Introduction

Chemotherapy-induced peripheral neuropathy (CIPN) remains a significant clinical challenge in cancer treatment, affecting up to 80% of patients receiving platinum-based agents such as cisplatin1. Characterized by mechanical allodynia, paresthesia, and sensory deficits in the distal extremities, CIPN not only impairs the quality of life but also often necessitates dose reductions or treatment discontinuation, thereby compromising the efficacy of cancer therapies2. Despite its high prevalence, the exact mechanisms underlying cisplatin-induced CIPN remain incompletely understood, and current treatments largely rely on systemic medications (e.g., gabapentin, duloxetine), which frequently cause central side effects3.

The GABAergic system plays a pivotal role in nociceptive modulation, with GABA serving as the primary inhibitory neurotransmitter in both central and peripheral tissues. GABAB receptors, metabotropic receptors that are widely expressed in sensory neurons and Schwann cells, regulate pain signaling by inhibiting neurotransmitter release and hyperpolarizing nociceptive terminals4. Activation of GABAB receptors by baclofen has shown analgesic effects in preclinical models of neuropathic pain; however, systemic administration is often limited by sedation, muscle relaxation, and cognitive impairment5. Given that CIPN symptoms are localized to the cutaneous regions2, targeting peripheral GABAB receptors through topical administration could potentially avoid central side effects, yet this approach remains underexplored.

Recent studies suggest that regional differences in GABAergic function may contribute to neuropathic pain, with peripheral and central GABAergic systems exhibiting distinct responses to injury6. In the context of cisplatin-induced CIPN, however, the spatial dynamics of GABA levels, GABA-synthesizing enzyme (GAD) expression, and GABAB receptor regulation remain unclear. In this study, we systematically characterized cisplatin-induced behavioral changes (mechanical, cold, and heat sensitivity) and regional GABAergic imbalance in the DRG and hind paw skin. Additionally, we evaluated the efficacy of intraplantar baclofen in reversing cisplatin-induced mechanical allodynia, testing the hypothesis that peripheral GABAB receptor activation could serve as a targeted therapy for CIPN.

Methods

Male Sprague-Dawley rats (6–8 weeks old, 100–120 g) were obtained from the Laboratory Animal Center of Yanbian University and housed under controlled conditions (22 ± 1 °C, 12 h light/dark cycle, 55 ± 5% humidity), with ad libitum access to food and water. Environmental enrichment (polyvinyl chloride tubes, wooden blocks) was provided to minimize stress. This study used only male rats to eliminate potential confounding effects of sex hormones on pain behavior and GABAergic signaling. All procedures were approved by the Institutional Animal Care and Use Committee (Approval No. YD20231212001) and conducted in accordance with ARRIVE guidelines. All experimental procedures involving animals were additionally performed in strict accordance with the Guide for the Care and Use of Laboratory Animals (National Institutes of Health, USA) and the editorial policies of Scientific Reports regarding experimental subjects. Euthanasia procedures conformed to the AVMA Guidelines for the Euthanasia of Animals (2020).

Cisplatin-induced Neuropathy Model. Cisplatin (Tocris, UK) was dissolved in 0.4% dimethyl sulfoxide (DMSO)/saline. A well-validated protocol for inducing CIPN was used7. Rats received intraperitoneal (i.p.) injections of cisplatin (2 mg/kg) or vehicle (0.4% DMSO/saline) once weekly for 4 consecutive weeks. This regimen was selected as it reliably produces robust mechanical allodynia without causing severe systemic toxicity or kidney failure.

Drug Administration. Baclofen (Tocris, UK) was dissolved in 0.9% saline (0.3, 1, or 3 µg/20 µL) and injected intraplantarly into the hind paw plantar surface using a 30-gauge needle (2 mm depth); controls received saline. Baclofen doses were selected based on pilot studies and previous literature demonstrating efficacy in peripheral pain models without motor side effects at these low, locally administered doses8.

Behavioral assessments were performed by a blinded experimenter. Mechanical allodynia was assessed using von Frey filaments (0.4–15 g) with the up-down method to determine the 50% paw withdrawal threshold (PWT)9. Cold allodynia was evaluated using the acetone test (0.05 mL), scored 0–3 (total 0–18) over 20 s10. Thermal hyperalgesia was measured as paw withdrawal latency (PWL) to radiant heat (50 mW, 20 s cutoff)10. Hind paw temperature was recorded using an infrared camera (IRIS-XP, Medicore) at a 30 cm focal distance, analyzing the mean temperature over a 1 cm² plantar region (emissivity = 0.98)11.

At 5 and 14 days post-cisplatin, rats were euthanized, and L4–L6 DRG and hind paw skin (1 cm²) were harvested. GABA levels were measured by HPLC (Agilent 1260) using a Gemini C18 column and fluorescence detection (excitation 330 nm, emission 450 nm)12. RT-PCR was performed for GAD, GABAB receptor, and β-actin (reference), with relative expression calculated via the 2^(-ΔΔCT) method13. Western blotting of skin lysates used antibodies against GABAB receptor (1:1000, Millipore) and β-actin (1:1000, Cell Signaling), with densitometry performed using ImageJ14. The original, unprocessed full-length gels for Western blotting, retaining complete loading controls and molecular weight markers, have been included in the Supplementary Information file to comply with the journal’s digital image and integrity policies.

For baclofen pretreatment, injections were administered 30 min before each cisplatin dose, with PWT measured daily for 21 days. For post-treatment, a single injection was given 14 days post-cisplatin, and PWT was measured at 15–180 min. The percentage of maximum possible effect (%MPE) was calculated as [(post-treatment PWT – baseline PWT)/(15 g – baseline PWT)] × 100. At the end of the experiment, rats were sacrificed by intraperitoneal injection of an overdose of sodium pentobarbital (150 mg/kg, Sigma-Aldrich, St. Louis, MO, USA), followed by cervical dislocation to confirm death. This procedure was approved by the Institutional Animal Care and Use Committee of Yanbian University (Approval No. YD20231212001) and conforms to the AVMA Guidelines for the Euthanasia of Animals (2020).

Data are presented as mean ± SEM. Statistical analyses were performed using two-way repeated measures ANOVA (for time courses) or one-way ANOVA (for single time points) with Bonferroni post-hoc tests (p < 0.05; GraphPad Prism 9).

Results

Following cisplatin administration, the 2 mg/kg dose significantly reduced the paw withdrawal threshold (PWT) starting at day 4 (5.2 ± 0.6 g vs. baseline 12.8 ± 0.5 g, p < 0.001), with persistence through day 21 (Fig. 1A). Lower doses (0.5 and 1 mg/kg) had no effect on PWT (p > 0.05 vs. baseline). In contrast, cold allodynia (Fig. 1B) and heat hyperalgesia (Fig. 1C) remained unchanged, with acetone test scores and paw withdrawal latency (PWL) showing no significant differences from baseline (p > 0.05). The detailed time-course data of these behavioral assessments, including individual animal data points and statistical significance analysis for each time point, are presented in Supplementary Fig. 1. Infrared imaging and quantitative analysis revealed a significant reduction in hind paw temperature at day 5 post-cisplatin (27.3 ± 0.4 °C vs. vehicle 29.0 ± 0.2 °C, p < 0.05), which normalized by day 7 (28.8 ± 0.3 °C, p > 0.05 vs. vehicle) (Fig. 2A, B).

Molecular analyses revealed a region-specific GABAergic imbalance in response to cisplatin. In the dorsal root ganglion (DRG), GABA levels were significantly increased at 14 days post-cisplatin (2.15 ± 0.18 nmol/mg tissue vs. naïve 1.23 ± 0.10 nmol/mg, p < 0.05; Fig. 3A), while hind paw skin exhibited decreased GABA levels at both 5 and 14 days (day 5: 0.82 ± 0.07 nmol/mg; day 14: 0.68 ± 0.06 nmol/mg vs. naïve 1.35 ± 0.11 nmol/mg, p < 0.05; Fig. 3B). RT-PCR showed that glutamate decarboxylase (GAD) mRNA was upregulated in DRG at 14 days (2.4 ± 0.3-fold vs. naïve, p < 0.05; Fig. 3C), but was undetectable in hind paw skin. Conversely, GABAB receptor mRNA in hind paw skin was downregulated at 14 days (0.4 ± 0.1-fold vs. naïve, p < 0.05; Fig. 3D), with a corresponding reduction in protein expression (0.3 ± 0.1-fold vs. naïve, p < 0.05; Fig. 3E).

Intraplantar baclofen pretreatment and post-treatment alleviated cisplatin-induced mechanical allodynia (Fig. 4). Pretreatment with baclofen (1 and 3 µg) delayed the onset of allodynia, with the 3 µg dose significantly increasing PWT on day 4 (8.7 ± 0.6 g vs. vehicle + cisplatin 5.2 ± 0.6 g, p < 0.05) and day 5 (8.1 ± 0.5 g vs. 4.9 ± 0.5 g, p < 0.05; Fig. 4A). Post-treatment with baclofen at 14 days post-cisplatin exerted a dose-dependent effect, with PWT peaking at 60 min (3 µg: 10.2 ± 0.7 g vs. vehicle 4.3 ± 0.5 g, p < 0.05; Fig. 4B, C). The 3 µg dose maintained efficacy through 120 min (8.5 ± 0.6 g, p < 0.05 vs. vehicle), with the percentage of maximum possible effect (%MPE) shown in Fig. 4D.

Notably, intraplantar baclofen (3 µg) administered 14 days post-cisplatin had no significant effect on hind paw temperature measured at 1 h post-injection, compared to the saline-treated chemotherapy-induced peripheral neuropathy (CIPN) group (p > 0.05; Fig. 5). This indicates that the antiallodynic effects of baclofen are independent of changes in peripheral skin temperature.

Fig. 1
Fig. 1The alternative text for this image may have been generated using AI.
Full size image

Cisplatin induces mechanical allodynia but not cold allodynia or heat hyperalgesia. (A) Mechanical paw withdrawal threshold (PWT, g) measured using von Frey filaments following administration of cisplatin (2 mg/kg) or vehicle. (B) Cold allodynia score assessed via the acetone test. (C) Paw withdrawal latency (PWL, s) in response to radiant heat. Arrows in all panels indicate the timing of cisplatin/vehicle intraperitoneal injection (Days 0, 7, 14, 21). Baseline (B) represents measurements prior to the first cisplatin administration. Data are presented as mean ± SEM (n = 8/group). Two-way repeated measures (RM) ANOVA with Bonferroni post-hoc test revealed a significant main effect of cisplatin treatment on PWT [F(1,14) = 31.26, p < 0.001, partial η²=0.46] and no significant effects on cold score [F(1,14) = 2.18, p = 0.16, partial η²=0.07] or PWL [F(1,14) = 1.83, p = 0.19, partial η²=0.06]. *p < 0.001 vs. Vehicle group (Cohen’s d = 1.82 for PWT at Day 21). Abbreviations: PWT, 50% paw withdrawal threshold; PWL, paw withdrawal latency.

Discussion

This study demonstrates that cisplatin-induced peripheral neuropathy (CIPN) is characterized by dose-dependent mechanical allodynia, acute hind paw temperature reduction, and a region-specific GABAergic imbalance—specifically, increased GABA and GAD in the DRG and decreased GABA and GABAB receptor expression in hind paw skin. Intraplantar baclofen alleviated mechanical allodynia via peripheral GABAB receptor activation, with no effect on temperature, supporting its potential as a topical therapy for CIPN.

Cisplatin-induced mechanical allodynia is consistent with clinical reports of CIPN, where mechanical hypersensitivity is a hallmark symptom1. The dose dependence suggests that cisplatin neurotoxicity is cumulative, aligning with clinical observations that higher cumulative doses correlate with more severe CIPN15. The preferential effect on mechanical sensitivity over cold or heat sensitivity points to a primary involvement of large myelinated Aβ fibers, which are known to be damaged by platinum derivatives and mediate mechanical sensation16.

The acute reduction in hind paw temperature observed at day 5 post-cisplatin likely reflects vascular dysfunction, as cisplatin disrupts endothelial cell function and reduces peripheral perfusion17. However, the normalization of temperature by day 7, despite the persistence of allodynia, suggests that vascular changes are not the primary drivers of mechanical hypersensitivity. This is consistent with studies linking CIPN to direct neurotoxicity rather than vascular changes18.

The regional GABAergic imbalance observed in this study is a key finding. Decreased GABA and GABAB receptors in hind paw skin likely impair peripheral inhibitory control, while the increased GABA and GAD expression in the DRG likely represent a compensatory central response. Peripheral GABA, synthesized by Schwann cells19, normally inhibits nociceptors via GABAB receptors, thereby limiting the activation of mechanosensitive channels such as Piezo220. The reduction in GABA and GABAB receptor expression in the skin would disinhibit these channels, exacerbating mechanical allodynia. Conversely, the increased DRG GABA and GAD expression—GAD being the rate-limiting enzyme in GABA synthesis21—likely represents an attempt to enhance central inhibitory signaling to compensate for peripheral GABAergic dysfunction. While this study focused on the GABAB receptor, it is important to note that GABAA receptors are also present in the periphery and contribute to pain modulation. Future studies should investigate potential cross-talk or complementary roles of GABAA and GABAB receptors in CIPN.

The regional GABAergic imbalance observed in this study is a key finding. Decreased GABA and GABAB receptors in hind paw skin likely impair peripheral inhibitory control, while the increased GABA and GAD expression in the DRG likely represent a compensatory central response. Peripheral GABA, synthesized by Schwann cells22, normally inhibits nociceptors via GABAB receptors, thereby limiting the activation of mechanosensitive channels such as Piezo223. The reduction in GABA and GABAB receptor expression in the skin would disinhibit these channels, exacerbating mechanical allodynia. Conversely, the increased DRG GABA and GAD expression—GAD being the rate-limiting enzyme in GABA synthesis24—likely represents an attempt to enhance central inhibitory signaling to compensate for peripheral GABAergic dysfunction. While our findings highlight the crucial role of GABAB receptors, it is noteworthy that GABAA receptors are also expressed in peripheral nociceptive pathways and contribute to pain modulation. However, GABAA-mediated analgesia typically involves fast, ionotropic inhibition that may be less effective in chronic neuropathic states compared to the sustained metabotropic signaling of GABAB receptors. This mechanistic distinction may explain why systemic benzodiazepines (GABAA-positive allosteric modulators) often show limited efficacy in neuropathic pain25, whereas targeted GABAB activation, as demonstrated here with intraplantar baclofen, produces more robust antiallodynic effects in our CIPN model. Future studies should investigate potential cross-talk or complementary roles of GABAA and GABAB receptors in CIPN to fully elucidate the complex GABAergic regulation of neuropathic pain26.

The efficacy of intraplantar baclofen confirms that residual peripheral GABAB receptors remain functional despite downregulation. The dose-dependent post-treatment effect (peaking at 60 min) and pretreatment delay of onset suggest a direct peripheral mechanism of action, as intraplantar administration minimizes systemic absorption27. The intraplantar doses of baclofen used in this study (0.3–3 µg) are several orders of magnitude lower than systemic doses required for central effects. Throughout our behavioral observations, no signs of motor impairment, sedation, or other adverse effects were noted, supporting the favorable safety profile of this localized administration route. The lack of effect on hind paw temperature further supports the notion that baclofen’s analgesic action is independent of vascular changes.

This study has several limitations that should be acknowledged. First, the experiments were conducted exclusively in male rats. While CIPN mechanisms are largely conserved, sex differences in GABAergic signaling and pain perception exist28, and future studies should include both sexes. Second, while our data strongly suggest the involvement of GABAB receptors, the absence of a selective GABAB receptor antagonist (e.g., SCH50911) to block the effect of baclofen means that receptor specificity, while highly likely, is not definitively proven. Third, although the concentration of DMSO (0.4%) in our vehicle is low and commonly used, we cannot entirely rule out its potential, albeit minimal, effects on membranes and pain modulation. Finally, future studies using single-cell RNA sequencing could identify the specific cellular sources (e.g., Schwann cells, keratinocytes, sensory neurons) of the observed GABAergic changes.

In conclusion, cisplatin-induced CIPN involves region-specific GABAergic dysregulation, and peripheral GABAB activation via intraplantar baclofen alleviates mechanical allodynia. These findings suggest that topical GABAB agonists may be a promising targeted therapy for CIPN, potentially avoiding the central side effects associated with systemic treatments.

Fig. 2
Fig. 2The alternative text for this image may have been generated using AI.
Full size image

Cisplatin causes a transient reduction in hind paw temperature. (A) Representative infrared image of the hind paw, with the plantar region of interest (ROI) outlined. (B) Temporal changes in hind paw temperature (°C) following cisplatin (2 mg/kg) or vehicle administration. Data are mean ± SEM (n = 6/group). Two-way RM ANOVA with Bonferroni post-hoc test showed a significant main effect of cisplatin on hind paw temperature [F(1,10) = 9.74, p < 0.01, partial η²=0.30]. *p < 0.05 vs. Vehicle group at the same time point (Cohen’s d = 1.23 for Day 14). Abbreviation: ROI, region of interest.

Fig. 3
Fig. 3The alternative text for this image may have been generated using AI.
Full size image

Region-specific GABAergic imbalance in cisplatin-induced neuropathy. (A) GABA levels in the dorsal root ganglion (DRG). (B) GABA levels in the hind paw skin. (C) mRNA expression of glutamate decarboxylase (GAD) in the DRG. (D) mRNA expression of the GABAB receptor in the hind paw skin. (E) GABAB receptor protein expression in the hind paw skin. N = naïve rats (no treatment); 5d and 14d = 5 and 14 days after the final cisplatin (2 mg/kg) injection, respectively. Data are mean ± SEM (n = 5/group). One-way ANOVA with Bonferroni post-hoc test demonstrated significant group effects for DRG GABA levels [F(2,12) = 8.67, p < 0.01, partial η²=0.35], hind paw skin GABAB receptor mRNA [F(2,12) = 7.92, p < 0.01, partial η²=0.33], and protein expression [F(2,12) = 6.85, p < 0.01, partial η²=0.30], with no significant effects on skin GABA levels [F(2,12) = 2.31, p = 0.14, partial η²=0.08] or DRG GAD mRNA [F(2,12) = 1.97, p = 0.18, partial η²=0.07]. *p < 0.05 vs. Naïve group (Cohen’s d = 1.31 for DRG GABA at 14d; Cohen’s d = 1.25 for skin GABAB protein at 14d). Abbreviations: GABA, γ-aminobutyric acid; DRG, dorsal root ganglion; GAD, glutamate decarboxylase; mRNA, messenger RNA.

Fig. 4
Fig. 4The alternative text for this image may have been generated using AI.
Full size image

Intraplantar baclofen alleviates cisplatin-induced mechanical allodynia. (A) Pretreatment paradigm: PWT over time following intraplantar administration of baclofen (0.3, 1, 3 µg/20 µL) or saline, administered 30 min prior to each cisplatin injection. (B, C) Post-treatment paradigm: PWT over time after a single intraplantar injection of baclofen or saline at 14 days post-final cisplatin. (D) Percentage of maximum possible effect (%MPE) calculated as [(post-treatment PWT – baseline PWT)/(15 g – baseline PWT)] × 100. Data are mean ± SEM (n = 8/group). Two-way RM ANOVA with Bonferroni post-hoc test revealed a significant main effect of baclofen treatment on PWT in both pretreatment [F(3,28) = 27.53, p < 0.001, partial η²=0.42] and post-treatment [F(3,28) = 22.81, p < 0.001, partial η²=0.38] paradigms. *p < 0.05 vs. Cisplatin + Saline group (Cohen’s d = 1.63 for 3 µg baclofen at 120 min post-treatment). Abbreviations: PWT, 50% paw withdrawal threshold; %MPE, percentage of maximum possible effect.

Fig. 5
Fig. 5The alternative text for this image may have been generated using AI.
Full size image

Intraplantar baclofen does not affect hind paw temperature. (A) Hind paw temperature measured at baseline (B) and 1 h (1 h) after intraplantar injection of baclofen (3 µg/20 µL) or saline in rats with cisplatin-induced peripheral neuropathy (CIPN; 14 days post-final cisplatin). Data are mean ± SEM (n = 6/group). Two-way RM ANOVA showed no significant main effect of baclofen treatment [F(1,10) = 0.92, p = 0.36, partial η²=0.08] or treatment×time interaction [F(1,10) = 0.75, p = 0.41, partial η²=0.06] (Cohen’s d = 0.31 for baclofen vs. saline at 1 h). Abbreviation: CIPN, cisplatin-induced peripheral neuropathy.