Abstract
Keywords
Background
Chemotherapy-induced peripheral neuropathy (CIPN) is a common dose-limiting complication of paclitaxel, affecting over 60% of patients shortly after treatment, with 41% developing persistent pain ≥3 months.1–4 Long-term surveys report CIPN prevalence of 55%–75% at 12 months, with variable analgesic responses. 5 Severe CIPN can necessitate dose reduction or treatment discontinuation, highlighting the urgent need to elucidate underlying mechanisms and develop effective therapies.
Paclitaxel is a widely applied chemotherapeutic drug for solid tumors, exerting its antineoplastic effect primarily through stabilization of microtubules via enhanced tubulin polymerization.6,7 Despite its therapeutic value, paclitaxel often induces CIPN, restricting clinical application, while the mechanistic basis remains only partially clarified. Current evidence indicates that perturbations in Ca2+ homeostasis contribute significantly to paclitaxel-induced neuropathic pain (PINP). 8 Paclitaxel initiates cytosolic Ca2+ oscillations in neuronal cells through activation of the phosphoinositide pathway and InsP3R, a process dependent on its interaction with the binding protein neuronal calcium sensor-1 (NCS-1). 9 Sustained Ca2+ elevation subsequently triggers activation of calpain and caspase-3/7, leading to apoptosis. 10
Calpains constitute a family of calcium-dependent cysteine proteases,11,12 among which calpain-1 (μ-calpain, CAPN1) and calpain-2 (m-calpain, CAPN2) represent the most thoroughly characterized isoform. 13 Their distinct activities are primarily defined by differences in Ca2+ sensitivity: CAPN1 is triggered at micromolar concentrations (1–10 μM), while CAPN2 requires millimolar levels (0.3–1 Mm).14,15 Calcium signaling exerts a central influence on apoptosis, with CAPN1 and CAPN2 acting as regulators of cell survival through their divergent Ca2+ threshold. 16 Beyond cell death pathways, both isoforms demonstrate dual actions in neuroinflammation, either intensifying proinflammatory cascades or, under specific conditions, inducing anti-inflammatory mechanisms that support neuronal repair.17,18
Evidence from earlier studies revealed temporally distinct alterations of CAPN1 and CAPN2 expression in dorsal root ganglia (DRG) in mirror-image pain rats: CAPN2 levels increased markedly within 0.5 h of injury, whereas CAPN1 was significantly reduced at 1 h. 19 In contrast, analysis of the spinal dorsal horn (SDH) demonstrated a pronounced rise in CAPN2 protein expression, while CAPN1 remained unchanged. 20 During tumorigenesis, there is often enhanced expression of calpain within tumor tissues,21,22 and the latter can confer resistance to the therapeutic effects of chemotherapeutic drugs. 23 However, whether it is a significant cause of PIPN remains not entirely clear. Recent studies demonstrated that cisplatin-induced neurotoxicity involved Sarm1 activation and calpain signaling, 24 pointing to a possible contribution of calpain to PINP.
In this study, a clinically relevant low-dose, dose-dense paclitaxel regimen (6 mg/kg, i.p., every other day, four injections) was administered to rats25–27 to assess calpain expression and activity in the DRG and spinal cord, together with the expression of pain-related molecules such as transient receptor potential V4 (TRPV4) and NCS-1. Both CAPN1 and CAPN2 proteins were markedly upregulated, accompanied by increased protease activity, as evidenced by the accumulation of spectrin breakdown products (SBDP) in the SDH. Immunofluorescence colocalization demonstrated that CAPN1 was primarily localized to neurons. Overexpression (OE) of CAPN1 in CaMKII-positive neurons of the normal spinal dorsal horn reproduced PINP, together with elevated NCS-1 levels. Collectively, the results indicate that CAPN1 may contribute to PINP through modulation of NCS-1 at the plasma membrane and activation of downstream inflammatory pathways.
Methods
Animals and ethical statement
Adult male Sprague–Dawley rats weighing 180–220 g were provided by the Laboratory Animal Center of Sun Yat-sen University (NO. SYXK [Yue] 2024-0081). Animals were maintained individually under controlled conditions (24 ± 1°C, 50%–60% relative humidity) with unrestricted access to standard chow and sterile water. A 12-h light/dark cycle was applied, with illumination from 06:00 to 18:00. All experimental procedures were approved by the Institutional Animal Care and Use Committee (IACUC) at Sun Yat-Sen University (Approval Nos.SYSU-IACUC-2024-000062) and were conducted in accordance with the guidelines of the National Institutes of Health (NIH) on animal care and ethical guidelines. Efforts were directed toward minimizing distress and reducing the number of animals used.
Paclitaxel-induced neuropathic pain model
Adult SD rats were randomly allocated into experimental groups for subsequent procedures. The paclitaxel group were administered intraperitoneal injections of paclitaxel (6 mg/kg, Sichuan Huiyu Pharmaceutical Co., Ltd., Neijiang, China) on days 1, 3, 5, and 7, yielding a cumulative dose of 24 mg/kg. 28 Control animals received an equal volume of saline intraperitoneally on the same schedule.
50% paw withdrawal threshold test
Mechanical sensitivity was determined using the up–down paradigm. 29 Prior to baseline assessment, rats were habituated to the testing apparatus for 15–20 min per session, repeated 2–3 times during the preceding week, and subsequently acclimated for 15–20 min before each evaluation. Animals were individually placed in acrylic enclosures with wire mesh floors. A series of von Frey filaments with logarithmically incremental stiffness (0.4, 0.6, 1.0, 2.0, 4.0, 6.0, 8.0, and 15.0 g) were applied perpendicularly to the plantar surface of the hind paw for 6–8 s, with a minimum interstimulus interval of 5 min between successive stimulations. Testing began with the 2.0 g filament. If no paw withdrawal was observed, the next higher force was applied; conversely, a lower force was used when a withdrawal response occurred. This up–down procedure was repeated until six response reversals were obtained. The 50% paw withdrawal threshold (PWT) was derived with Cell G software according to the formula: 50% PWT (g) = [10^(Xf + kδ)]/10,000, where Xf represents the logarithmic value of the final von Frey filament, k denotes a tabulated constant determined by the response sequence, and δ corresponds to the mean logarithmic interval between successive filaments.
Calpain activity assay
Calpain enzymatic activity was evaluated indirectly through measuring levels of the SBDP, which arises from calpain-mediated cleavage of the cytoskeletal protein spectrin.30,31 The accumulation of SBDP serves as a reliable biochemical indicator of calpain activation.
Western blot
Western blot analysis was conducted following the procedures detailed in our previous studies. 32 Following anesthesia with 0.4% sodium pentobarbital (40 mg/kg body weight, i.p.), L4–L6 DRG and spinal cord samples (L5 spinal segment) were collected, briefly snap-frozen in liquid nitrogen for 10 s, and homogenized in SDS lysis buffer supplemented with a protease and phosphatase inhibitor cocktail (Roche Molecular Biochemicals, Indianapolis, IN, USA). Extracted Proteins were separated with SDS-PAGE and transferred onto PVDF membranes. After blocking for 1 h at room temperature, membranes were incubated overnight at 4°C with the designated primary antibodies: CAPN1 (Rabbit, 1:1000, Abcam, ab39170), CAPN2 (Rabbit, 1:1000, Abcam, ab39165), β-actin (Rabbit, 1:10000, Proteintech, 66009-1-lg), GAPDH (Mouse, 1:3000, Novus, NB100-56875), TRPV4 (Rabbit, 1:1000, EpiZyme, p011399), NCS-1 (Rabbit, 1:1000, Thermo Fisher, PA5-92988), and SBDP (Mouse, 1:500, Sigma-Aldrich, MAB1622). Following extensive washing, membranes were exposed to appropriate secondary antibodies at room temperature for 1 h. Immunoreactive signals were visualized using enhanced chemiluminescence (ECL) and quantified with ImageJ analysis software.
Immunohistochemistry
Cellular localization of CAPN1 in the SDH was examined using immunofluorescence double-labeling, following a previously reported protocol. 19 Rats were anesthetized with 0.4% sodium pentobarbital (40 mg/kg, i.p.) and perfused transcardially with 4% paraformaldehyde through the ascending aorta. The lumbar 5 spinal segment was dissected, post-fixed in 4% paraformaldehyde at 4°C for 30 min, and subsequently cryoprotected in 30% sucrose at 4°C for 3 days. Frozen sections were generated with a LEICA CM1900 cryostat (Germany) after dehydration and maintained at 4°C until use. Immunohistochemical staining was conducted with primary antibodies against calpain1 (Rabbit, 1:200, Abcam, catalogue no.: ab39170), NeuN (Mouse, 1:200, CST, catalogue no.: 94403), GFAP (Mouse, 1:400, CST, catalogue no.: 3670S), or Iba1 (Goat, 1:500, Abcam, catalogue no.: ab5076). Following overnight incubation at 4°C, sections were subsequently treated with the appropriate secondary antibodies for 1 h at room temperature. Imaging was performed with a fluorescence microscope equipped with a CCD camera (Leica DFC350FX/DMIRB, Germany), and data analysis was carried out using Leica IM50 software (Germany). Colocalization of CAPN1 with individual markers was quantified, and intensity–distance plots were constructed to illustrate spatial intensity profiles.
Intraspinal drug microinjection
For intraspinal delivery of adeno-associated virus (AAV), vectors encoding CAPN1 OE (serotype 2/9, rAAV-CaMKIIa-Capn1-EGFP) or control (rAAV-CaMKIIa-EGFP) were obtained from BrainVTA (BrainVTA Co., Ltd., Wuhan, China) and administered 21 days prior to experimentation. The gene sequence employed for CAPN1 OE is as follows: rat Capn1, NM_019152.2, 2142 bp. The control sequence utilized was: CCTAAGGTTAAGTCGCCCTCG. Anesthesia was induced with 4% isoflurane in oxygen and maintained with 1.5%–2% isoflurane throughout the procedure. Following exposure of the L4–L6 vertebrae, the vertebral column was stabilized using a stereotaxic frame. A limited laminectomy was then carried out, followed by a careful incision of the arachnoid to visualize the spinal cord. Viral infusion was delivered through glass micropipettes (tip diameter 30–40 μm) attached to a 10 μL syringe at a constant rate of 30 nL/min. The viral suspension was administered using glass micropipettes with a tip diameter of 30–40 μm, which were connected to a 10-μL microsyringe. Infusion was carried out at a controlled rate of 30 nL per minute. Injection sites corresponding to laminae I–III of the spinal dorsal horn were determined using stereotaxic coordinates (ML, ±0.75 mm from the central vessel; DV, –0.25 mm). 33 Four injection sites were targeted, each receiving 150 nL of AAV, spaced 1 mm apart longitudinally, with the rostral and caudal extremes excluded. The micropipette remained in position for 10 min after each infusion before withdrawal, and the incision was subsequently closed by layered suturing.
Data analysis
Protein expression was quantified by measuring the gray intensity of Western blot bands with Tanon GIS software, followed by validation through image intensity analysis and colocalization assessment using ImageJ. Quantitative outcomes derived from Western blotting, immunostaining, and behavioral evaluations were processed with GraphPad Prism (version 9.5.1) and presented as mean ± standard error of the mean (SEM). Statistical evaluation of Western blotting and immunostaining results was assessed through one-way ANOVA with Dunnett’s post hoc test, or Student’s
Results
Decreased mechanical withdrawal thresholds but unaltered CAPN1/CAPN2 levels in DRGs of PINP rats
Pain-related behaviors and calpain protein expression in DRGs were subsequently examined in the PINP model. Rats in the paclitaxel group received intraperitoneal injections (i.p.) of 6 mg/kg every other day (days 1, 3, 5, 7), yielding a cumulative dose of 24 mg/kg, while the vehicle group was administered an equivalent volume of normal saline under the same protocol (Figure 1(a)). Body weight did not differ significantly between the paclitaxel and vehicle groups (Figure 1(b)). In contrast, mechanical withdrawal thresholds of bilateral hind paws declined as early as day 1 following paclitaxel administration (left,

Establishment of the PINP model and assessment of pain behaviors and calpain expression in DRGs. (a) Experimental design outlining paclitaxel administration and behavioral testing. (b) Changes in body weight between vehicle-treated and paclitaxel-treated rats. (c, d) Mechanical withdrawal thresholds of the left (c) and right (d) hind paws measured over time after paclitaxel administration. *
Increased calpain expression/activation & TRPV4/NCS-1 in PINP rat spinal cord
To determine whether spinal mechanisms contribute to PINP pathology, calpain expression and activity were further analyzed in spinal cord tissues. At 28 days after paclitaxel treatment, protein levels of both CAPN1 and CAPN2 were significantly increased (

Expression of calpain proteins, calpain enzymatic activity, TRPV4, and NCS-1 in the spinal cord of PINP rats. (a–c) Western blot images (a) with quantitative analysis (b, c) showing CAPN1 and CAPN2 protein levels in spinal cord tissues 28 days post-paclitaxel treatment. (d–f) Western blot images (d) and quantitative analyses (e, f) of spectrin degradation fragments (150 kDa and 136 kDa) used as indicators of calpain activity. (g, h) Quantitative evaluation of TRPV4 (g) and NCS-1 (h) protein expression in the spinal cord, corresponding to Western blot images in (d). Data presented as mean ± SEM. **
CAPN1 strong co-localization with neurons, weak with glia cells in spinal dorsal horn
Previous research indicates that TRPV4 activation elevates intracellular Ca2+ levels and triggers the release of inflammatory mediators, while NCS-1 contributes to neuropathic pain hypersensitivity by modulating Ca2+ signaling.9,34–36 The increased expression of TRPV4 and NCS-1 (Figure 3(d), (g), (h)) implies enhanced intracellular Ca2+ dynamics; however, whether this alteration represents a downstream consequence of calpain activation or serves as an upstream regulatory event remains unresolved. Unlike CAPN2, CAPN1 is activated by micromolar Ca2+ concentrations (1–10 μM). The extent to which CAPN1 activation influences CAPN2 expression and the possible interdependence between these two isoforms have not been addressed. In the present study, the functional significance of CAPN1 upregulation in the spinal cord of PINP rats (Figure 2) remains uncertain, leaving open the possibility of either anti-inflammatory or pro-inflammatory contributions.

Cellular localization of CAPN1 in the spinal cord detected by immunofluorescence double-labeling (a) Representative immunofluorescence images illustrating co-staining of CAPN1 (specific for calpain1), NeuN (neuronal nuclei marker), and DAPI (nuclear labeling). (b, c) Intensity–distance plots with corresponding colocalization analysis from (a), indicating NeuN fluorescence distribution along the distance axis, yielding a Pearson’s correlation coefficient (Pearson’s Rr) of 0.75 and an overlap coefficient (Overlap_R) of 0.84. (d) Representative images showing CAPN1 co-staining with GFAP (astrocyte marker) and DAPI. (e, f) Intensity–distance plots and colocalization analysis corresponding to (d), revealing GFAP fluorescence distribution along the distance axis with Pearson’s Rr of 0.06 and Overlap_R of 0.47. (g) Representative images showing CAPN1 co-staining with Iba1 (microglia marker) and DAPI. (h) Intensity–distance plot and colocalization analysis corresponding to (g), demonstrating Iba1 fluorescence distribution along the distance axis, with Pearson’s Rr of 0.05 and OverlapR of 0.26. Scale bars are provided in each image. (j–m) Representative immunofluorescence images, acquired under identical settings, showing marked microglial activation in the SDH of the PTX-induced pain model. Sholl analysis for the processes intersections (k) and max length (l), and quantification for the body area (m) of microlia are conducted. **
Future investigations will prioritize two directions: first, clarification of the relationship between elevated CAPN1 expression in the spinal cord of PINP rats and the manifestation of pain behaviors; second, evaluation of the potential interplay between CAPN1 and CAPN2, as well as their involvement in TRPV4/NCS-1–mediated signaling.
The cellular distribution of CAPN1 in the SDH was examined using double immunofluorescence staining. In neurons (Figure 3(a), CAPN1+NeuN+DAPI), CAPN1 exhibited extensive overlap with NeuN. 76.69% ± 1.39% of NeuN-positive neurons (Red) exhibit colocalization with CAPN1 immunofluorescent positive signals (Green). Quantitative assessment (Figure 3(c)) yielded a Pearson’s correlation coefficient of 0.75 and an overlap coefficient of 0.84, reflecting a strong spatial association and substantial signal coincidence between CAPN1 and NeuN. The NeuN intensity–distance profile (Figure 3(b)) further confirmed the consistent co-localization pattern of CAPN1 within neuronal populations.
By contrast, the CAPN1 positive signals rarely exhibit colocalization with the astrocyte marker GFAP and the microglia marker Iba1. In astrocytes (Figure 3(d), CAPN1+GFAP+DAPI) and microglia (Figure 3(g), CAPN1+Iba1+DAPI), colocalization analysis (Figure 3(f) and (i)) revealed very low Pearson’s correlation coefficients (0.06 and 0.05, respectively) and modest overlap coefficients (0.47 and 0.26, respectively), indicating limited correspondence between CAPN1 and GFAP or Iba1. The intensity–distance plots of GFAP and Iba1 (Figure 3(e) and (h)) displayed discrete distribution patterns, aligning with the minimal co-localization of CAPN1 in astrocytic and microglial compartments.
Although CAPN1 exhibited only minimal colocalization with GFAP and Iba1, we still observed pronounced microglial activation in the PTX-induced pain model. A Sholl analysis was conducted, the number of branching nodes in the processes of microglia and their processes max length were markedly decreased (Figure 3(j)–(l)), while the cell bodies of microglia marked by the Iba1 positive signal were significantly enlarged (Figure 3(m)), exhibiting an amoeboid activated state. This finding suggests that PINP may also be implicated in microglia-mediated neuroinflammatory responses.
Capn1 overexpression in CaMKII+ spinal neurons induces mechanical allodynia
CaMKII-positive neurons within the SDH have been identified as essential mediators of pain signaling and synaptic plasticity regulation.37–39 To examine the role of CAPN1, intraspinal microinjection (DV, −0.25 mm; ML, ±0.75 mm) was performed to induce targeted overexpression of Capn1 (Capn1 OE) or deliver the control virus (Ctrl V) in CaMKII-positive neurons. On day 21 following viral transfection, administration of PTX (6 mg/kg, i.p.) or saline (Vehicle, i.p.) was carried out on days 1, 3, 5, and 7, after which behavioral responses were assessed (Figure 4(i)). Compared with the Ctrl V + Vehicle group, the Ctrl V + PTX group exhibited a marked reduction in bilateral mechanical hindpaw withdrawal thresholds beginning on day 1, which persisted until day 28 (Figure 4(a) and (d)). Bilateral Capn1 OE with Vehicle (Capn1 OE + Vehicle group) similarly led to a significant threshold decrease relative to Ctrl V + Vehicle, with onset on day 3 post-injection (Figure 4(b) and (e)). Partial recovery occurred in the later phase, as the right paw threshold reduction remained significant only until day 11 (Figure 4(e)). In contrast, Capn1 OE combined with PTX (Capn1 OE + PTX group) produced a sustained bilateral threshold reduction from day 3 through the endpoint (Figure 4(c) and (f)). Although significant group differences in right paw thresholds were detected at isolated time points (days 1, 5, and 28; Figure 4(h)), no group differences were identified in left paw thresholds across the experimental timeline (Figure 4(g)).

Effects of Capn1 overexpression (OE) in spina CaMKII-positive neurons on mechanical hindpaw withdrawal thresholds in rats (a–c) Comparison of left hindpaw mechanical withdrawal thresholds between the control virus + paclitaxel (Ctrl V + PTX) group and the control virus + vehicle (Ctrl V + Vehicle) group (a), or the Capn1 OE + Vehicle group and the Ctrl V + Vehicle group (b), or the Capn1 OE + PTX group and the Ctrl V + Vehicle group (c). (d–f) Comparison of right hindpaw mechanical withdrawal thresholds between the Ctrl V + PTX group and the Ctrl V + Vehicle group (d), or the Capn1 OE + Vehicle group and the Ctrl V + Vehicle group (e), or the Capn1 OE + PTX group and the Ctrl V + Vehicle group (f). The “Ctrl V + vehicle” groups across the three panels in a–c and d–f respectively represent the same left control group and right control group. *
Spinal Capn1 OE increases the protein levels of CAPN1, CAPN2, SBDP, and NCS-1
Spinal viral transfection was carried out from Day −20 to Day 0 according to the protocol illustrated in Figure 5(i), followed by intraperitoneal injections of PTX (6 mg/kg) or saline on Days 1, 3, 5, and 7. Animals were sacrificed on Day 14, and spinal tissues were harvested for analysis (Figure 5(a)). Western blotting was performed to evaluate the expression of CAPN1, CAPN2, SBDP, TRPV4, and NCS-1, with β-actin as the loading control.

Effects of PTX exposure and Capn1 overexpression on spinal expression of CAPN1, CAPN2, SBDP, TRPV4, and NCS-1. (a) Schematic diagram illustrating the experimental protocol. Representative Western blot bands (b) and quantitative analysis of CAPN1 (c), CAPN2 (d), SBDP at 150 kDa (e) and 136 kDa (f), TRPV4 (g), and NCS-1 (h) protein expression in the spinal cord tissues of rats. Data are presented with statistical significance indicated as *
Relative to the Ctrl V + Vehicle group, both the Ctrl V + PTX and Capn1 OE + PTX groups displayed a significant elevation in CAPN1 (
Discussion
Analysis in this study revealed that CAPN1 and CAPN2 expression in DRGs of paclitaxel-treated rats did not change significantly, whereas both proteins exhibited marked upregulation in the SDH, accompanied by heightened calpain activity and increased expression of NCS-1 and TRPV4. CAPN1 localization was predominantly neuronal, and its overexpression in CaMKII-positive neurons of naïve rats directly induced mechanical hypersensitivity to a degree comparable with that observed in paclitaxel-treated animals. Furthermore, Capn1 OE elevated NCS-1 expression and initiated a cascade characterized by CAPN2 upregulation and sustained calpain hyperactivation.
Calpains are Ca2+-dependent cysteine proteases with dual roles in regulating neuroinflammation. 12 CAPN2 has been associated with pro-inflammatory signaling40,41 whereas CAPN1 has frequently been linked to neuroprotective mechanisms.41–44 Our previous work demonstrated that L5 ventral root transection (L5-VRT) causes downregulation of CAPN1 and upregulation of CAPN2 in DRG neurons, with CAPN2 driving increased calpain activity and IL-6 expression. The reduction of CAPN1 is hypothesized to diminish its protective role, thereby acting synergistically with CAPN2 upregulation to promote the onset of chronic pain. 19 Although CAPN1 may exert protective actions in certain contexts, excessive activation in conjunction with CAPN2 can drive inflammatory responses and neurodegeneration. 45 In this work, using a clinically relevant paclitaxel regimen (6 mg/kg i.p., administered every other day for four injections), bilateral mechanical hypersensitivity was observed in rats from day 1 and persisted throughout the 28-day study period (Figure 1(c) and (d)). In contrast to the L5-VRT model, paclitaxel-induced chronic pain was not accompanied by changes in CAPN1 or CAPN2 protein expression in bilateral DRGs (Figure 1(e)–(g)). Previous study has also reported that paclitaxel induced an elevation in both innate immune responses and adaptive immune responses in DRGs of mice. 46 Furthermore, CAPN1 is activated in the mice DRGs 6 h after a single high-dose paclitaxel injection (60 mg/kg, i.p.). 47 Differences in dose, regimen, evaluation time, and species likely account for the inconsistency with those results. Based on these observations, subsequent investigations were directed toward the spinal cord.
In PINP rats, CAPN1 and CAPN2 expression in the SDH was markedly elevated, accompanied by increased calpain protease activity as well as upregulation of NCS-1 and TRPV4 (Figure 2). Notably, CAPN1 displayed aberrant induction absent in the previous L5-VRT model, where dorsal horn inflammation was instead linked to CAPN2 elevation. 19 This distinction raised the question of CAPN1’s functional involvement in PINP. Immunofluorescence double labeling demonstrated predominant neuronal expression of CAPN1 (Figure 3(a)–(c)). Viral-mediated overexpression of CAPN1 in CaMKII-positive neurons of the dorsal horn in naïve rats produced a direct reduction in mechanical pain threshold (Figure 4), while simultaneously enhancing NCS-1 expression and further increasing CAPN2 expression and activity (Figure 5).
The evidence indicates a dual regulatory role of CAPN1 in neuroinflammatory processes. Beyond previously reported anti-inflammatory effects,48–50 CAPN1 may also potentiate pro-inflammatory signaling. Studies in cellular and murine models of neuroinflammation and Parkinson’s disease have shown that silencing CAPN1 or CAPN2 substantially decreases IL-6 and IL-1β expression. 45 Additional evidence from sickle cell disease models demonstrated that CAPN1 gene deletion mitigated chronic pain behaviors. 51 Mechanistically, CAPN1 appears to influence NF-κB signaling. While canonical NF-κB activation typically proceeds through IKK-dependent phosphorylation and subsequent proteasomal degradation of IκBα, 52 CAPN1 can degrade IκBα directly in a Ca2+-dependent fashion. 53 Furthermore, high glucose conditions were recently shown to promote HK2-mediated phosphorylation of IκBα, strengthening its interaction with CAPN1 and accelerating degradation, thereby intensifying NF-κB activation. 54
NCS-1 is a key Ca2+-binding protein. 55 Although CAPN1 overexpression did not directly alter TRPV4 protein levels (Figure 5), previous studies demonstrated that NCS-1 modulated TRPV4 channel activity; elevated NCS-1 enhanced TRPV4-dependent Ca2+ influx, intensifying intracellular Ca2+ signaling and contributing to cellular injury and inflammatory responses.9,36 The CAPN2 upregulation and calpain hyperactivation observed in this study may therefore occur as a downstream consequence of NCS-1 elevation (Figure 5). Synergistic interaction between NCS-1 and TRPV4 amplifies Ca2+ signaling, driving CAPN1-mediated degradation of IκBα, subsequent NF-κB activation, and transcription of pro-inflammatory genes. 52 Activated NF-κB additionally promotes NLRP3 inflammasome assembly, resulting in IL-1β maturation and pyroptotic cell death.56–58 It has been demonstrated that activation of the NLRP3 inflammasome depends on calpain activity, especially that of CAPN1.59,60 Recent studies have revealed that chemotherapeutic agents can induce the accumulation of double-stranded DNA (dsDNA) in the cytoplasm. These dsDNA molecules bind to SARM1, leading to the degradation of NAD+ and nerve cell death, ultimately triggering CIPN. 61 These suggest that chemotherapy-induced nerve cell death is a significant underlying cause of neuropathic pain.
Evidence from the present study indicates that paclitaxel appears to enhance NCS-1 expression in the SDH via its interaction with NCS-1, 9 thereby enhancing intracellular Ca2+ signaling and calpain activity. CAPN1, activated under micromolar Ca2+ conditions, contributes to neuroinflammatory injury by promoting NF-κB activation and NLRP3 inflammasome assembly, while concurrently reinforcing a positive feedback loop through further induction of NCS-1. This self-perpetuating cycle may drive PINP progression (Figure 6).

Interaction among NCS-1, TRPV4, and CAPN1 in paclitaxel-induced neuropathic pain (PINP). Paclitaxel enhances NCS-1 expression in spinal cord neurons, activating intracellular Ca2+ signaling and enhancing calpain protease activity. Activated CAPN1 further enhances NCS-1 expression, thereby intensifying TRPV4-dependent Ca2+ influx and CAPN2 activity. This self-reinforcing loop promotes neuroinflammatory responses and drives both initiation and progression of PINP.
Limitations
Although the data suggest that CAPN1 overexpression enhances calcium influx and subsequently activates CAPN2 to promote neuropathic pain, a systematic assessment of the temporal expression dynamics of CAPN1 and CAPN2 following paclitaxel administration was not performed. Future studies employing CAPN1 knockdown approaches will be necessary to clarify its specific contribution to the development of paclitaxel-induced peripheral neuropathy.
Conclusion
In summary, the present study identifies aberrant activation of CAPN1 in SDH neurons, rather than in DRG, as a principal driver of PINP. Through interaction with NCS-1 and TRPV4, CAPN1 may enhance Ca2+ signaling and activates pro-inflammatory pathways involving NF-κB and NLRP3 inflammasome assembly, thereby sustaining neuroinflammation and chronic pain. Collectively, the results provide new mechanistic insights into PINP and point to the spinal calpain-1/NCS-1/TRPV4 axis as a potential therapeutic target for prevention and intervention.
Footnotes
Abbreviations
AAV: adeno-associated virus; CAPN1: calpain-1; CAPN2: calpain-2; CaMKII: Calcium/Calmodulin-dependent protein kinase II; CIPN: chemotherapy-induced chronic pain; DEGs: differentially expressed genes; DRG: dorsal root ganglia; FDR: false discovery rate; GEO: Gene Expression Omnibus; IL-6: Interleukin-6; InsP3R: inositol-1,4,5-trisphosphate receptor; L5-VRT: L5 ventral root transection; NCS-1: neuronal calcium sensor-1; NLRP3: NOD-like receptor family pyrin domain containing 3; OE: Overexpression; PINP: paclitaxel-induced neuropathic pain; PWT: paw withdrawal threshold; SBDP: spectrin breakdown product; SD: Sprague–Dawley; SDH: spinal dorsal horn; SEM: standard error of the mean; TRPV4: Transient receptor potential V4.
Author contributions
Authors Y Z and S-X C designed the study and wrote the protocol. Authors Y-N Z, S-X C, Y-H Z, and J-Q Y performed the experiment and analyzed the data. Authors Y Z and S-X C managed the literature searches and Figures drawing. Authors Q-Y L and S-X C performed Sholl analysis and Figures drawing. Authors Y Z and S-X C wrote the manuscript. Authors Y-N Z and S-X C revised the manuscript. All authors read, contributed to and have approved the final manuscript.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China (Grant Nos. 82371236, 81870873, and 81801111), the Natural Science Foundation of Guangdong Province of China (Grant Nos. 2024A1515012479 and 2022A1515012198), and Shenzhen Science and Technology Program (Grant No. JCYJ20250604174812016).
