Sinphar Pharmaceutical CO, LTD, formulated in a 1:1 mixture of Cremophor EL and dehydrated ethanol. Before administration, paclitaxel was diluted in 0.9% sodium chloride to a final concentration of 1 mg/mL. For assessing behavioral changes, mechanical withdrawal thresholds, thermal sensitivity, and motor coordination before and after paclitaxel treatments were measured weekly for 5 weeks by the investigators blinded to the experimental groups. The behavioral data of von Frey, tail immersion, and rotarod test presented as normalized to baseline before paclitaxel treatments.

2.7.1. Von Frey test

Electronic von Frey device (Part #2390; IITC Life Science, Woodland Hills, CA) was used to evaluate paclitaxel-induced mechanical allodynia. Before behavioral testing, mice were individually placed into Plexiglas chambers (10 cm L × 10 cm W × 13.5 cm H) on an elevated mesh platform for habituation. Blunt end von Frey filament with an electric device was applied to the midplantar surface of mouse hind paws for detecting the mechanical pressure. Mechanical thresholds of both hind paws were automatically recorded when nociceptive behavioral responses, including withdrawal or shaking, were observed. For each mouse, we recorded 8 trial responses and averaged 5 values closest to the median.

2.7.2. Tail immersion test

For assessing tail thermal sensitivity, the tail immersion assay was performed. For each testing session, each mouse gently restrained by wrapping a chux while the distal one-third of the tail was immersed in a water bath maintained at 49 ± 0.5°C, and the latency of tail withdraw was scored (the cutoff for a response is 10 seconds). The latency of tail withdrawal scored for 5 trials separated by 10 minutes, and 5 trials were averaged.

2.7.3. Rotarod performance test

The rotarod test was used to assess sensory motor coordination. Mice were placed on the accelerating rotarod until they stopped or fell. The rod speed gradually increased, starting at 4 rpm and accelerating to 40 rpm in 5 minutes. The training was conducted for 4 days with 5 trials per day. The baseline of rotarod performance was recorded on the last day of training by measuring the fall latency for individual 5 trials. After paclitaxel treatments, the rotarod test was conducted weekly for 5 weeks to determine whether sensory motor coordination was affected.

2.8. Immunofluorescence for image analyses

After ensuring the death of animals following CO2 euthanasia, mouse DRG and foot paw were isolated after paclitaxel treatment or neurobehavioral tests. Dissected tissues were fixed with 4% paraformaldehyde for 4 hours and cryoprotected with 30% sucrose in PBS for 12 hours at 4°C. These samples were embedded in optimal cutting temperature compound (OCT, CM 3050S; Leica, Wetzlar, Germany) and sliced into 20-μm-thick sections. Slides were washed with 0.05% Triton X-100 in PBS for 30 minutes and then incubated with CAS-BlockTM solution (008120; Life Technologies, Carlsbad, CA) at room temperature for 1 hour. The information of antibodies was listed as described in Supplementary Table 2 (available at http://links.lww.com/PAIN/A961). For double immunofluorescence staining, 2 primary antibodies from different species were mixed and incubated with the sections of the sample at 4°C overnight. After washing primary antibodies with PBS, the samples were then incubated with secondary antibodies (Alexa-488, 594, 1:200; Invitrogen) and Hoechst 33342 (10 mg/mL, H1399; Invitrogen) for 1 hour at room temperature. Immunofluorescence images were acquired using a 40x objective by a multiphoton laser scanning microscope (FV1000MPE; Olympus, Shinjuku City, Japan).

2.8.1. Immunofluorescence intensity analysis in dorsal root ganglion

For quantification of immunostaining using ImageJ (NIH), analyzing the fluorescence magnitudes of CD68, IL-20, IL-20 receptors, store-operated Ca2+ channels, and transient receptor potential calcium channels in mouse DRG was conducted by an observer blinded to the experimental design, and the expression of fluorescence intensities was normalized to the images of the control group.

2.8.2. Quantification of intraepidermal nerve fiber staining

Intraepidermal nerve fiber (IENF) density was calculated in the mouse hind paw intraplantar skin. Frozen 20-μm sections of mouse plantar skins were incubated with rabbit anti-PGP9.5 (1:100; ab108986, Abcam, Cambridge, United Kingdom) and goat anti-Collagens VI (1:100, AB769; Chemicon, Bengaluru, Karnataka) for IENF and the epidermal basement membrane zone staining, respectively. Then, these samples were followed and stained with secondary antibodies (Alexa488 for IENF and Alexa594 for Collagens VI, 1:200; Invitrogen) at room temperature for 1 hour. The immunofluorescence images were acquired using a deconvolution microscope (DeltaVision; GE Health, Chicago, IL) with a 40x objective. Five IENF pictures were randomly selected from each mouse of different treatment groups. The density of IENF was analyzed by a bioimaging technician blinded to experimental design and expressed as the number of IENF divided by the length of the basement membrane (IENF/mm).

2.9. Histopathologic evaluation of the sciatic nerve

After neurobehavioral tests, mice were euthanized according to Institutional Animal Care and Use Committee protocol, and mouse sciatic nerve samples were removed, fixed in 2.5% glutaraldehyde for 1 to 2 hours, and postfixed in 1% osmium tetraoxide solution for 1 hour at 4°C. These samples were then dehydrated in increasing concentrations of ethanol and embedded in Epon 812 epoxy resin (#14120, Electron Microscopy Sciences, Hatfield, PA). The sciatic nerve sections (90 nm) were prepared and stained with lead citrate and uranyl acetate. These specimens were examined and photographed using a transmission electron microscope with ×5000 magnification for analyzing axon numbers or ×15,000 magnification for assessing the myelination integrity (H7650; Hitachi, Chiyoda City, Japan) by a bioimaging technician. The myelinated fibers of mouse sciatic nerve were quantified by an observer blinded to the experimental design, and the results were presented as axon numbers per mm.2 For further analyzing the myelination integrity of mouse sciatic nerve, the G-ratio was used to calculate the ratio of inner axon circumference to outer myelin circumference.

2.10. Serum cytokine and chemokine detection

2.10.1. Human cytokine and chemokine detection

Venous blood samples were obtained from cancer patients before (first time point), 2 (second time point), and 7 days (third time point) after first paclitaxel treatment (at a dose of 175 mg/m2). These samples were collected in BD Vacutainer SST II-serum tubes (BD Biosciences, San Jose, CA) and then centrifuged at 3000g for 10 minutes to yield serum. Serum levels of IL-20 cytokine were measured by commercial human IL-20 quantikine ELISA kit (DL200, R&D system) following the manufacturer's protocols. The lower detection limits for human IL-20 were 16.6 pg/mL. Levels of RANTES, MCP-1, monokine induced by gamma interferon (known as MIG), IL-8, and interferon gamma-induced protein 10 (IP-10) chemokines were further measured by using cytometric bead array (human chemokine array, Cytometric Bead Array, BD Biosciences, San Jose, CA) and calculated by flow cytometers (Coulter CytoFLEX S Flow Cytometry; Beckman, Brea, CA). The levels of cytokine and chemokine were expressed as normalized to baseline.

2.10.2. Mouse cytokine and chemokine detection

Before blood collection, mice were randomized to different subgroups and sequentially treated with vehicle (saline) or anti-IL-20 mAb (7E, 5 mg/kg) 6 hours before paclitaxel treatment (4.5 mg/kg) using 4 intraperitoneal injections given on alternate days. After euthanasia with CO2, mouse blood samples were obtained and then centrifuged at 3000g for 10 minutes to yield serum. Levels of cytokine and chemokine in mouse serum were measured by immunology multiplex assay (MCYTOMAG-70K; Millipore, Burlington, MA) and analyzed by Luminex 200.

2.11. Intracellular calcium measurement

ND7/23 DRG cells were plated at a density of 3000 cells/cm2 on the glass-bottom dish and incubated with 2 µM fura-2/acetoxymethylester (Fura-2/AM, F1201; Invitrogen) in DMEM medium for 30 minutes at 37°C and then washed with Ca2+-free HEPES solution (pH 7.40). The Fura-2/AM was excited alternatively between 340 and 380 nm. The emission wavelength was 510 nm. The fluorescence images were detected by the Olympus IX71 inverted microscope. Intracellular calcium measurement was monitored using the Polychrome IV monochromator (Till Photonics GmbH, Munich, Germany) and calculated by using the TILLvisION 4.0 program (Till Photonics). For [Ca2+]i measurement, ND7/23 cells were stimulated with 5 nM IL-20 (CYT-379; ProSpec) or 2 µM thapsigargin (10522; Cayman) in Ca2+-free buffer for 10 minutes. To further measure IL-20–induced endoplasmic reticulum store depletion (store-operated Ca2+ entry [SOCE]), these cells were readmitted with 2 mM Ca2+ and recorded for 5 minutes. For basal [Ca2+]i measurement, ND7/23 cells in the area on the glass-bottom dish were selected to determine the background fluorescence level and subtracted from each of the regions of interest (ROIs). The results of [Ca2+]i measurement were expressed as ratios between 340 and 380 fluorescence signals (340 nm/380 nm). The mean ratio was estimated in each ROI and then averaged between all ROIs to calculate the mean trace. The SOCE Δ[Ca2+]i was expressed as maximum minus the minimum ratio after IL-20 and 2 mM Ca2+ addition, respectively.

2.12. Cancer cell viability

Human cervical cancer, SiHa cells were seeded at a density of 1 × 105 cells/well onto 6-well plates for counting cell number or 5 × 103 cells/well into 96-well plates for cell viability. After seeding overnight, these cells were treated with paclitaxel, anti-IL-20 mAb (7E, 2 μg/mL), anti-IL-20R1 mAb (51D, 2 μg/mL), paclitaxel plus anti-IL-20 mAb, or paclitaxel plus anti-IL-20R1 mAb and incubated for 0, 24, 48, or 72 hours. The effects of IL-20 mAb or IL-20R1 mAb on suppressing the viability of cancer cells by paclitaxel were detected by methylthiazolyldiphenyl-tetrazolium bromide assay following the manufacturer's procedures (CT02; Sigma-Aldrich). The cell viability was estimated using SpectraMax i3x microplate reader to measure optical density at 570 nm with a reference wavelength of 630 nm. Cancer cell viability was presented as normalized to the optical density from the control.

2.13. Tumor xenograft models

In mouse tumor xenograft models, with 5 × 105 murine breast cancer, 4T-1 or 5 × 106 human cervical cancer SiHa cells were subcutaneously implanted in the basolateral flank of 10-week-old female BALB/c mice or NOD.CB17-Prkdcscid/NcrCr (NOD/SCID) mice, respectively. When tumor sizes reached approximately 100 mm3, mice were randomized into different groups to receive paclitaxel (4.5 mg/kg per injection, cumulative dosage: 18 mg/kg), anti-IL-20 mAb (7E, 5 mg/kg), anti-IL-20R1 mAb (51D, 6 mg/kg), or vehicle injection on 4 alternative days. Tumor size was measured with an external digital caliper every other day by an observer blinded to experimental design. Tumor volume (mm3) was estimated by using the formula: 1/2 (the shortest tumor diameter)2 × (the longest tumor diameter).

2.14. Statistical analysis

Data are presented as the mean ± SEM. Statistical analyses were calculated using statistics software SPSS Statistics 15.0 (IBM SPSS Statistics). To analyze behavioral responses, differences between groups were compared using a two-tailed Student's t-test or two-way analysis of variance followed by the Bonferroni post hoc analysis. The criterion for significance was P < 0.05.

3. Results

3.1. Serum IL-20 level is associated with chemotherapy-induced peripheral neuropathy risk

We first determined if the proinflammatory cytokine IL-20 functioned as a predictive biomarker of CIPN. Serum levels of IL-20 were sequentially measured in healthy subjects and gynecological cancer patients undergoing paclitaxel treatment (Fig. 1A; Supplementary Table 3, available at http://links.lww.com/PAIN/A961). The Quality of Life Questionnaire-CIPN twenty-item scale (QLQ-CIPN20), nerve conduction study, and heat pain thresholds were used to assess neuropathy and peripheral nerve function before, 1, and 6 months after paclitaxel treatment, respectively. No differences were noted between healthy subjects and gynecological cancer patients in baseline serum IL-20 level and body surface area (Supplementary Table 3, available at http://links.lww.com/PAIN/A961). For healthy controls, serum IL-20 levels did not significantly change within 1 week (Fig. 1A). By contrast, 2 patterns of changes in serum IL-20 levels were observed in cancer patients treated with chemotherapy. In group A, the serum IL-20 levels significantly increased 1 day after chemotherapy and returned to basal levels within 1 week. Simultaneous measurement of chemokines indicated that the serum levels of MCP-1, MIG, and IP-10 were increased 1 week after chemotherapy in group A (Supplementary Fig. 1, available at http://links.lww.com/PAIN/A961). By contrast, serum IL-20 levels did not significantly change in group B (Fig. 1A), and only serum MCP-1 levels were increased 1 week later in this group (Supplementary Fig. 1, available at http://links.lww.com/PAIN/A961). These results are consistent with the concept that IL-20, a pleiotropic inflammatory cytokine, displays the chemotactic characteristics, upregulating the expression of chemokines in inflammatory diseases.32,43 Importantly, cancer patients in group A had an increased CIPN score, decreased the amplitude of nerve conduction, and increased the heat pain threshold in the 6-month follow-up (Fig. 1B). This implies that increased serum IL-20 levels are accompanied by an increased risk of CIPN.

Figure 1.
Figure 1.:
Elevated serum IL-20 levels are associated with severe sensory pain in cancer patients after paclitaxel treatment. (A) Sequential measurement of serum IL-20 within 1 week for healthy controls (n = 12) and gynecological cancer patients treated with paclitaxel (n = 10 from group A; n = 5 from group B). **P < 0.01 compared with serum IL-20 levels on day 2, one-way repeated-measures analysis of variance (ANOVA). (B) Patients (group A) with high serum IL-20 expression on the day 2 reported higher pain scores based on the Quality of Life Questionnaire-CIPN twenty-item scale (QLQ-CIPN20, left), had lower nerve conduction amplitude of the median sensory nerve (middle), and displayed higher heat pain thresholds (right). *P < 0.05 vs baseline, one-way repeated-measures ANOVA. (C) Representative images (right) and quantitative analyses (left) of mouse dorsal root ganglion (DRG) neurite outgrowth after IL-20 and/or paclitaxel treatment. Scale bar, 50 μm. Data are shown as the mean ± SEM (n > 5/group). #P < 0.001 vs vehicle group; *P < 0.05; ***P < 0.001 vs paclitaxel treatment, two-way ANOVA. CIPN, chemotherapy-induced peripheral neuropathy.

To assess potential neurotoxicity, we studied the in vitro effect of IL-20 on DRG, a major target of chemotherapy-induced neuron damage. As shown in Figure 1C, paclitaxel and recombinant IL-20 exerted the dose-dependent toxic effects on neurite outgrowth of DRG. Importantly, the synergistic neurotoxicity of paclitaxel and IL-20 was observed. Together, the results suggest that changes in serum IL-20 after chemotherapy are associated with the development of CIPN.

3.2. IL-20 levels fluctuate with paclitaxel treatment in a mouse model of chemotherapy-induced peripheral neuropathy

Macrophage infiltration and cytokine secretion in pain circuits have been implicated in the development of CIPN.23,29,44,45 To elucidate the relevance of IL-20 production to paclitaxel treatment, mouse models of CIPN were applied,7 and time-course changes in serum IL-20 levels after treatment were investigated. After a single paclitaxel (4.5 mg/kg) injection, serum IL-20 levels from C57BL/6J female mice were elevated on day 2 and reduced on day 15 (Fig. 2A). Similarly, serum IL-20 levels also varied with 4 paclitaxel injections in mouse models of CIPN (Fig. 2A). To further examine activation of the neuroimmune system as a potential cause of paclitaxel-induced neuropathy, transcript levels of inflammatory cytokines and chemokines were estimated by real-time PCR in the mouse spinal cord, DRG, and foot paws on days 0, 2, 8, and 15 after paclitaxel treatment. Compared with those in vehicle control groups, paclitaxel caused a dramatic increase in IL-20 levels on day 8 and subsequently activated TNF-α on day 15 in DRG (Supplementary Fig. 2B, available at http://links.lww.com/PAIN/A961). Furthermore, IL-20 expression also increased in the mouse foot paw and spinal cord on day 15 after paclitaxel treatment (Supplementary Figs. 2A and C, available at http://links.lww.com/PAIN/A961). These findings revealed that paclitaxel induces the expression of a wide variety of inflammatory and chemokine genes in peripheral nerve tissues, with elevated IL-20 levels in DRG.

Figure 2.
Figure 2.:
Serum IL-20 cytokine levels fluctuate after paclitaxel treatment in a mouse model. (A) Measurement of mouse serum IL-20 levels after paclitaxel treatment (n > 3/group). *P < 0.05 compared with serum IL-20 levels on day 0, one-way ANOVA. (B) Representative images and (C) quantification of CD68 (a macrophage marker), IL-20, IL-20 receptor-1 (IL-20R1), IL-20R2, and IL-22R1 immunoreactivity in dorsal root ganglion (DRG) sections of mice on day 0, 2, 8, or 15 after the first paclitaxel treatment. Scale bar, 20 μm. N = 5, ***P < 0.001 vs immunoreactivity on day 0, one-way ANOVA. All data are presented as the mean ± SEM. ANOVA, analysis of variance.

To further assess whether paclitaxel induces the infiltration of macrophage and the overexpression of IL-20 and its receptor in DRG, thus, mouse DRG were immunostained for the macrophage marker CD68, IL-20, and IL-20 receptors (including IL-20R1, IL-20R2, and IL-22R1) (Fig. 2B). Paclitaxel indeed triggered the time-dependent infiltration of macrophages and upregulated IL-20 in mouse DRG neurons, which reached its peak on day 8 after paclitaxel treatment (Fig. 2C). Furthermore, the expression of IL-20R1, but not IL-20R2 and IL-22R1, was significantly associated with responsiveness to paclitaxel treatment (Fig. 2C). Cumulatively, data from CIPN mouse models suggest that proinflammatory IL-20 levels fluctuate with paclitaxel treatment in the mouse serum, spinal cord, DRG, and peripheral tissues.

3.3. IL-20 potentiates paclitaxel-induced inflammatory responses partly through MAPK signaling

We conducted the in vitro experiments to elucidate the linkage of elevated IL-20 to paclitaxel-induced inflammatory responses in the DRG. As shown in Figure 3A and B, ND7/23 DRG cells and Raw 264.7 macrophages were incubated with recombinant mouse IL-20 protein at different time intervals. After 10 hours of IL-20 incubation, the mRNA expression of TNF-α and MCP-1 significantly increased in ND7/23 DRG cells. Meanwhile, IL-20 treatment transiently induced TNF-α and MCP-1 expression in Raw 264.7 cells. The above data also support that IL-20, as a regulator, triggers proinflammatory cytokine and chemokine secretion in DRG and macrophage cells.

Figure 3.
Figure 3.:
IL-20 incubation upregulated paclitaxel-induced inflammatory responses. Quantitative analysis of cytokines and chemokines from (A) ND7/23 DRG and (B) Raw 264.7 macrophage cells with IL-20 treatment (200 ng/mL) at different time intervals (n = 3/group). The expression levels of TNF-α (top) and MCP-1 (bottom) were analyzed using RT-qPCR. *P < 0.05; **P < 0.01; ***P < 0.001, compared with control, by one-way ANOVA. (C) ND7/23 DRG cells were incubated with IL-20 or paclitaxel and analyzed using immunoblotting with specific antibodies against phospho-P38 (top), against phospho-ERK1/2 (bottom), and β-Actin antibody was used as an internal control. Elevated inflammatory responses after incubation of IL-20 and paclitaxel in (D) ND7/23 DRG and (E) Raw 264.7 macrophage cells were decreased by ERK1/2 (PD98059, 25 μM) or P38 inhibitors (SB202190, 20 μM). N = 3/group. *P < 0.05; **P < 0.01; ***P < 0.001, vs vehicle group; #P < 0.01, vs group with IL-20 and paclitaxel treatment, by one-way ANOVA. ANOVA, analysis of variance; DRG, dorsal root ganglia.

Mitogen-activated protein kinases (MAPKs), including ERK and p38, are involved in mediating cytokine-induced nociceptive signaling cascade and in the development of paclitaxel-induced neurotoxicity.24 For further studying whether IL-20 acts as an upstream regulator in aggravating paclitaxel-induced inflammatory responses by enhancing MAPKs signaling, ND7/23 DRG cells were treated with IL-20 and paclitaxel. As shown in Figure 3C, paclitaxel has dosage-dependent effects in the activation of p38 and ERK1/2 kinases. Furthermore, the incubation of IL-20 magnified the activation of the MAPK signaling pathway provoked by paclitaxel.

To further study whether MAPK signaling pathways contribute to IL-20 exposure in the exacerbation of paclitaxel-induced neuroinflammatory responses, the ERK1/2 and p38 MAPK inhibitor were applied to study their effects. Paclitaxel or IL-20 treatment alone stimulated inflammatory TNF-α, IL-6 cytokines, and MCP-1 chemokine expression in both ND7/23 DRG and Raw 264.7 macrophage cells (Figs. 3D and E). Incubation of paclitaxel and IL-20 aggravated the activation of these cytokines and chemokines (Figs. 3D and E). The application of ERK1/2 and p38 MAPK inhibitors diminished these elevated inflammatory responses by blocking the MAPK signaling pathway (Figs. 3D and E). The above results suggested that IL-20 upregulated inflammatory cytokine expression partly through activating MAPK signaling pathway.

3.4. The neuroprotective effects of targeting IL-20 in vivo alleviate chemotherapy-induced peripheral neuropathy by suppressing inflammatory cascades

Previous studies have reported that inhibiting proinflammatory cytokine signaling can attenuate nociceptive hypersensitivity and inflammation associated with nerve injury.25,44 However, whether IL-20 upregulation contributes to paclitaxel-induced peripheral neuropathy remains unclear. To determine whether an anti-IL-20 therapeutic strategy could alleviate paclitaxel-induced pain, mouse models of paclitaxel-induced neuropathy, including von Frey, tail immersion, and rota-rod tests were conducted. C57BL/6J female mice at 7 weeks of age received 5 mg/kg of an anti-IL-20 mAb 6 hours before the intraperitoneal injection of paclitaxel (4.5 mg/kg) on 4 consecutive days (Fig. 4A). Paclitaxel causes both peripheral motor and sensory neuropathies,23,29,39 and in our study, this included behavioral dysfunction associated with mechanical sensitivity (Fig. 4B), thermal sensation (Fig. 4C), and motor coordination (Fig. 4D). Anti-IL-20 mAb administration before paclitaxel treatment attenuated mechanical allodynia (Fig. 4B) and prevented thermal hypoesthesia (Fig. 4C). Moreover, paclitaxel-induced motor coordination defects were significantly improved by anti-IL-20 mAb treatment, based on the latency required for mice to fall from the rota rod (Fig. 4D). We also designed experiments to study whether the humanized anti-IL-20 mAb could dampen paclitaxel-induced neuropathy. Similar to the protective effect of mouse anti-IL-20 mAb, pretreatment with a humanized anti-IL-20 mAb also alleviated sensory impairment and coordination deficits (Supplementary Fig. 3, available at http://links.lww.com/PAIN/A961). These results demonstrated that anti-IL-20 mAbs protect against the development of paclitaxel-induced sensory and motor polyneuropathy.

Figure 4.
Figure 4.:
Systemic blockade of IL-20 prevents paclitaxel-induced neuropathy and peripheral nerve damage. (A) Experimental design to assess the protective effects of an anti-IL-20 monoclonal antibody (mAb) on paclitaxel-induced neuropathy. (B–D) Administration of the anti-IL-20 mAb reduced paclitaxel-induced mechanical allodynia, thermohypesthesia, and motor dysfunction. Right: quantification of the mechanical threshold, thermal sensitivity, and motor coordination in the fifth week (n = 10 mice/group). *P < 0.05; **P < 0.01; ***P < 0.001, by two-way ANOVA. (E) Transmission electron micrographs of the sciatic nerve. Bottom: the measurement of G-ratio represents the ratio of axon diameter (dashed line) to myelin (solid line). N, the nucleus of the Schwann cell. Black dashed arrow, dying axons. White dashed arrow, axons with detached compact myelin. White arrow, myelin debris. Scale bar, 4 μm (top), 1 μm (middle and bottom). (F) Quantification of myelinated fiber density and (G) the G-ratio in mouse sciatic nerves (n = 5 mice/group). (H) Immunofluorescence images and (I) quantification of intraepidermal nerve fibers suggest that the anti-IL-20 mAb reduced paclitaxel-induced epidermal innervation loss (n = 5 mice/group). Scale bar, 30 μm. **P < 0.01; ***P < 0.001, by two-way ANOVA. ANOVA, analysis of variance.

Nerve tissue ultrastructure was studied at the fifth week after paclitaxel treatment. As shown in Figure 4E, paclitaxel caused nerve damage to both myelinated and nonmyelinated fibers, and nerve tissues showed varying degrees of dying axons (black dashed arrow), axons with detached compact myelin (white dashed arrow) and accumulated myelin debris (white arrow). By contrast, anti-IL-20 mAb treatment alleviated this nerve damage (Fig. 4E). Quantitative analysis of myelinated fiber densities in mouse sciatic nerves revealed a 24% decrease in paclitaxel-treated mice compared with those in controls (P < 0.01). By contrast, the anti-IL-20 mAb partially rescued paclitaxel-induced axon loss (Fig. 4F). To assess myelination integrity, we calculated the G-ratios of myelinated fibers in sciatic nerves. Paclitaxel treatment significantly decreased this (axon diameter < 5 μm) compared with that in controls (P < 0.05), whereas the anti-IL-20 mAb remarkably alleviated this axonal demyelination (P < 0.05; Fig. 4G).

Several lines of data suggest that morphologic changes in skin innervations are associated with clinical measures of small-fiber neuropathy.4,21,38 An elevation in proinflammatory cytokines is one cause of small fiber neuropathy.40 Accordingly, we proposed that IL-20 inhibition is a potential therapeutic modality for paclitaxel-induced IENF loss. Therefore, IENF density was calculated after paclitaxel treatment. IENFs (white arrow) and the epidermal basement membrane zone (white dashed line) in the mouse hind paw intraplantar skin were immunolabeled for PGP9.5 and collagen IV, respectively (Fig. 4H). Paclitaxel treatment caused a 61% IENF loss in the intraplantar epidermis. By contrast, there was no difference in IENF density in paclitaxel-treated animals cotreated with the anti-IL-20 mAb, as compared to that in controls (P < 0.01; Fig. 4I). Hence, the anti-IL-20 mAb was protective against paclitaxel-induced neuropathy.

Furthermore, we conducted additional experiments to assess whether blocking IL-20 dampens paclitaxel-induced inflammatory responses, alters macrophage polarization, and decreases macrophage infiltration. Results showed that paclitaxel treatment enhanced serum levels of IP-10, MIG chemokines, and proinflammatory TNF-α, IL-1β, and IL-20p70 cytokines, but not anti-inflammatory cytokines (Figs. 5A-B). The anti-IL-20 antibody also dampened the inflammatory response to paclitaxel (Figs. 5A-B). Consistent with previous findings, paclitaxel triggered the infiltration of proinflammatory M1 macrophages into DRG tissues (Figs. 5C and F). However, anti-IL-20 treatment before each paclitaxel injection abolished paclitaxel-induced M1 macrophage recruitment in DRG neurons (Figs. 5C and F). These findings suggest that paclitaxel drastically impaired sciatic nerve function attenuated skin innervation and induced proinflammatory cascades, whereas the anti-IL-20 mAb prevented this peripheral nerve damage and inflammation.

Figure 5.
Figure 5.:
Blocking IL-20 signaling by antibody injections significantly diminishes systemic inflammation and reduces paclitaxel-induced M1 macrophage recruitment in mouse dorsal root ganglion (DRG) neurons after paclitaxel treatment. (A) Representative photographs showing serum cytokine and chemokine levels isolated from mice treated with vehicle, anti-IL-20 mAb, paclitaxel, or anti-IL-20 mAb and paclitaxel on day 8 after the first paclitaxel injection. (B) Quantification of the expression of chemokines (top), proinflammatory cytokines (middle), and anti-inflammatory cytokines (bottom panel, n = 5 mouse/group). Data are shown as the mean ± SEM. *P < 0.05, vs vehicle group, based on two-way ANOVA. (C and D) Representative photomicrographs showing immunostaining of DRG sections from mice treated with vehicle, anti-IL-20 mAb, paclitaxel, or anti-IL-20 mAb and paclitaxel on day 8 after the first paclitaxel injection. Sections were triple labeled with (C) anti-NOS2 (red, M1-like macrophage marker), anti-CD68 (green, macrophage marker), and DAPI (blue) or (D) anti-Arg-1 (red, M2-like macrophage marker), anti-CD68 (green, macrophage marker), and DAPI (blue). Inserts show the lower magnification of DRG sections, and the dashed squares represent the enlarged view. Scale bar, 50 μm. Quantification of the number of (E) macrophages and (F) NOS2+/CD68+ (M1) and Arg-1+/CD68+ (M2) macrophages (n > 5 sample/group). Data are shown as the mean ± SEM. *P < 0.05; ***P < 0.001 vs paclitaxel group, based on two-way ANOVA. ANOVA, analysis of variance; mAb, monoclonal antibody.

We also investigated whether blocking IL-20 signaling could reverse paclitaxel-induced neurotoxicity. Anti-IL-20 (7E, 5 mg/kg) or anti-IL-20R1 mAbs (51D, 6 mg/kg) were administered twice weekly after neuropathy was induced, and mouse behavior tests were performed (Fig. 6A). Two weeks after paclitaxel injection, the mice exhibited a significant decrease in mechanical threshold (Fig. 6B) and an increase in thermal threshold (Fig. 6C). Anti-IL-20 or anti-IL-20R1 mAbs partially alleviated mechanical allodynia and reduced thermohypesthesia (Figs. 6B and C). In addition, paclitaxel treatment dramatically diminished the 30% time on the rota rod at the fifth week (Fig. 6D). Anti-IL-20 or anti-IL-20R1 mAbs remarkably improved paclitaxel-induced motor coordination deficits (Fig. 6D). We also assessed the beneficial effect of anti-IL-20 and anti-IL-20R1 mAbs on axon demyelination in the sciatic nerve (Fig. 6E) and loss of cutaneous innervations (Fig. 6F) after paclitaxel treatment. Mice receiving 4 doses of paclitaxel exhibited a decrease in myelinated fiber densities (P < 0.01; Fig. 6G) and a decline in the G-ratio of sciatic nerves (P < 0.001; Fig. 6H). Meanwhile, IENFs were significantly depleted after paclitaxel treatment (P < 0.001; Fig. 6I). However, inhibition of IL-20 signaling after paclitaxel treatment alleviated axon loss (Fig. 6G) and remarkably reversed paclitaxel-induced demyelination (Fig. 6H). Furthermore, anti-IL-20 and anti-IL-20R1 mAbs moderately ameliorated decreases in skin innervation in mouse hind paws caused by paclitaxel (Fig. 6I). These findings demonstrated that the severity of paclitaxel-induced neuropathy could be improved by anti-IL-20 or anti-IL-20R1 mAbs.

Figure 6.
Figure 6.:
Immunotherapeutic strategies targeting IL-20 restore paclitaxel-induced nerve damage. (A) Experimental design to assess the therapeutic effect of anti-IL-20 therapy on paclitaxel-induced polyneuropathy. (B–D) Administration of an anti-IL-20 monoclonal antibody (mAb) or anti-IL-20R1 mAb after paclitaxel treatment alleviated mechanical allodynia, thermohypesthesia, and motor dysfunction. Right: quantification of the mechanical threshold, thermal sensitivity, and motor coordination in the fifth week (n = 10 mice/group). *P < 0.05; **P < 0.01; ***P < 0.001, by two-way ANOVA. (E) Histological analysis of the sciatic nerve revealed that paclitaxel-induced axon loss and demyelination were improved by anti-IL-20 mAb treatment. Scale bar, 4 μm (for top panels), 1 μm (for middle and bottom panels). (F) Representative images are showing that anti-IL-20 and anti-IL-20R1 mAb treatment alleviated paclitaxel-induced skin innervation loss. Scale bar, 30 μm. (G) Quantification of myelinated fiber density and (H) the G-ratio in the sciatic nerves. (I) Quantification of intraepidermal nerve fibers in the glabrous skin of hind paw footpad. N = 5, **P < 0.01; ***P < 0.001, by two-way ANOVA. Data represent mean ± SEM. ANOVA, analysis of variance.

3.5. IL-20R1 deficiency protects against paclitaxel-induced nerve damage

To further confirm the critical involvement of IL-20 in CIPN, we analyzed the effect of IL-20R1 deficiency on paclitaxel-induced peripheral neuropathy by using IL-20R1-knockout mice. After the protocol is shown in Figure 4A, both IL-20R1+/+ and IL-20R1−/− mice developed hyperalgesia in response to mechanical stimulation (Fig. 7A). However, mechanical withdrawal thresholds were gradually recovered in IL-20R1−/− mice (Fig. 7A). Furthermore, IL-20R1−/− mice exhibited diminished thermohypesthesia and coordination deficits when compared with IL-20R1+/+ mice (Figs. 7B and C). To further confirm that IL-20R1 deficiency can prevent paclitaxel-induced neuroinflammation, we performed immunofluorescence analysis of mouse DRG neurons, by labeling with macrophage markers CD68, IL-20, or IL-20R1. Paclitaxel treatment induced macrophage recruitment, IL-20 accumulation, and IL-20R1 overexpression in DRG neurons of IL-20R1+/+ mice (Figs. 7D–G). By contrast, decreases in macrophage recruitment and IL-20 expression in DRG were observed in IL-20R1−/− mice treated with paclitaxel (Figs. 7D–F).

Figure 7.
Figure 7.:
IL-20R1 deficiency protected mice from paclitaxel-induced polyneuropathy and peripheral nerve degeneration. (A) IL-20R1 deficiency decreased paclitaxel-induced mechanical hypersensitivity. Right: Quantification of mechanical withdrawal threshold 5 weeks after the last paclitaxel injection. (B) IL-20R1 knockout prevented mice from paclitaxel-induced thermohypesthesia. Right: Quantification of thermal sensitivity in the fifth week. (C) The effect of IL-20 deficiency on paclitaxel-induced motor dysfunction. Right: Quantification of motor coordination assessed by the rotarod test in the fifth week. Data are presented as mean ± SEM. N = 10, ***P < 0.001 vs paclitaxel group, by two-way ANOVA. (D) Double-staining of CD68 (macrophage marker, green) and NeuN (neuron marker, red) in mouse dorsal root ganglion (DRG) neurons. Inserts indicate the lower magnification, and the dash squares represent the enlarged view of DRG sections. Scale bar, 20 µm. Right: Quantitative analyses of macrophage density in DRG neurons at day 8 after the first paclitaxel administration. Data are shown as mean ± SEM (n = 6 samples/group). ***P < 0.001 vs paclitaxel group, by two-way ANOVA. (E) Double-staining of IL-20 (red) and IL-20R1 (green) in DRG neurons from IL-20R1 knockout mice after paclitaxel treatments. Scale bar, 50 µm. (F) Quantitative analyses of IL-20 and (G) IL-20R1 in DRG from paclitaxel-treated IL-20R1 knockout and wildtype mice. Data are shown as mean ± SEM (n = 6 samples/group). ***P < 0.001 vs paclitaxel group, by two-way ANOVA. (H) Representative transmission electron micrographs of the sciatic nerve from IL-20R1 knockout mice after paclitaxel-induced axon loss and demyelination. Scale bar, 4 μm (for top panels); 1 μm (for middle and bottom panels). (I) Quantification of density and (J) the G-ratio of myelinated fibers in the sciatic nerves from IL-20 knockout and wildtype mice treated with paclitaxel or vehicle injections. Data are shown as mean ± SEM. N = 5, *P < 0.05; **P < 0.01; ***P < 0.001 vs paclitaxel group, by two-way ANOVA. ANOVA, analysis of variance.

After completing behavioral experiments, we assessed the ultrastructure of mouse sciatic nerves (Figs. 7H–J). In terms of myelinated fiber densities and the integrity sciatic nerve myelination, there were no significant differences between IL-20R1+/+ and IL-20R1−/− mice. Paclitaxel treatment induced severe axonal degeneration and myelin damage in the sciatic nerves of IL-20R1+/+ mice (Figs. 7I and J). By contrast, after paclitaxel treatment, IL-20R1−/− mice exhibited no significant deficits in sciatic nerve morphology compared to that in IL-20R1+/+ mice (Figs. 7H-J).

3.6. IL-20 modulates calcium homeostasis and contributes to chemotherapy-induced neurotoxicity

Based on the previous findings, paclitaxel-induced macrophage activation and IL-20 secretion were attenuated in the DRG of IL-20R1-knockout mice. Thus, we examined whether IL-20R1 deficiency protects DRG neurons against IL-20- or paclitaxel-induced neurotoxicity in vitro. Compared to that in wild-type DRG controls (Fig. 1C), exposing IL-20R1-deficient DRG neurons to paclitaxel attenuated neurite outgrowth (Figs. 8A and B). Furthermore, IL-20 did not cause neurite damage in IL-20R1-knockout DRG (Figs. 8A and B). This suggested that IL-20 is also a critical factor in modulating neurite outgrowth.

Figure 8.
Figure 8.:
The potential effect of IL-20 on paclitaxel-induced neurite damage through the remodeling of calcium homeostasis. (A) Representative images and (B) quantitative analysis of neurite outgrowth from IL-20R1-knockout mouse dorsal root ganglion (DRG) with IL-20 cytokine and/or paclitaxel treatment (n = 5/group). Scale bar, 50 mm. Green, anti-β-III tubulin antibody. #P < 0.001 vs vehicle, by two-way ANOVA. (C) Both IL-20 and thapsigargin (TG) induced store-operated Ca2+ entry (SOCE) activation. (D) Paclitaxel potentiated IL-20-induced SOCE activation. Representative intracellular calcium ([Ca2+]i) measurement and quantitative analysis of SOCE amplitude in DRG neurons (n > 50 cells/group). Arrow, adding 2 μM of TG or 200 ng/mL of IL-20. ***P < 0.001, by t-tests. (E) Fluorescence imaging and quantitative analysis of SOCE compartments and transient receptor potential (TRP) channels in DRG neurons from paclitaxel-treated mice. Scale bar, 50 μm. *P < 0.05; *P < 0.01; ***P < 0.001, by t-tests. (F–I) Quantification of the fluorescence intensity of Orai1, STIM1, TRPC1, and TRPA1 in DRG neurons from mice after vehicle, paclitaxel, IL-20 monoclonal antibody (mAb), or IL-20R1 mAb treatment. Data are shown as the mean ± SEM. *P < 0.001, two-way ANOVA. ANOVA, analysis of variance; SOCE, store-operated calcium entry.

An important mechanism for the regulation of neurite outgrowth is the control of intracellular calcium homeostasis.11 Under inflammatory conditions, perturbed calcium homeostasis results in neurite damage. Previous studies have shown that SOCE components participate in modulating calcium homeostasis in response to inflammation and partly contribute to chemotherapy-induced neuropathic pathologies.7,10,30 Therefore, we investigated whether IL-20 augmented paclitaxel-induced neurite damage is mediated by the dysregulation of calcium homeostasis. Pretreatment with a Ca2+-free solution and thapsigargin, a sarco/endoplasmic reticulum Ca2+ ATPase pump inhibitor, transiently increased [Ca2+]i by blocking calcium incorporation into endoplasmic reticulum Ca2+ stores. The sequential application of 2 mM Ca2+ solution also triggered calcium entry in DRG neurons. Accordingly, IL-20 increased intracellular Ca2+ levels and induced SOCE activation, demonstrating that both thapsigargin and IL-20 can trigger SOCE activity in DRG (Fig. 8C). Previous findings showed that paclitaxel perturbs SOCE activity, contributing to neurotoxicity.7,10 We further examined the additive effect of IL-20 on paclitaxel-induced SOCE dysregulation. IL-20 significantly enhanced SOCE activation in DRG neurons pretreated with paclitaxel for 24 hours (Fig. 8D). This indicated that IL-20 could promote paclitaxel-induced neurotoxicity by regulating calcium homeostasis in DRG neurons.

Store-operated Ca2+ channels (including STIM1 and Orai1) and transient receptor potential calcium channels, including TRPC (canonical), TRPA (ankyrin), TRPV (vanilloid), and TRPM (melastatin), have been reported as targets for the development of neuropathy.1,2,9,30,33 Thus, we examined if paclitaxel regulates intracellular calcium homeostasis by modulating these calcium channels. The expression of calcium channels in mouse DRG was assessed by immunofluorescence on day 8 after paclitaxel treatment. Compared to those in control mice, the numbers of Orai1+, STIM1+, TRPC1+, TRPA1+, TRPV1+, TRPV4+, and TRPM2+ DRG neurons from paclitaxel-treated mice were significantly increased (Fig. 8E). This demonstrated that IL-20 enhances [Ca2+]i and triggers SOCE. Accordingly, the effect of IL-20 on calcium channel upregulation and SOCE component expression was examined. Anti-IL-20 or anti-IL-20R1 mAbs alone did not affect the expression of SOCE components (Figs. 8F–I and Supplementary Figs. 4A–D, available at http://links.lww.com/PAIN/A961). However, anti-IL-20 or anti-IL-20R1 mAb treatment before each paclitaxel injection prevented the paclitaxel-induced increase in Orai1+, STIM1+, TRPC1+, and TRPA1+ DRG neurons (Figs. 8F–I and Supplementary Figs. 4A–D, available at http://links.lww.com/PAIN/A961). These results demonstrated the protective mechanism of anti-IL-20 mAb on paclitaxel-induced neurotoxicity, through the modulation of SOCE components and TRP calcium channels.

3.7. Effect of IL-20 blockade on the antitumor efficacy of paclitaxel

IL-20 was reported to regulate tumor growth and metastasis.15,42 To prove that an anti-IL-20 mAb would not affect paclitaxel efficacy in vivo, BALB/c mice were subcutaneously injected with 4T-1 cancer cells. When tumor volumes reached approximately 100 mm3, mice were administrated with anti-IL-20 or anti-IL-20R1 mAbs 6 hours before paclitaxel treatment on 4 alternating days. Tumors in the vehicle group exhibited rapid growth, whereas paclitaxel alone significantly inhibited the growth of breast cancer cells (Fig. 9A). Blocking IL-20 signaling inhibited 4T-1 murine breast cancer cell growth but had no additive effect with paclitaxel (Fig. 9A). We also tested whether the inhibition of IL-20 signaling alters human cervical tumor growth in vitro and in vivo. Blocking IL-20 signaling did not affect SiHa cell growth and did not influence the antitumor effect of paclitaxel (Supplementary Figs. 5B and D, available at http://links.lww.com/PAIN/A961). These findings demonstrated that IL-20 has different effects on the growth of various cancer cells and that an IL-20 antibody does not affect paclitaxel efficacy.

Figure 9.
Figure 9.:
IL-20 blockade alleviates paclitaxel-induced mechanical allodynia in a mouse breast tumor xenograft model. (A) Inhibiting IL-20 signaling retarded breast cancer progression in a murine 4T-1 tumor xenograft model. Right: quantitative analyses of breast tumor volume on day 20 post-inoculation (n = 5/group). ***P < 0.001 vs vehicle group, by two-way ANOVA. (B) Representative procedure to assess cancer- or paclitaxel-induced mechanical allodynia in tumor-bearing mice. Paw withdrawal threshold to mechanical stimulus, as measured using a von Frey filament, in the ipsilateral or contralateral hind paw of tumor-bearing mice. (C) After paclitaxel treatment, tumor-bearing mice exhibited hypersensitivity to mechanical stimuli both in the ipsilateral or contralateral hind paw. Blocking (D) IL-20 and (E) IL-20 receptors prevented paclitaxel-induced mechanical allodynia but did not alleviate cancer-induced pain. Right: quantification of mechanical withdrawal threshold on day 16 after 4T-1 breast tumor cell injection. Data are presented as the mean ± SEM. #P < 0.05; ##P < 0.01; ###P < 0.001 vs ipsilateral side; ***P < 0.001 paclitaxel group, by two-way ANOVA. ANOVA, analysis of variance.

We further explored whether the anti-IL-20 mAb attenuates paclitaxel-induced mechanical hyperalgesia in 4T-1 tumor-bearing mice by mimicking the conditions cancer patients experience after chemotherapy (Fig. 9B). Control animals exhibited a lower mechanical threshold on the ipsilateral but not contralateral hind paw, indicating that tumor-bearing mice developed cancer-associated pain (P < 0.001; Fig. 9C). Compared to that in controls, tumor-bearing mice treated with paclitaxel demonstrated increased sensitivity to mechanical stimuli both on the ipsilateral and contralateral hind paw, demonstrating the toxic effect of paclitaxel on mechanical sensation (Fig. 9C). Notably, anti-IL-20 or anti-IL-20R1 mAb treatment alleviated paclitaxel-induced mechanical allodynia but did not protect tumor-bearing mice from cancer-induced mechanical hypersensitivity (Figs. 9D and E).

4. Discussion

This study highlights the novel role of the potent proinflammatory cytokine IL-20 in chemotherapy-induced peripheral neurotoxicity. We showed that IL-20 is critical for the pathological development of paclitaxel-induced neuropathy. This conclusion is supported by the following evidence, as illustrated in Figure 10. (1) Serum expression levels of IL-20 were found to be closely associated with the clinical outcome of paclitaxel-induced neuropathy. (2) Although it is difficult to discriminate the increase of IL-20 levels is due to cancer itself or the chemotherapy in the human study, paclitaxel induced the fluctuation of serum IL-20 levels in mice without carrying tumor xenograft. (3) Targeting IL-20 signaling with a monoclonal antibody in vivo alleviated CIPN by suppressing inflammatory cascades and immune cell migration. (4) Genetic deletion of the IL-20 receptor prevented paclitaxel-induced sensory and motor deficits. (5) Paclitaxel was found to upregulate IL-20, which resulted in the dysregulation of Ca2+ homeostasis and contributed to chemotherapy-induced neurotoxicity. This is the first study to show the function of IL-20 in paclitaxel-induced neurotoxicity, as evidenced by cell-line in vitro studies, mouse models, and the analyses of clinical samples. Therefore, anti-IL-20 and anti-IL-20R1 antibodies might represent potent therapeutics for the clinical prevention and treatment of paclitaxel-induced neuropathy. However, since IL-20R1 is also shared by IL-24, which is a tumor suppressor, an anti-IL-20 antibody might be a better option for the treatment of neuropathy after chemotherapy.

Figure 10.
Figure 10.:
The involvement of IL-20 in paclitaxel-induced pathophysiological alterations. Top: normal control. After paclitaxel treatment, increased IL-20 levels were accompanied by proinflammatory cascade activation, SOCE channels sensitization, and peripheral nerve damage (middle panel). However, targeting IL-20 signaling with the monoclonal antibodies limited the activation of inflammatory cascades, attenuated the sensitization of the SOCE channels, and subsequentially alleviated chemotherapy-induced neurotoxicity (bottom). SOCE, store-operated calcium entry.

Neuroinflammation is a hallmark of CIPN pathogenesis. Excessive inflammation in damaged nervous tissue after chemotherapy contributes to the initiation and maintenance of neuropathy.27 IL-20 has been reported as an autocrine factor that stimulates dendritic cell activation and subsequently regulates proinflammatory cytokines and chemokines release.20,35 This study is to test the hypothesis that modulating proinflammatory mediators by targeting IL-20 might attenuate the progression of CIPN. We showed that increased IL-20 levels after chemotherapy were accompanied by proinflammatory cascade activation, such as TNF-α, IL‐1β, and MCP-1, which are important mediators in the development of CIPN. The inhibition of IL-20 signaling by neutralizing antibodies or receptor-blocking antibodies attenuated the activation of inflammatory cascades provoked by paclitaxel treatment. Supporting information shown as follows: first, chemotherapy-induced IL-20 levels were elevated in a time-dependent manner, which reached its peak level on day 2 after single paclitaxel treatment. Furthermore, serum levels of IP-10, MIG chemokines, and proinflammatory TNF-α, IL-1β, and IL-20p70 cytokines were increased 1 week after chemotherapy. In addition, the results from the in vitro study also indicated that IL-20 potentiated paclitaxel-induced proinflammatory cytokines in DRG and macrophage cells. However, the blockade of IL-20 signaling with neutralizing antibodies or receptor antibodies limited the activation of inflammatory cascades. We thus demonstrated that IL-20 functions as a pleiotropic inflammatory cytokine and plays a role in modulating the production of proinflammatory cytokines and chemokines after chemotherapy.

Paclitaxel was found to increase macrophage infiltration and elevate proinflammatory cytokine production, which caused damage to the DRG or distal nerve endings, leading to neuropathy. Increasing evidence demonstrates that different macrophage phenotypes, either proinflammatory M1 or immunosuppressive M2, generate disparate effects on axon terminals.3,19 Activated M1 macrophages significantly reduce, whereas M2 macrophages facilitate neurite extension in DRG neurons. Previous studies revealed that paclitaxel could drive macrophage polarization into the M1 phenotype and stimulate macrophages to express high levels of NO, TNF-α, IL-1β, and the chemokine MCP-1.10,27 In this study, we found that paclitaxel treatment enhanced M1 macrophage population and triggered a proinflammatory cascade comprising IL-20, TNF-α, and MCP-1 in DRG. Notably, blocking the activation of IL-20 or IL-20R1 repressed paclitaxel-induced M1 macrophage infiltration in mouse DRG. Moreover, the inflammatory cytokines released from activated macrophages also modulated neurite outgrowth. We further confirmed that IL-20 is a key modulator of the progression of paclitaxel-induced neuron degeneration. In our studies, paclitaxel or IL-20 alone exasperated neurite loss. Furthermore, IL-20 potentiated the neurotoxic effect of paclitaxel on DRG neurite length. Importantly, the blockade of IL-20 with an IL-20 antibody improved sciatic nerve injury and restored intraepidermal nerve loss provoked by paclitaxel in vivo. These findings support the notion that IL-20 plays a critical role in the pathogenesis of paclitaxel-induced M1 macrophage infiltration and neurite degeneration.

Accumulating evidence suggests that molecular components of store-operated Ca2+ channels are involved in modulating the pathological process of neuropathic and inflammatory pain.30 An amplification of SOCE activity and the aggregation of store-operated Ca2+ channels were found to be enhanced by inflammatory cytokines. Moreover, pharmacologic inhibition of SOCE activity alleviates mechanical allodynia and relieves inflammatory pain.2,30 In this study, IL-20 provoked a transient calcium response in Ca2+-free medium and induced [Ca2+]i increases upon Ca2+ restoration. Similarly, paclitaxel sensitizes SOCE activity through the aggregation of store-operated Ca2+ channels, including STIM1 and Orai1 in DRG sensory neurons, which was found to be reversed by an IL-20 antibody. Besides store-operated Ca2+ channels, proinflammatory modulators also contribute to sensitization to neuropathy by activating TRP channels.1,2 These function as modulators of intracellular Ca2+ signaling and are involved in mediating different sensations, including the sensations of pain, temperature, and chemical and mechanical pressure.5 TRPC1, TRPA1, and TRPV4 have been reported to regulate SOCE and are also involved in paclitaxel-induced neurotoxicity.1,9,33 These TRP channels are expressed in DRG neurons and are essential for the nociceptive response to a mechanical or thermal stimulus after paclitaxel treatment. Pharmacological inhibition of the TRP channels, including TRPC1, TRPA1, TRPV1, or TRPV4, can attenuate paclitaxel-evoked mechanical allodynia.1,9,33 We found that TRPC1, TRPA1, TRPV1, and TRPV4 channels are activated in the DRG of paclitaxel-treated animals. More importantly, IL-20 or IL-20R1 mAb treatment attenuated the paclitaxel-induced sensitization of the TRP channels. These findings suggest that paclitaxel induces IL-20 production, activates SOCE channels, and sensitizes TRP channels, which results in the development of CIPN.

In the animal models of neuropathy, unobserved heterogeneity, including intersubject variability, often contributes to the data variations.34 To reduce individual heterogeneity (or bias), ratio measurements are not uncommonly used in the tests of heat pain or von Frey filaments to adjust for biologically relevant individual difference.28,31,36 Consequently, by considering the intersubject variability, the animal models of CIPN are measured at the ratio level (ie, measure/baseline measurement) for each given laboratory mouse to control the unmeasured underlying characteristics of the animal in this study.

In summary, we demonstrate that IL-20 is a potent modulator of the proinflammatory response after paclitaxel treatment. Our study provides valuable insights into potential therapeutic interventions to suppress paclitaxel-induced neuroinflammatory pain by blocking IL-20 signaling. An anti-IL-20 human mAb has been used in clinical trials for rheumatoid arthritis and psoriasis, and its safety and biocompatibility are well documented.26,37 Regarding the unmet needs of CIPN patients, our humanized IL-20 antibody showed promise as a potent therapeutics to prevent or treat paclitaxel-induced neurotoxicity in the near future.

Conflict of interest statement

The authors have no conflicts of interest to declare.

Appendix A. Supplemental digital content

Supplemental digital content associated with this article can be found online at http://links.lww.com/PAIN/A961.

Acknowledgments

The authors thank Miss Hsiao-Tzu Chang for her technical assistance. The authors appreciate the technical services provided by the “Bioimaging Core Facility of the National Core Facility for Biopharmaceuticals, Ministry of Science and Technology, Taiwan.” This work was supported by the Ministry of Science and Technology (MOST 104-2320-B-006-015-MY3, MOST 107-2319-B-006-001 and MOST 108-2319-B-006-001 to M.-R. Shen and MOST 107-2823-8-006-002 to M.-S. Chang), National Health Research Institutes (NHRI-108A1-CACO-02181811), the Ministry of Health Welfare (MOHW107-TDU-B-211-114018, MOHW108-TDU-B-211-124018), National Cheng Kung University, and National Cheng Kung University Hospital, Taiwan.

References

[1]. Alessandri-Haber N, Dina OA, Chen X, Levine JD. TRPC1 and TRPC6 channels cooperate with TRPV4 to mediate mechanical hyperalgesia and nociceptor sensitization. J Neurosci 2009;29:6217–28.
[2]. Basso L, Altier C. Transient receptor potential channels in neuropathic pain. Curr Opin Pharmacol 2017;32:9–15.
[3]. Block ML, Zecca L, Hong JS. Microglia-mediated neurotoxicity: uncovering the molecular mechanisms. Nat Rev Neurosci 2007;8:57–69.
[4]. Boehmerle W, Huehnchen P, Peruzzaro S, Balkaya M, Endres M. Electrophysiological, behavioral and histological characterization of paclitaxel, cisplatin, vincristine and bortezomib-induced neuropathy in C57Bl/6 mice. Sci Rep 2014;4:6370.
[5]. Bourinet E, Altier C, Hildebrand ME, Trang T, Salter MW, Zamponi GW. Calcium-permeable ion channels in pain signaling. Physiol Rev 2014;94:81–140.
[6]. Burmeister AR, Johnson MB, Marriott I. Murine astrocytes are responsive to the pro-inflammatory effects of IL-20. Neurosci Lett 2019;708:134334.
[7]. Chen LH, Sun YT, Chen YF, Lee MY, Chang LY, Chang JY, Shen MR. Integrating image-based high-content screening with mouse models identifies 5-hydroxydecanoate as a neuroprotective drug for paclitaxel-induced neuropathy. Mol Cancer Ther 2015;14:2206–14.
[8]. Chen WY, Chang MS. IL-20 is regulated by hypoxia-inducible factor and up-regulated after experimental ischemic stroke. J Immunol 2009;182:5003–12.
[9]. Chen Y, Yang C, Wang ZJ. Proteinase-activated receptor 2 sensitizes transient receptor potential vanilloid 1, transient receptor potential vanilloid 4, and transient receptor potential ankyrin 1 in paclitaxel-induced neuropathic pain. Neuroscience 2011;193:440–51.
[10]. Chen YF, Chen LH, Yeh YM, Wu PY, Chen YF, Chang LY, Chang JY, Shen MR. Minoxidil is a potential neuroprotective drug for paclitaxel-induced peripheral neuropathy. Sci Rep 2017;7:45366.
[11]. Chisholm SP, Cervi AL, Nagpal S, Lomax AE. Interleukin-17A increases neurite outgrowth from adult postganglionic sympathetic neurons. J Neurosci 2012;32:1146–55.
[12]. Chiu YS, Wei CC, Lin YJ, Hsu YH, Chang MS. IL-20 and IL-20R1 antibodies protect against liver fibrosis. Hepatology 2014;60:1003–14.
[13]. Hosoi T, Wada S, Suzuki S, Okuma Y, Akira S, Matsuda T, Nomura Y. Bacterial endotoxin induces IL-20 expression in the glial cells. Brain Res Mol Brain Res 2004;130:23–9.
[14]. Hsu YH, Chen WY, Chan CH, Wu CH, Sun ZJ, Chang MS. Anti-IL-20 monoclonal antibody inhibits the differentiation of osteoclasts and protects against osteoporotic bone loss. J Exp Med 2011;208:1849–61.
[15]. Hsu YH, Hsing CH, Li CF, Chan CH, Chang MC, Yan JJ, Chang MS. Anti-IL-20 monoclonal antibody suppresses breast cancer progression and bone osteolysis in murine models. J Immunol 2012;188:1981–91.
[16]. Hu LY, Mi WL, Wu GC, Wang YQ, Mao-Ying QL. Prevention and treatment for chemotherapy-induced peripheral neuropathy: therapies based on CIPN mechanisms. Curr Neuropharmacol 2019;17:184–96.
[17]. Huang HW, Jou IM, Wang CK, Chen PY, Wang WC, Lin CC. Power spectral analyses of index finger skin blood perfusion in carpal tunnel syndrome and diabetic polyneuropathy. Exp Diabetes Res 2011;2011:465910.
[18]. Hwang HW, Wang WC, Lin CC. The influences of inter-trial interval on the thermal and thermal pain thresholds in quantitative sensory testing. Acta Neurol Taiwan 2012;21:152–7.
[19]. Kigerl KA, Gensel JC, Ankeny DP, Alexander JK, Donnelly DJ, Popovich PG. Identification of two distinct macrophage subsets with divergent effects causing either neurotoxicity or regeneration in the injured mouse spinal cord. J Neurosci 2009;29:13435–44.
[20]. Kragstrup TW, Andersen T, Heftdal LD, Hvid M, Gerwien J, Sivakumar P, Taylor PC, Senolt L, Deleuran B. The IL-20 cytokine family in rheumatoid arthritis and spondyloarthritis. Front Immunol 2018;9:2226.
[21]. Lauria G, Dacci P, Lombardi R, Cazzato D, Porretta-Serapiglia C, Taiana M, Sassone J, Dalla Bella E, Rinaldo S, Lettieri C, Eleopra R, Devigili G. Side and time variability of intraepidermal nerve fiber density. Neurology 2015;84:2368–71.
[22]. Lavoie Smith EM, Barton DL, Qin R, Steen PD, Aaronson NK, Loprinzi CL. Assessing patient-reported peripheral neuropathy: the reliability and validity of the European Organization for Research and Treatment of Cancer QLQ-CIPN20 Questionnaire. Qual Life Res 2013;22:2787–99.
[23]. Lees JG, Makker PG, Tonkin RS, Abdulla M, Park SB, Goldstein D, Moalem-Taylor G. Immune-mediated processes implicated in chemotherapy-induced peripheral neuropathy. Eur J Cancer 2017;73:22–9.
[24]. Li Y, Zhang H, Kosturakis AK, Cassidy RM, Zhang H, Kennamer-Chapman RM, Jawad AB, Colomand CM, Harrison DS, Dougherty PM. MAPK signaling downstream to TLR4 contributes to paclitaxel-induced peripheral neuropathy. Brain Behav Immun 2015;49:255–66.
[25]. Li Y, Zhang H, Zhang H, Kosturakis AK, Jawad AB, Dougherty PM. Toll-like receptor 4 signaling contributes to Paclitaxel-induced peripheral neuropathy. J Pain 2014;15:712–25.
[26]. Lundblad MS, Overgaard RV, Gothberg M, Fjording MS, Watson E. Clinical pharmacokinetics of the anti-interleukin-20 monoclonal antibody NNC0109-0012 in healthy volunteers and patients with psoriasis or rheumatoid arthritis. Adv Ther 2015;32:228–38.
[27]. Makker PG, Duffy SS, Lees JG, Perera CJ, Tonkin RS, Butovsky O, Park SB, Goldstein D, Moalem-Taylor G. Characterisation of immune and neuroinflammatory changes associated with chemotherapy-induced peripheral neuropathy. PLoS One 2017;12:e0170814.
[28]. Mo M, Erdelyi I, Szigeti-Buck K, Benbow JH, Ehrlich BE. Prevention of paclitaxel-induced peripheral neuropathy by lithium pretreatment. FASEB J 2012;26:4696–709.
[29]. Montague K, Malcangio M. The therapeutic potential of monocyte/macrophage manipulation in the treatment of chemotherapy-induced painful neuropathy. Front Mol Neurosci 2017;10:397.
[30]. Munoz F, Hu H. The role of store-operated calcium channels in pain. Adv Pharmacol 2016;75:139–51.
[31]. Murthy SE, Loud MC, Daou I, Marshall KL, Schwaller F, Kuhnemund J, Francisco AG, Keenan WT, Dubin AE, Lewin GR, Patapoutian A. The mechanosensitive ion channel Piezo2 mediates sensitivity to mechanical pain in mice. Sci Transl Med 2018;10:eaat9897.
[32]. Myles IA, Fontecilla NM, Valdez PA, Vithayathil PJ, Naik S, Belkaid Y, Ouyang W, Datta SK. Signaling via the IL-20 receptor inhibits cutaneous production of IL-1beta and IL-17A to promote infection with methicillin-resistant Staphylococcus aureus. Nat Immunol 2013;14:804–11.
[33]. Qi Z, Wang Y, Zhou H, Liang N, Yang L, Liu L, Zhang W. The central analgesic mechanism of YM-58483 in attenuating neuropathic pain in rats. Cell Mol Neurobiol 2016;36:1035–43.
[34]. Rice ASC, Finnerup NB, Kemp HI, Currie GL, Baron R. Sensory profiling in animal models of neuropathic pain: a call for back-translation. PAIN 2018;159:819–24.
[35]. Rich BE, Kupper TS. Cytokines: IL-20 - a new effector in skin inflammation. Curr Biol 2001;11:R531–534.
[36]. Sandi C, Pinto RM, Al-Mahdawi S, Ezzatizadeh V, Barnes G, Jones S, Rusche JR, Gottesfeld JM, Pook MA. Prolonged treatment with pimelic o-aminobenzamide HDAC inhibitors ameliorates the disease phenotype of a Friedreich ataxia mouse model. Neurobiol Dis 2011;42:496–505.
[37]. Senolt L, Leszczynski P, Dokoupilova E, Gothberg M, Valencia X, Hansen BB, Canete JD. Efficacy and safety of anti-interleukin-20 monoclonal antibody in patients with rheumatoid arthritis: a randomized phase IIa trial. Arthritis Rheumatol 2015;67:1438–48.
[38]. Sorensen L, Molyneaux L, Yue DK. The relationship among pain, sensory loss, and small nerve fibers in diabetes. Diabetes Care 2006;29:883–7.
[39]. Staff NP, Grisold A, Grisold W, Windebank AJ. Chemotherapy-induced peripheral neuropathy: a current review. Ann Neurol 2017;81:772–81.
[40]. Uceyler N, Kafke W, Riediger N, He L, Necula G, Toyka KV, Sommer C. Elevated proinflammatory cytokine expression in affected skin in small fiber neuropathy. Neurology 2010;74:1806–13.
[41]. Wei CC, Chen WY, Wang YC, Chen PJ, Lee JY, Wong TW, Chen WC, Wu JC, Chen GY, Chang MS, Lin YC. Detection of IL-20 and its receptors on psoriatic skin. Clin Immunol 2005;117:65–72.
[42]. Wei CC, Hsu YH, Li HH, Wang YC, Hsieh MY, Chen WY, Hsing CH, Chang MS. IL-20: biological functions and clinical implications. J Biomed Sci 2006;13:601–12.
[43]. Wolk K, Haugen HS, Xu W, Witte E, Waggie K, Anderson M, Vom Baur E, Witte K, Warszawska K, Philipp S, Johnson-Leger C, Volk HD, Sterry W, Sabat R. IL-22 and IL-20 are key mediators of the epidermal alterations in psoriasis while IL-17 and IFN-gamma are not. J Mol Med (Berl) 2009;87:523–36.
[44]. Zhang H, Boyette-Davis JA, Kosturakis AK, Li Y, Yoon SY, Walters ET, Dougherty PM. Induction of monocyte chemoattractant protein-1 (MCP-1) and its receptor CCR2 in primary sensory neurons contributes to paclitaxel-induced peripheral neuropathy. J Pain 2013;14:1031–44.
[45]. Zhang H, Li Y, de Carvalho-Barbosa M, Kavelaars A, Heijnen CJ, Albrecht PJ, Dougherty PM. Dorsal root ganglion infiltration by macrophages contributes to paclitaxel chemotherapy-induced peripheral neuropathy. J Pain 2016;17:775–86.
Keywords:

IL-20; Paclitaxel; Neuroprotection; Chemotherapy-induced neuropathy

Supplemental Digital Content

© 2020 International Association for the Study of Pain