Danshenol B alleviates central post-stroke pain by regulating the PIK3CG/NLRP3 signaling pathway.

Background

CPSP is a type of neuropathic pain directly associated with stroke lesions that occurs in both ischemic and hemorrhagic stroke patients [ 1 ]. CPSP affects approximately 1–35% of stroke survivors, with markedly higher incidences reported when lesions involve critical regions such as the medulla or thalamus, where prevalence rates can exceed 24% and 50%, respectively [ 2 , 3 ]. CPSP symptoms may emerge immediately after stroke or develop progressively, typically between one and twelve months [ 4 ]. Furthermore, approximately 35–51% of patients experience evoked pain, often triggered by mechanical allodynia or cold hyperalgesia [ 5 ]. In basic research, CPSP models are most commonly established by injecting collagenase or autologous blood into the brain [ 6 , 7 ]. Studies have shown that the pathophysiology of CPSP is complex, involving inflammatory responses [ 2 , 8 ], microglial activation [ 6 ], apoptosis, neurovascular and blood-brain barrier disruptions [ 9 ], oxidative stress [ 10 ], and free radical production [ 11 ]. Current treatment options in Western medicine for CPSP include pharmacological agents, such as amitriptyline, lamotrigine, and gabapentin, as well as physical interventions including motor cortex stimulation, spinal cord stimulation, repetitive transcranial magnetic stimulation, and transcranial direct current stimulation [ 12 – 14 ]. Despite these interventions, many patients continue to experience suboptimal outcomes. Traditional Chinese medicine (TCM) has gained increasing attention due to its favorable safety profile and widespread acceptance in clinical practice.

As a TCM, Danshen exhibits significant anti-inflammatory, antioxidant, and antiapoptotic properties, which are directly relevant to the pathogenesis of CPSP [ 15 ]. Most research on Danshen in the context of stroke has focused on ischemic stroke, owing to its well-established efficacy in promoting blood circulation and removing stasis. In contrast, its application in hemorrhagic stroke remains underexplored, largely due to the prevailing concern that blood-activating and stasis-resolving herbs may exacerbate bleeding. However, a 2019 randomized trial involving 537 hemorrhagic stroke patients published in The Lancet reported that antithrombotic therapy did not increase the risk of recurrent bleeding [ 16 ], suggesting that such concerns may be overstated. Emerging evidence also indicates that compound Danshen preparations can alleviate pain through mechanisms involving signal transduction pathways, the endocrine system, and the nervous system [ 17 ]. Thus, exploring the role of individual Danshen constituents in the context of cerebral hemorrhage and pain may yield valuable insights. In 1997, Tezuka et al. isolated a bioactive compound, Danshenol B (C₂₂H₂₆O₄), from the methanol extract of Salvia miltiorrhiza Bunge, which exhibited strong aldose reductase (AR) inhibitory activity [ 18 ]. Despite this, research on Danshenol B remains limited, and its pharmacological effects are not well characterized. Although no studies to date have examined analgesic potential or mechanisms of Danshenol B, it is worth noting that AR plays a key role in the development of sensory abnormalities in diabetic neuropathy [ 19 ]. Additionally, data from the Traditional Chinese Medicine Systems Pharmacology (TCMSP) platform suggest that Danshenol B may target the Mu-type opioid receptor, highlighting its potential relevance for pain management. Collectively, these findings support the hypothesis that Danshenol B may possess therapeutic value in the treatment of CPSP.

Danshen targets numerous proteins, and the pathophysiology of stroke involves a wide range of molecular pathways. One such key target is PIK3CG (also known as PI3Kγ), a catalytic subunit of phosphoinositide 3-kinase that plays a pivotal role in regulating inflammation and oxidative stress. PIK3CG has been implicated in various disease states, including autoimmune disorders and multiple myeloma [ 20 – 22 ]. Recent studies have demonstrated that PIK3CG is significantly upregulated in ischemic stroke lesions, where it activates microglia and astrocytes, contributing to neuroinflammation and stroke progression [ 23 , 24 ]. Knockdown of PIK3CG has been shown to reduce blood–brain barrier permeability and decreases infarct volume [ 25 ]. Furthermore, PIK3CG-mediated activation of the Akt signaling pathway has been associated with bone cancer pain, suggesting that PIK3CG may play a role in both stroke and pain pathophysiology [ 26 ]. NLRP3, which is involved in inflammasomes and pyroptosis [ 27 , 28 ], is implicated in conditions such as depression [ 29 ], arthritis [ 30 ], and atherosclerosis [ 31 ]. Inhibition of the HIF-1α/NLRP3 signaling pathway has been shown to alleviate pain, anxiety, and depression in hemorrhagic stroke patients [ 6 ]. NLRP3 interacts with the Stimulator of Interferon Genes (STING) protein to promote microglial pyroptosis during ischemic stroke [ 32 ]. Additionally, certain TCMs provide neuroprotection by inhibiting NLRP3 during stroke [ 33 ]. Collectively, these findings highlight the critical roles of both PIK3CG and NLRP3 in the molecular mechanisms underlying stroke and pain, and support the rationale for investigating the PIK3CG/NLRP3 axis as a therapeutic target in CPSP.

In this study, we established a mouse model of CPSP and performed transcriptomic sequencing to investigate gene expression changes associated with the condition. Using network pharmacology analysis, we identified key active components of Danshen and their potential molecular targets relevant to CPSP. Among these, Danshenol B—a major bioactive compound—was selected for further evaluation. We assessed its therapeutic efficacy through oral administration in CPSP mice. Additionally, we explored the underlying molecular mechanisms by which Danshenol B exerts its analgesic effects, with a particular focus on the regulation of the PIK3CG/NLRP3 signaling pathway. Based on these investigations, we hypothesize that Danshenol B alleviates CPSP by inhibiting the PIK3CG/NLRP3 signaling pathway, offering a novel therapeutic strategy for post-stroke pain management.

Methods

Animal preparation

Male C57BL/6J mice, weighing 25–30 g each, were obtained from Beijing Vital River, China. The mice were housed under a 12-hour light–dark cycle at a temperature of 24 °C ± 1 °C and humidity levels of 40–60%, with ad libitum access to food and water. The mice were acclimated for a minimum of 7 days before the commencement of the experiments. All animal care procedures were approved by the Zhengzhou University Animal Care and Use Committee and complied with the International Association for the Study of Pain guidelines. Ethical approval was obtained from the Zhengzhou University Ethics Committee (Approval No. ZZUIRB2024-191).

Intracranial injection

After the mouse was anesthetized with 3% isoflurane (RWD Life Technology Co., Ltd., China), it was placed in a stereotaxic apparatus (Harvard Apparatus, Cambridge, MA), the head was secured with ear bars, and anesthesia was maintained with 1.5% isoflurane. The skin was disinfected, an incision was made, and hydrogen peroxide was used to remove the periosteum. To induce central post-stroke pain, as previously described [ 7 ], a glass electrode-tipped microliter syringe (7001KH, Hamilton, USA) was used to inject collagenase IV (Coll) (0.01 U/10 nL, Sigma‒Aldrich, St. Louis, MO, USA) into the right VPL/VPM at the following coordinates: lateral: 1.52 mm, bregma: -1.67 mm, depth: -3.6 mm. The control group was injected with an equal volume of saline. The injection was performed over 10 min at a constant rate. After injection, the syringe was retracted by 0.1 mm, held in place for an additional 10 min to prevent reflux, and then slowly withdrawn.

For mechanistic studies, lentiviral particles encoding either a PIK3CG overexpression construct (LV-pik3cg-OE, SC-424436-LAC, Santa Cruz Biotechnology, CA, USA), PIK3CG shRNA (sh-pik3cg, SC-39130-V, Santa Cruz Biotechnology), NLRP3 overexpression construct (LV-NLRP3-OE, SC-432122-LAC, Santa Cruz Biotechnology), Control Lentiviral Activation Particles (LV-NC, SC-437282, Santa Cruz Biotechnology) or Control CRISPR/Cas9 Plasmid (sh-NC, SC-418922, Santa Cruz Biotechnology) were injected into the same stereotaxic coordinates in separate groups of CPSP mice. These injections were administered after CPSP model establishment to modulate PIK3CG or NLRP3 expression in vivo and to investigate their roles in the analgesic effects of Danshenol B. Following each injection, the skull was moistened with saline, the incision was sutured, and the wound was disinfected with iodine tincture. The mice were then placed in a 36 °C recovery chamber for postoperative monitoring and care. The rationale for PIK3CG overexpression was to determine whether it is functionally required for the analgesic effect of Danshenol B.

Administration of Danshenol B

For drug administration, Danshenol B (purity of 99%, HY-122963, MCE, USA, CAS: 189308-09-6) was diluted in PBS. A 6-gauge gavage needle was used to aspirate the diluted Danshenol B (for the control group, the same volume of PBS was used). Gently grasp the mouse and fix its head with the mouse facing upward. The gavage needle was inserted from the upper jaw and gently pushed inward. Once the needle reaches halfway through the needle, it feels the absence of resistance; then, the drug is injected, and the gavage needle is withdrawn. The drug was administered once daily for 21 consecutive days.

Modified neurological severity score

The modified neurological severity score (mNSS) was used to assess neurological function [ 34 ]. The evaluation included measurements of motor skills, sensory perception, balance, and reflexes, with scores ranging from 0 to 18. Lower scores indicate better neurological status, whereas higher scores reflect more severe deficits. Detailed scoring criteria are provided in Supplementary Table S1 .

Behavioral tests

Von Frey testing

Mechanical sensitivity was assessed via von Frey filaments (Stoelting, Wood Dale, IL, USA) with forces of 0.07 g and 0.4 g [ 35 ]. Paw withdrawal frequency (PWF) was measured by applying von Frey filaments to the central region of the plantar surface of the hind paw. The number of paw withdrawals was recorded over 10 trials to quantify sensitivity to mechanical stimuli.

Thermal hyperalgesia test

Thermal hyperalgesia was measured by recording the paw withdrawal latency (PWL) in response to a heat source. The mice were placed on a raised glass platform, and a radiant heat source was directed onto the plantar surface of the hind paw through the glass plate. The withdrawal time of the hind paw following radiant heat application was recorded. Each hind paw was tested individually, with a 10-minute interval between tests.

Cold plate assay

The animals were placed on a plate precooled to 0 °C, and a timer was started immediately upon placement. The timer was stopped when the animals exhibited behaviors such as paw lifting, flinching, or guarding. The duration from initial placement to the observed flinch was recorded as the PWL in response to painful cold stimuli.

Nissl staining

Nissl staining was used to visualize Nissl bodies within the neuronal cytoplasm. Tissue sections were immersed in Nissl staining solution (Beyotime, Shanghai, China) for 10 min at room temperature. After staining, the sections were washed twice with distilled water and then rinsed with 70% ethanol to remove excess dye. Images of stained Nissl bodies were acquired using an Olympus DP80 microscope (Olympus, Japan). Image analysis and processing were conducted using ImageJ software (version 2.9.0; National Institutes of Health, Bethesda, MD, USA).

Tissue collection and RNA extraction

The study used two groups of mice, each comprising three biological replicates. Right-sided VPL/VPM tissues were separately extracted from each mouse. Each sample consisted of pooled tissues from three mice, resulting in nine mice per treatment group. Following the manufacturer’s guidelines, total RNA was isolated via a miRNeasy kit, which includes on-column genomic DNA digestion (Qiagen, Valencia, CA, USA). The RNeasy Micro Kit (Qiagen) was used for RNA purification, and concentrations were measured with a NanoDrop 2000 Spectrophotometer (Thermo Scientific, Wilmington, DE, USA). Sample quality was assessed via A260/280 ratios (1.97–2.08) with an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA).

RNA sequencing

Total RNA (1.0 µg per sample) was subjected to rRNA depletion via the Ribo-Zero rRNA Removal Kit (Human/Mouse/Rat; Illumina, San Diego, CA, USA). Eukaryotic mRNA was enriched via magnetic beads conjugated with oligo (dT) and fragmented randomly with fragmentation buffer. First-strand cDNA was synthesized from mRNA via random hexamer primers (Thermo Fisher Scientific, USA). A mixture of buffer, dNTPs, RNase H, and DNA polymerase I was subsequently added to synthesize second-strand cDNA, which was then purified via AMPure XP beads (Beckman Coulter, Brea, CA). The purified double-stranded cDNA was processed for end repair, A-tailing, and adapter ligation. This was followed by size selection via AMPure XP beads and enrichment of the cDNA library via PCR. RNA sequencing was performed on the Illumina Nova6000 platform (Illumina, San Diego, USA) using 2 × 150 bp paired-end reads, resulting in a minimum of 20 million reads per sample. Over 21.08 million reads were obtained in each group, ranging from 21.08 to 27.47 million in the Saline group and 22.55 to 27.03 million in the Coll group.

After trimming, the reads were aligned to the ENSEMBL reference mouse genome (GRCm38.95). Mapped reads distribution across genomic regions was visualized using HISAT2 (version 2.2.1) with the parameters --dta -p 6 --max [ 36 ]. The aligned reads were classified into exonic, intronic, and intergenic regions. Supplementary Fig. 1A shows the distribution of these categories in the Saline and Coll groups from the VPL/VPM. The overall dispersion of the samples in each group was similar (Supplementary Fig. 1B ). The Pearson correlation coefficient between samples within each group was approximately 1 (Supplementary Fig. 1C ). Principal component analysis revealed a clear separation between the Saline and Coll groups along PC1, which explained 96.4% of the total variance, indicating robust transcriptional reprogramming in the ipsilateral VPL/VPM 14 days after collagenase injection (Supplementary Fig. 1D ).

Gene expression levels were quantified using featureCounts and analyzed with DESeq2 (v1.30.1) for differential expression between the Saline and Collagenase groups [ 37 , 38 ]. Genes with a fold change ≥ 2 and an adjusted P -value (FDR) < 0.05 were considered significantly differentially expressed. Identified DEGs were classified as upregulated or downregulated based on their expression trends.

The RNA-seq data were uploaded with the Gene Expression Omnibus accession number GSE281041 .

Functional enrichment analysis of differentially expressed genes

For functional analysis, approximately 409 DEGs ( P < 0.05, fold change ≥ 2) from the VPL/VPM were categorized using Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway and Gene Ontology (GO) analyses via clusterProfiler (version 3.10.1, minGSSize = 1, maxGSSize = 10000, pAdjustMethod=“fdr”). Genetic regulatory networks were elucidated by organizing hierarchical categories into biological processes (BP), molecular functions (MF), and cellular components (CC). Gene set enrichment analysis (GSEA) was conducted via clusterProfiler (v3.10.1, nPerm = 1000). The GSEA utilized KEGG pathways and GO terms from the BP, CC, and MF categories as gene sets of interest, with the log2FC of each differential group serving as the ranked background gene set.

Protein‒protein interaction (PPI) network construction

To better understand the functional relationships among DEGs, interactions among significant DEGs in the VPL/VPM were predicted using the STRING database (version 12.0). Using Cytoscape (version 3.10.1), a network was constructed from the top 50 DEGs exhibiting the strongest correlations. The Centiscape plugin was used to determine the degree of connection for each node, which was represented by the node size. Nodes are color-coded red for upregulated genes and blue for downregulated genes.

Collection of danshen targets

We used the TCMSP (version 2.3) database to identify Danshen’s active constituents, selecting those with oral bioavailability ≥ 30% and drug likeness ≥ 0.183. We then retrieved the SMILES codes for 42 Danshen components from the PubChem database ( https://pubchem.ncbi.nlm.nih.gov/ , accessed on 14 July 2024). We uploaded these SMILES codes to the SwissTargetPrediction database ( http://www.swisstargetprediction.ch/ , accessed on 14 July 2024) to identify molecular targets. We cross-referenced these targets with DEGs to pinpoint Danshen’s therapeutic targets for CPSP treatment and visualized the results with a Venn diagram ( https://www.bioinformatics.com.cn/ ). The 11 identified therapeutic targets were uploaded to STRING for protein–protein interaction network construction and then imported into Cytoscape for topological analysis using the Centiscape and Network Analyzer plugins.

Molecular docking verification

SDF format ligand files for the four Danshen components were sourced from the PubChem database ( https://pubchem.ncbi.nlm.nih.gov/ , accessed on 23 July 2024). Four-dimensional structural models for the mouse proteins PIK3CG (A0A1W2P8F6), caspase-1 (CASP1, P29452 ), Bruton’s tyrosine kinase (BTK, P35991 ), and Matrix metalloproteinase-9 (MMP9, P41245 ) were sourced from the UniProt database ( https://www.uniprot.org/ , accessed on 23 July 2024). Molecular docking analysis using ZDOCK was performed to assess the interactions between Danshen components and their target proteins. The ZDOCK module results identified docking sites and computed ZDOCK scores.

Quantitative real-time RT-PCR

Total RNA isolated from the VPL/VPM was treated with DNase I (New England Biolabs, Ipswich, MA, USA) and reverse-transcribed using the RevertAid First Strand cDNA Synthesis Kit (Thermo), utilizing oligo(dT) or specific RT primers. A 2 µL cDNA template was amplified by real-time PCR using primers (Sangon Biotech, Shanghai, China) listed in Supplementary Table S2 . Each sample was analyzed in triplicate using a 20 µL reaction mixture, which included 250 nM of both forward and reverse primers, 10 µL of Thermo Scientific Maxima SYBR Green qPCR Master Mix (2×, Rox solution), and 20 ng of total cDNA. The reactions were performed using a 7500 Fast Real-Time PCR Detection System (Applied Biosystems, USA). The thermal cycling protocol consisted of an initial step at 95 °C for 3 min, followed by 40 cycles of 95 °C for 10 s, 60 °C for 30 s, and 72 °C for 30 s. The mRNA ratios were calculated using the ΔΔCt method (2 −ΔΔCt ). All the data were normalized to Gapdh , which is stable in mice after CPSP.

Western blot

Bilateral VPL/VPM tissues were collected from the mice and homogenized in ice-cold lysis buffer. After electrophoresis, the proteins were transferred onto PVDF membranes (Millipore, Darmstadt, Germany) at 250 mA. The membranes were blocked for 2 h at room temperature and then incubated overnight at 4 ℃ with primary antibodies against rabbit PIK3CG (1:1000, ab302958, Abcam), rabbit NLRP3 (1:1000, ab263899, Abcam), and rabbit GAPDH (1:1000, ab9485, Abcam). The membranes were then incubated with peroxidase-conjugated secondary antibodies for 2 h at room temperature. The immune complexes were detected using an enhanced chemiluminescence kit (Beyotime Biotechnology, Shanghai, China). Images were captured using a Bio-Rad ChemiDoc MP imaging system (Bio-Rad, USA). ImageJ software (version 2.9.0, National Institutes of Health, Bethesda, MD) was used to convert the images to 8-bit grayscale for further processing. The gel was calibrated using the “uncalibrated OD” function, after which the gel image was inverted. The target band was then selected, and its area was measured to obtain the area value.

Drug affinity responsive target stability (DARTS)

VPL/VPM tissues were collected from mice, and protein lysates were prepared as previously described. Danshenol B (purity 99%, HY-122963, MCE; CAS: 189308-09-6) was dissolved in 0.1% DMSO to prepare a 10 mM stock solution. For treatment, Danshenol B was added to 30 µg of total protein lysate to a final concentration of 10 µM, while control samples received an equal volume of 0.1% DMSO. The reaction volume was adjusted to 100 µL and incubated at 4 °C for 30 min to allow sufficient interaction between the small molecule and target proteins. Trypsin (2.5%, Thermo Scientific, USA; CAS: 15090046) was then added at a 1:200 (enzyme: protein, w/w) ratio, and the mixture was incubated on ice for 20 min. The reaction was terminated by adding 25 µL of 5× SDS-PAGE loading buffer (Beyotime, P0015L), followed by heating at 95 °C for 5 min to denature the proteins and halt proteolysis. The denatured samples were then subjected to Western blot analysis as previously described. Protein bands were visualized and quantified by densitometric analysis using ImageJ software.

Statistical analysis

To minimize bias, all behavioral assessments (including the von Frey test, thermal hyperalgesia test, cold plate assay, and mNSS scoring) were performed by investigators blinded to the treatment groups. For molecular experiments, including RNA extraction, RT-qPCR, and Western blot analysis, tissue samples were assigned coded identifiers, and data acquisition and analysis were conducted in a blinded manner. Mice that died during or after the collagenase IV injection procedure were excluded from the study. In addition, for behavioral and molecular analyses, data points deviating more than two standard deviations from the group mean were considered outliers and excluded from the final analysis.

Data were randomly collected and presented as the mean ± SD using GraphPad Prism 9. The normality of distribution for each dataset was assessed before analysis using the same software. Normality was tested using the D’Agostino–Pearson omnibus, Anderson–Darling, Shapiro–Wilk, and Kolmogorov–Smirnov tests with the Dallal–Wilkinson–Lillie for P value. Statistical analyses included two-tailed, unpaired Student’s t -tests and one-way and two-way ANOVAs with repeated measures. Significant results from ANOVAs were further analyzed using post hoc Tukey tests. P -values less than 0.05 were considered significant.

Results

Injection of collagenase IV into the VPL/VPM led to nociceptive hypersensitivities

A central post-stroke pain model was created by injecting collagenase IV (0.01 U/10 nL) into the right VPL/VPM junction in mice (Coll group). The control group received saline (10nL) injections (Saline group). The mNSS was evaluated in both groups 1 day before injection and at 1, 3, and 14 days post-injection. No significant differences were observed between the two groups, indicating that normal motor function, balance, and reflexes remained intact (Supplementary Fig. 2 , Table 1 ). Nissl bodies in the VPL/VPM region were examined 3 (Coll 3d group) and 14 (Coll 14d group) days postinjection. Compared with the Naive group, the Coll 3d and Coll 14d groups exhibited significant loss of Nissl bodies in the right VPL/VPM ( P = 0.03, P < 0.001; Figs. 1 A–E). We repeated the Saline and collagenase IV injections and evaluated pain behavior in the mice. In the Coll group, mechanical allodynia in the contralateral hind paw emerged on day 7 postinjection ( * P < 0.05, ** P < 0.01; Figs. 1 F, H), while no abnormalities were observed in the ipsilateral hind paw compared to the Naive group (Figs. 1 G, I). Cold hyperalgesia was similarly observed in the Coll group ( ** P < 0.01, *** P < 0.001; Fig. 1 J). However, neither the Saline nor the Coll group exhibited heat hyperalgesia compared to the Naive group (Figs. 1 K, L). These results confirm the successful establishment of a CPSP model in mice.

Table 1 The mNSS of mice in the saline and coll groups on the first day post-injection Mean S.D. Min. Max. P Motor tests Saline 0.1765 0.3930 0 1 0.6406 Coll 0.1176 0.3321 0 1 Sensory tests Saline 0 0 0 0 Coll 0 0 0 0 Beam balance tests Saline 0.4118 0.6183 0 2 0.6537 Coll 0.5294 0.8745 0 3 Reflex absence and abnormal movements Saline 0.1176 0.3321 0 1 0.6406 Coll 0.1765 0.3930 0 1 Note: Motor tests, Unpaired t -test, t = 0.4714, df = 32; Sensory tests, Unpaired t -test; Beam balance tests, Unpaired t -test, t = 0.4529, df = 32; Reflex absence and abnormal movements, Unpaired t -test, t = 0.4714, df = 32. Each group consisted of 17 mice

Fig. 1 Successfully established a CPSP mouse model. ( A ): Timeline of drug administration, behavioral assessments, and Nissl staining. ( B – D ): Nissl staining was employed to examine Nissl bodies in the VPL/VPM of mice from the Naive, Coll 3d, and Coll 14d groups. The left image has a scale bar of 1000 μm, and the right image has a scale bar of 200 μm. Solid lines in the images mark the boundaries of different nuclei. n = 6. ( E ): The number of pixels within the Nissl body–depleted region was quantified for each group (one-way ANOVA followed by Tukey’s test. F (2, 15) = 48.71, P < 0.001, n = 6). ( F – I ): PWF in the bilateral hind paws of mice was assessed using 0.07 g ( F – G ) and 0.4 g ( H – I ) von Frey filaments (two-way ANOVA followed by Tukey’s test. * P < 0.05, ** P < 0.01, *** P < 0.001 vs. Naive group. n = 12). ( J ): PWL in mice was evaluated using a cold plate (two-way ANOVA followed by Tukey’s test. ** P < 0.01, *** P < 0.001 vs. Naive group. n = 12). ( K – L ): PWL in the bilateral hind paws of mice was assessed using a thermal radiometer (two-way ANOVA followed by Tukey’s test. n = 12)

Altered expression profiles of mRNAs in the VPL/VPM after CPSP

We repeated the saline and collagenase IV injections and collected VPL/VPM tissue from the Saline and Coll group mice for transcriptomic sequencing on the 14th day post-injection (Fig. 2 A). Significant changes in mRNA expression within the VPL/VPM were observed after CPSP. Specifically, about 409 DEGs (347 upregulated, 62 downregulated) presented substantial changes in expression levels (Fig. 2 B). Clustered heatmaps of DEGs revealed distinct gene expression patterns in the VPL/VPM after CPSP (Fig. 2 C). G protein-coupled receptors (GPCRs) and ion channels are crucial for transmitting and modulating nociceptive information. Among the DEGs, 133 were identified as GPCR mRNAs, and 226 were identified as ion channel mRNAs in the VPL/VPM. The heatmaps display the top 15 upregulated and downregulated DEGs for both GPCRs (Fig. 2 D) and ion channels (Fig. 2 E) in the VPL/VPM.

Fig. 2 Statistical results of DEGs in the VPL/VPM of CPSP model mice after transcriptomic sequencing. ( A ): Saline or collagenase IV was injected into the VPL/VPM of two groups of mice, and RNA sequencing was performed on the ipsilateral VPL/VPM tissues 14 days post-injection. ( B ): The mRNA sequencing data between the Saline and Coll groups were plotted as a Volcano plot. n = 3 (each sample from 3 mice). ( C ): Heatmaps of significantly DEGs in the VPL/VPM from mice on day 14 post-CPSP were generated. n = 3 (each sample from 3 mice). ( D – E ): The top 15 upregulated and downregulated genes of GPCRs ( D ) and ion channels ( E ) in the VPL/VPM on day 14 after CPSP are shown. n = 3 (each sample from 3 mice). The up-and down-regulated genes are colored in yellow and blue, respectively

Functional enrichment analysis of the DEGs after CPSP

To analyze the functional enrichment of these DEGs, we used clusterProfiler for GO, KEGG and euKaryotic Ortholog Groups (KOG) pathway analyses to categorize the upregulated and downregulated DEGs by distinct processes. We screened the transcriptomic data for CPSP-related pathways and found that the upregulated DEGs were predominantly enriched in inflammation-associated pathways according to GO analysis (Fig. 3 A). In contrast, the downregulated DEGs were mainly associated with oxidative stress-related pathways (Fig. 3 B). KEGG analysis further revealed that the upregulated DEGs were primarily enriched in pathways related to inflammation and apoptosis (Fig. 3 C). Notably, no CPSP-relevant pathways were enriched among the downregulated DEGs in the KEGG analysis.

Fig. 3 Analysis results of CPSP-related pathways in GO and KEGG analyses. ( A ): Perform GO analysis on the upregulated DEGs after CPSP modeling, and screen for CPSP-related pathways, listing the top 10. ( B ): Perform GO analysis on the downregulated DEGs after CPSP modeling, and screen for CPSP-related pathways, listing the top 10. ( C ): Perform KEGG analysis on the upregulated DEGs after CPSP modeling, and screen for CPSP-related pathways

Enrichment analyses of KOG, GO, and KEGG pathways revealed distinct functional distributions of the DEGs. Specifically, KOG enrichment analysis showed that 48 upregulated DEGs were associated with signal transduction pathways, while 26 upregulated DEGs were related to cytoskeletal organization and extracellular structure (Supplementary Fig. 3 A, Supplementary Table S3 ). In contrast, the 62 downregulated DEGs were dispersed across multiple functional categories without significant enrichment in any single category (Supplementary Fig. 3 B, Supplementary Table S4 ). GO enrichment analysis revealed that the upregulated DEGs were enriched in immune system processes, biological adhesion, and cell killing within the biological process category (Supplementary Fig. 3 C, Supplementary Table S5 ). For molecular function, they were enriched in electron carrier activity and chemoattractant activity (Supplementary Fig. 3 C, Supplementary Table S5 ). Downregulated DEGs were mainly linked to biological adhesion, rhythmic processes, and synapse part functions (Supplementary Fig. 3D, Supplementary Table S6 ). In the network diagram of DEGs and GO pathways, the top five pathways with the smallest P -values were chosen for analysis. The analysis revealed upregulated DEGs such as Cst7 , newGene_6227, Cd5l , Clec7a , Mmp12 , and Saa3 (Supplementary Figs. 4 4 A, C, E), and downregulated DEGs including Gbp3 , Ifi203 , Irgm2 , Oasl2 , Gbp9 , Cplx3 , Kcnq5 , and Trh (Supplementary Figs. 4 A, B, D, F), all of which showed significant changes. The pathways associated with each gene are listed in Supplementary Table S7 . To systematically identify the lineage programs altered during CPSP and avoid missing genes that may have biological significance despite not being significantly differentially expressed, we performed GSEA. GO pathways were used as the gene sets of interest, and the top five gene sets with the lowest P -values were identified. Functions such as signal transduction, response to stimuli (Supplementary Figs. 5 A, C), the extracellular region (Supplementary Fig. 5 B), and metalloendopeptidase activity (Supplementary Fig. 5 C) were found to play important roles in CPSP. The results of the GO enrichment analysis were visualized using directed acyclic graphs, with branches illustrating inclusion relationships from broader to narrower categories. The top 10 GO enrichments functioned as master nodes. These were presented alongside related GO terms and systematically organized categories. Directed acyclic graphs were created for biological processes, molecular functions, and cellular components within the VPL/VPM (Supplementary Fig. 6 ).

KEGG pathway analysis showed that the upregulated DEGs were enriched mainly in the phagosome and tuberculosis pathways (Supplementary Fig. 3 E, Supplementary Table S8 ), while the downregulated DEGs were enriched in hepatitis C and the NOD-like receptor signaling pathway (Supplementary Fig. 3 F, Supplementary Table S9 ). The top five pathways with the lowest P -values were chosen for analysis in the network diagram of DEGs and KEGG pathways. The study revealed that upregulated DEGs such as Atp6v0d2 , Clec7a , and Itgax (Supplementary Fig. 4 G) and downregulated DEGs such as Irgm2 , Gbp3 , and Oasl2 (Supplementary Fig. 4 H) exhibited significant changes. The pathways related to each gene are listed in Supplementary Table S10 . Similarly, GSEA was performed using KEGG pathways as the gene sets of interest, and the top five gene sets with the lowest P values were identified. Pathways related to cancer, viral infection, and the PI3K-Akt signaling pathway were found to exhibit significant changes in CPSP (Supplementary Fig. 7 ).

To understand the functional interactions among DEGs and their potential involvement in neuropathic pain, a PPI network was constructed using the STRING database ( https://string-db.org/ ). The network was constructed using the top 50 protein-coding DEGs with the strongest correlations in the VPL/VPM. As illustrated in Supplementary Fig. 8, the upregulated DEGs, including Ptprc , Itgam , Fcgr3 , Cd44 , and Cd86 are crucial molecules among the hub genes in the VPL/VPM network (Supplementary Table S11 ).

Screening potential therapeutic targets for CPSP using components of Danshen

Among the 42 identified components of Danshen (Supplementary Table S12 ), 235 targets were predicted using the Swiss Target Prediction database (Supplementary Table S13 ). A network in which Danshen components were mapped to their targets was constructed with Cytoscape 3.9.1 (Fig. 4 A). Notably, tanshinaldehyde and (6 S)-6-(hydroxymethyl)-1,6-dimethyl-8,9-dihydro-7 H-naphtho[8,7-g]benzofuran-10,11-dione (Mol name: MOL007155) exhibited the highest number of associated potential targets at 100 each, followed by Danshenol B (94), Danshenol A (73), luteolin (58), and sclareol (48). The study identified 409 DEGs in the ipsilateral VPL/VPM of CPSP model mice. After excluding 11 novel genes, 398 DEGs remained. A Venn diagram was created to show overlaps between the predicted targets of Danshen and the DEGs in CPSP. The analysis revealed 11 potential candidate targets for treating CPSP with Danshen (Fig. 4 B).

Fig. 4 Transcriptomic intervention analysis of Danshen components in a mouse model of CPSP. ( A ): Construction of the Danshen–targets network. ( B ): Venn diagram of predicted Danshen targets and DEGs. ( C ): PPI network of interesting 11 targets. ( D ): KEGG pathway analysis of 11 key targets. ( E ): GO enrichment analysis of 11 key targets. ( F ): Danshen-targets-CPSP network diagram

The PPI network for the 11 intersecting targets was created using the STRING database and visualized using Cytoscape 3.9.1 (Fig. 4 C). The network consisted of 11 nodes and 76 edges, with node size and color representing their degree values. Btk , Mmp9 , Casp1 , Ctss , Ccr5 , and Pik3cg presented the highest degree values among the targets, at 10, 9, 9, 8, 8, and 8, respectively (Supplementary Table S14 ). Topological analysis revealed that Btk had the highest betweenness centrality score of 0.1589, followed by Pik3cg at 0.070 and Mmp9 and Casp1 at 0.048, suggesting that Btk is the most crucial component within this network (Supplementary Table S14 ).

To comprehensively understand how Danshen components affect CPSP at the system level, KEGG and GO enrichment analyses were performed on the 11 intersecting targets. In the KEGG enrichment analysis, the top 15 pathways with the lowest P -values ( P < 0.05) were selected, including primary immunodeficiency ( P < 0.001), the chemokine signaling pathway ( P < 0.01), NF-κB signaling pathway ( P < 0.01), and toxoplasmosis ( P < 0.01) (Fig. 4 D, Supplementary Table S15 ). Additionally, GO enrichment analysis revealed that the targets were involved in 742 biological processes, 45 cellular components, and 80 molecular functions (Supplementary Table S16 ). As shown in Fig. 4 E, the main enriched terms in the BP category included regulation of the inflammatory response ( P < 0.001), positive regulation of the inflammatory response ( P < 0.001), and myeloid leukocyte activation ( P < 0.001). The CC category was significantly enriched in the extrinsic component of the membrane and the cytoplasmic side of the plasma membrane, both with notable results ( P < 0.001). In the MF category, significant enrichment was observed for nonmembrane-spanning protein tyrosine kinase activity ( P < 0.001).

Molecular docking between danshen components and therapeutic targets

To identify the key Danshen components for CPSP, an interaction network involving 21 Danshen components and 11 intersecting targets was constructed (Fig. 4 F). The remaining 21 Danshen components do not target these 11 genes. Among the 21 Danshen components, MOL007155 exhibited the highest degree value at 8, followed by tanshinaldehyde and Danshenol B at 6, and sclareol at 4 (Supplementary Table S17 ). Topological analysis revealed that MOL007155 had the highest betweenness centrality score of 494.43, followed by tanshinaldehyde at 303.51, Danshenol B at 286.89, and sclareol at 100.09 (Supplementary Table S17 ). In conclusion, MOL007155, tanshinaldehyde, Danshenol B, and sclareol are likely the crucial Danshen components for combating CPSP.

Molecular docking was used to assess the binding affinities between four Danshen components—MOL007155, tanshinaldehyde, Danshenol B, and sclareol—and the key target proteins BTK, CASP1, MMP9, and PIK3CG (Fig. 5 ). The docking scores ranged from − 7.111 to -9.619 kcal/mol, all below − 7.000 kcal/mol, indicating strong binding affinities [ 39 ]. The binding pockets of BTK, CASP1, MMP9, and PIK3CG are tightly occupied by these components and are stabilized through hydrogen bonds and hydrophobic interactions (Supplementary Table S18 ). This finding indicates the effective binding of Danshen components to core targets via hydrogen bonds and hydrophobic interactions. Notably, the strongest binding affinity was between MOL007155 and BTK (-9.619 kcal/mol), followed by Danshenol B and PIK3CG (-9.127 kcal/mol) (Fig. 5 ).

Fig. 5 Molecular docking of main Danshen components with BTK, CASP1, MMP9, PIK3CG. ( A ): Interaction diagram between BTK and Tanshinaldehyde. ( B ): Interaction diagram between CASP1 and Tanshinaldehyde. ( C ): Interaction diagram between MMP9 and Tanshinaldehyde. ( D ): Interaction diagram between PIK3CG and Tanshinaldehyde. ( E ): Interaction diagram between BTK and Sclareol. ( F ): Interaction diagram between CASP1 and Sclareol. ( G ): Interaction diagram between MMP9 and Sclareol. ( H ): Interaction diagram between PIK3CG and Sclareol. ( I ): Interaction diagram between BTK and MOL007155. ( J ): Interaction diagram between CASP1 and MOL007155. ( K ): Interaction diagram between MMP9 and MOL007155. ( L ): Interaction diagram between PIK3CG and MOL007155. ( M ): Interaction diagram between BTK and Danshenol B. ( N ): Interaction diagram between CASP1 and Danshenol B. ( O ): Interaction diagram between MMP9 and Danshenol B. ( P ): Interaction diagram between PIK3CG and Danshenol B

Administration of danshenol B effectively alleviates CPSP in mice

To confirm the consistency of transcriptomic sequencing results, we performed RT-qPCR to quantify the mRNA expression levels of Btk , Casp1 , Mmp9 , and Pik3cg in the ipsilateral VPL/VPM of Saline and Coll groups on days 3 and 14 after modeling. On day 3, only Mmp9 mRNA ( P = 0.007) expression was significantly elevated in the Coll 3d group. By day 14, the mRNA levels of Btk ( P = 0.0154), Casp1 ( P = 0.0140), Mmp9 ( P = 0.0190), and Pik3cg ( P = 0.0174) were significantly increased in the Coll group compared to the Saline group (Figs. 6 A–E). Given the challenges in acquiring MOL007155 and the high expression of PI3K pathway genes in the CPSP model, as shown in Supplementary Fig. 7 A, we focused on the interaction between PIK3CG and Danshenol B to investigate its role in treating CPSP. A DARTS assay using VPL/VPM tissue from naïve mice demonstrated that Danshenol B binds to PIK3CG and enhances its resistance to proteolytic degradation, further confirming the direct interaction between Danshenol B and the PIK3CG protein ( P = 0.0171, Supplementary Fig. 9 ). By day 14 post-surgery, PIK3CG protein was significantly higher in the ipsilateral VPL/VPM of the Coll 14d group than in controls ( P = 0.0092, Figs. 6 F–G). To investigate the regulatory effects of Danshenol B on PIK3CG and CPSP, we orally administered varying concentrations (5/10/50 mg/kg) of Danshenol B to mice (Coll + Danshenol B group) once daily for 21 consecutive days, while the control group received PBS (Coll + PBS group). We assessed PIK3CG protein expression in the ipsilateral VPL/VPM and evaluated pain-related behaviors in mice (Fig. 6 H). Continuous administion of Danshenol B (50 mg/kg) for 14 days alleviated Coll-induced mechanical allodynia ( ## P < 0.01, ### P < 0.001, Figs. 6 I, K) and cold hyperalgesia ( ### P < 0.001, Fig. 6 M) in the contralateral hind paw, without altering the ipsilateral hind paw’s pain threshold (Fig. 6 J, L). Lower doses of 5 mg/kg and 10 mg/kg did not alleviate pain (Figs. 6 I–M). After completing the behavioral test (at 21 days post-injection), ipsilateral VPL/VPM tissues were collected from mice in the Coll + PBS group and the Coll + Danshenol B (50 mg/kg) group for Western blot analysis and RT-qPCR. The PIK3CG protein level in the ipsilateral VPL/VPM was significantly lower in the Coll + Danshenol B group compared to the Coll + PBS group ( P = 0.0109, Figs. 6 N–O). However, no significant changes were observed in the mRNA levels of Btk , Casp1 , Mmp9 , and Pik3cg (Supplementary Fig. 10 ). These results suggest that continuous administration of Danshenol B (50 mg/kg) reduces PIK3CG and alleviates mechanical allodynia and cold hyperalgesia.

Fig. 6 Danshenol B can alleviate mechanical allodynia and cold hyperalgesia in CPSP. ( A ): Saline or collagenase IV was injected into the VPL/VPM of mice, and the mRNA and protein expression in the ipsilateral VPL/VPM were evaluated 14 days post-injection. ( B – E ):The mRNA levels of Btk ( B ), Casp1 ( C ), Mmp9 ( D ), and Pik3cg ( E ) (unpaired t -test, t = 2.054, df = 10. P = 0.07, n = 6 in B left; unpaired t -test, t = 2.915, df = 10. P = 0.0154, n = 6 in B right; unpaired t -test, t = 0.6782, df = 10. P = 0.51, n = 6 in C left; unpaired t -test, t = 2.973, df = 10. P = 0.0140, n = 6 in C right; unpaired t -test, t = 3.4, df = 10. P = 0.007, n = 6 in D left; unpaired t -test, t = 2.794, df = 10. P = 0.0190, n = 6 in D right; unpaired t -test, t = 1.198, df = 10. P = 0.26, n = 6 in E left; unpaired t -test, t = 2.846, df = 10. P = 0.0174, n = 6 in E right). ( F – G ): Fourteen days post-injection, PIK3CG protein in the ipsilateral VPL/VPM of the Coll group was higher compared to the Saline group (unpaired t -test, t = 3.152, df = 11; P = 0.0092, n = 6 in Saline group, n = 7 in Coll group). (H): Timeline of drug administrations and pain behavior assessment time points. ( I – J ): PWF for contralateral ( I ) and ipsilateral ( J ) hind paws in each group, tested using 0.07 g von Frey filaments ( I : two-way ANOVA followed by Tukey’s test, F (16, 220) = 3.587, P < 0.001. J: two-way ANOVA followed by Tukey’s test, F (16, 220) = 0.6591, P = 0.83; * P < 0.05, ** P < 0.01, *** P < 0.001 vs. Saline group; ## P < 0.01, ### P < 0.001 vs. Coll + PBS group; n = 12). ( K – L ): PWF for contralateral ( K ) and ipsilateral ( L ) hind paws in each group, tested using 0.4 g von Frey filaments (K: two-way ANOVA followed by Tukey’s test, F (16, 220) = 5.003, P < 0.001. L: two-way ANOVA followed by Tukey’s test, F (16, 220) = 0.4380, P = 0.97; ** P < 0.01, *** P < 0.001 vs. Saline group; # P < 0.05, ## P < 0.01 vs. Coll + PBS group; n = 12). ( M ): PWL for each group of mice, measured using a 0 °C cold plate (two-way ANOVA followed by Tukey’s test, F (16, 220) = 12.15, P < 0.001; * P < 0.05, *** P < 0.001 vs. Saline group; ### P < 0.001 vs. Coll + PBS group; n = 12). ( N – O ): 21 days post-injection, PIK3CG protein in the ipsilateral VPL/VPM of mice was lower in the Coll + Danshenol B group compared to the Coll + PBS group (unpaired t -test, t = 3.633, df = 6; P = 0.0109, n = 4)

Danshenol B alleviates pain by inhibiting the PIK3CG/NLRP3 signaling pathway

To investigate the role of PIK3CG in Danshenol B’s pain-relieving effects, we administered 50 mg/kg of Danshenol B (Coll + Danshenol B group) or PBS (Coll + PBS group) daily, starting one day after CPSP model establishment, alongside lentiviral overexpression of PIK3CG (Coll + Danshenol B + LV-pik3cg-OE group). The control group received a negative control virus (Coll + Danshenol B + LV-NC group) (Fig. 7 A). Twenty-one days after collagenase IV injection, green fluorescent protein (GFP)-positive neurons, infected by the virus, were predominantly observed in the VPL/VPM (Fig. 7 B). After 21 days, the Coll + Danshenol B + LV-pik3cg-OE group showed increased mechanical and cold pain sensitivity in the contralateral hind paw compared to the Coll + Danshenol B + LV-NC group ( P < 0.001, Figs. 7 C, E, G). No differences were observed in PWF of the ipsilateral hind paw among the groups (Figs. 7 D, F). After completing the behavioral assessments on day 21, ipsilateral VPL/VPM tissues from the Coll + Danshenol B group, Coll + Danshenol B + LV-NC group, and Coll + Danshenol B + LV-pik3cg-OE group were collected for Western blot analysis. Western blot revealed elevated PIK3CG protein in the ipsilateral VPL/VPM of the Coll + Danshenol B + LV-pik3cg-OE group, relative to the Coll + Danshenol B ( P = 0.0339) and Coll + Danshenol B + LV-NC ( P = 0.0071) groups (Fig. 7 H).

Fig. 7 Danshenol B alleviates CPSP by regulating PIK3CG protein. ( A ): Collagenase IV was injected into the VPL/VPM of mice, and one day later, either the PIK3CG overexpression virus (LV-pik3cg-OE) or the control virus (LV-NC) was injected into the VPL/VPM of the two groups of mice. Meanwhile, mice were administered Danshenol B (50 mg/kg) or PBS daily via oral gavage. Pain behaviors were assessed, and on day 21, PIK3CG protein levels in the ipsilateral VPL/VPM were measured. ( B ): Observation of lentiviral-mediated GFP gene expression in VPL/VPM brain tissue using a fluorescence microscope. scale bar = 200 μm. ( C – D ): PWF for contralateral ( C ) and ipsilateral ( D ) hind paws in each group, tested using 0.07 g von Frey filaments (C: two-way ANOVA followed by Sidak test, F (12, 210) = 14.31, P < 0.001. D: two-way ANOVA followed by Sidak test, F (12, 210) = 0.6969, P = 0.75; n = 15). ( E – F ): PWF for contralateral ( E ) and ipsilateral ( F ) hind paws in each group, tested using 0.4 g von Frey filaments (E: two-way ANOVA followed by Sidak test, F (12, 210) = 13.48, P < 0.001. F: two-way ANOVA followed by Sidak test, F (12, 210) = 0.9567, P = 0.49; n = 15). ( G ): PWL for each group of mice, measured using a 0 °C cold plate (two-way ANOVA followed by Sidak test, F (12, 210) = 27.49, P < 0.001; n = 15). ( H ): 21 days post-injection, PIK3CG protein increased in the Coll + Danshenol B + LV-pik3cg-OE group relative to the Coll + Danshenol B + LV-NC group (one-way ANOVA, F (2, 9) = 8.958, P = 0.0072; n = 4)

To investigate PIK3CG’s regulatory effects on pain, we administered lentiviral shRNA to knockdown PIK3CG in the ipsilateral VPL/VPM (Coll + sh-pik3cg group), post-CPSP model establishment, with controls receiving a control virus (Coll + sh-NC group) (Fig. 8 A). By day 14 post-injection, the PWF to 0.07 g von Frey filament stimulation in the contralateral hind paw was lower in the Coll + sh-pik3cg group than in the controls ( P = 0.04, Fig. 8 B). On days 14 and 21, the PWF to 0.4 g von Frey filament stimulation in the contralateral hind paw was reduced in the Coll + sh-pik3cg group compared to the Coll + sh-NC group ( P < 0.001, Fig. 8 D). sh-pik3cg injection did not alter the PWF in the ipsilateral hind paw (Fig. 8 C, E) but increased the PWL ( P < 0.001, Fig. 8 F). After completing pain behavior assessments on day 21, ipsilateral VPL/VPM tissues from the two groups of mice were collected for immunofluorescence staining, RT-qPCR, and Western blot analysis. GFP expression was primarily observed in the VPL/VPM, confirming successful viral infection in the targeted area (Fig. 8 G). shRNA injection effectively decreased Pik3cg mRNA ( P = 0.0247, Fig. 8 H) and protein ( P = 0.0004, Fig. 8 I) levels in the VPL/VPM.

Fig. 8 Danshenol B alleviates CPSP by inhibiting the PIK3CG/NLRP3 signaling pathway. ( A ): Timeline schematic for CPSP model establishment, virus injection, and pain behavior testing time points. ( B – C ): PWF for contralateral ( B ) and ipsilateral ( C ) hind paws in each group, tested using 0.07 g von Frey filaments (B: two-way ANOVA followed by Sidak test, F (3, 66) = 1.021, P = 0.39. C: two-way ANOVA followed by Sidak test, F (3, 66) = 0.2457, P = 0.86; n = 12). ( D – E ): PWF for contralateral ( D ) and ipsilateral ( E ) hind paws in each group, tested using 0.4 g von Frey filaments ( D : two-way ANOVA followed by Sidak test, F (3, 66) = 13.47, P < 0.001. E: two-way ANOVA followed by Sidak test, F (3, 66) = 0.2635, P = 0.85; n = 12). ( F ): PWL for each group of mice, measured using a 0 °C cold plate (two-way ANOVA followed by Sidak test, F (3, 66) = 38.81, P < 0.001; n = 12). ( G ): Observation of lentiviral-mediated GFP gene expression in VPL/VPM brain tissue using a fluorescence microscope. scale bar = 500 μm. ( H ): 21 days post-injection, Pik3cg mRNA decreased in the Coll + sh-pik3cg group relative to the Coll + sh-NC group (unpaired t -test, t = 2.641, df = 10; n = 6). ( I ): 21 days post-injection, PIK3CG protein decreased in the Coll + sh-pik3cg group relative to the Coll + sh-NC group (unpaired t -test, t = 5.252, df = 10; n = 6). ( J ): 14 days post-injection, NLRP3 protein increased in the Coll group relative to the Saline group (unpaired t -test, t = 2.853, df = 13; n = 7 in Saline group, n = 8 in Coll group). ( K ): 21 days post-injection, NLRP3 protein decreased in the Coll + Danshenol B group relative to the Coll + P group (unpaired t -test, t = 16.95, df = 4; n = 3). ( L ): 21 days post-injection, NLRP3 protein increased in the Coll + Danshenol B + LV-pik3cg-OE group relative to the Coll + Danshenol B + LV-NC group (one-way ANOVA, F (2, 9) = 36.02, P < 0.001; n = 4). ( M ): 21 days post-injection, NLRP3 protein decreased in the Coll + sh-pik3cg group relative to the Coll + sh-NC group (unpaired t -test, t = 7.932, df = 4; n = 3)

Research indicates that PIK3CG activation of NLRP3 contributes to septic myocardial injury [ 40 ]. To investigate whether a similar regulatory relationship exists in CPSP, we examined NLRP3 protein expression following PIK3CG modulation in the CPSP model. NLRP3 protein was significantly elevated in the ipsilateral VPL/VPM of the Coll group compared to controls ( P = 0.0136, Fig. 8 J). Danshenol B treatment decreased NLRP3 compared to the Coll + PBS group in the Coll + Danshenol B group ( P < 0.0001, Fig. 8 K). However, PIK3CG overexpression reversed the Danshenol B-induced reduction in NLRP3 ( P < 0.001, Fig. 8 L). Knockdown of PIK3CG in the CPSP model also reduced NLRP3 ( P = 0.0014, Fig. 8 M).

PIK3CG modulates the development of CPSP through NLRP3-mediated signaling pathways

To investigate whether the regulatory effect of PIK3CG on CPSP is mediated through NLRP3, we first established a CPSP mouse model and, seven days later, injected lentiviruses encoding PIK3CG-targeting shRNA along with an NLRP3 overexpression construct into the ipsilateral VPL/VPM. Pain-related behaviors and NLRP3 protein expression in the VPL/VPM were subsequently assessed (Fig. 9 A). Fourteen days after viral injection, the PWF of the contralateral hind paw was significantly increased in the NLRP3 overexpression group compared to the control group ( P = 0.007 and P < 0.001; Figs. 9 B, D), while the PWL was significantly reduced ( P < 0.001; Fig. 9 F). In contrast, no significant difference in the PWF of the ipsilateral hind paw was observed between the two groups (Figs. 9 C, E). Consistently, Western blot analysis revealed a significant elevation in NLRP3 protein expression in the VPL/VPM of the NLRP3 overexpression group ( P < 0.001; Fig. 9 G–H). These findings indicate that NLRP3 overexpression reverses the analgesic effect induced by PIK3CG knockdown, thereby confirming that NLRP3 functions as a downstream effector of PIK3CG in the regulation of CPSP.

Fig. 9 NLRP3 overexpression counteracted the pain-relieving effects resulting from PIK3CG knockdown. ( A ): Timeline schematic for CPSP model establishment, virus injection, and pain behavior testing time points. ( B – C ): PWF for contralateral ( B ) and ipsilateral ( C ) hind paws in each group, tested using 0.07 g von Frey filaments ( B : two-way ANOVA followed by Sidak test, F (2, 44) = 3.381, P = 0.04. C: two-way ANOVA followed by Sidak test, F (2, 44) = 0.1997, P = 0.82; n = 12). ( D – E ): PWF for contralateral ( D ) and ipsilateral ( E ) hind paws in each group, tested using 0.4 g von Frey filaments ( D : two-way ANOVA followed by Sidak test, F (2, 44) = 6.671, P = 0.003. E: two-way ANOVA followed by Sidak test, F (2, 44) = 0.5016, P = 0.61; n = 12). ( F ): PWL for each group of mice, measured using a 0 °C cold plate (two-way ANOVA followed by Sidak test, F (2, 44) = 20.04, P < 0.001; n = 12). ( G ): 14 days after virus injection, NLRP3 protein increased in the Coll + sh-pik3cg + LV-NLRP3-OE group relative to the Coll + sh-pik3cg + LV-NC group (unpaired t -test, t = 16.80, df = 4; n = 3)

Elevated PIK3CG protein and mRNA levels were observed in the ipsilateral VPL/VPM of the CPSP model. Administration of Danshenol B (50 mg/kg) significantly attenuated mechanical allodynia and cold hyperalgesia, accompanied by reduced PIK3CG expression in the VPL/VPM. Overexpression of PIK3CG counteracted the analgesic effects of Danshenol B, whereas PIK3CG knockdown alleviated pain. Notably, PIK3CG overexpression led to increased NLRP3 levels, while PIK3CG knockdown suppressed NLRP3 expression. Furthermore, the analgesic effect resulting from PIK3CG knockdown was abolished by concurrent NLRP3 overexpression. These findings suggest that Danshenol B alleviates mechanical allodynia and cold hyperalgesia in the CPSP model by reducing PIK3CG and consequently lowering NLRP3 levels.

Discussion

In this study, we integrated transcriptomic sequencing, network pharmacology, and molecular docking analyses to identify potential therapeutic targets of Danshen and its active components involved in the treatment of CPSP. We demonstrated that oral administration of Danshenol B (50 mg/kg), a key bioactive compound from Danshen, significantly alleviated CPSP symptoms in mice. Mechanistic investigations revealed that Danshenol B exerts its analgesic effect by downregulating the PIK3CG/NLRP3 signaling pathway, which is known to mediate neuroinflammatory processes and chronic pain. These results provide new evidence for the therapeutic value of Danshenol B in CPSP and highlight a novel molecular mechanism underlying the analgesic effects of traditional Chinese medicine–derived compounds.

We established a CPSP model by injecting collagenase IV into the right VPL/VPM of mice, which led to significant mechanical allodynia and cold hyperalgesia, but not heat hyperalgesia. This finding suggests that the VPL/VPM nuclei may not be critical in transmitting heat pain. Previous studies have implicated the ventral tegmental area–nucleus accumbens circuit in the modulation of thermal nociception [ 41 ]; however, no direct evidence has demonstrated the involvement of VPL/VPM in heat pain processing. Consistent with this, functional MRI studies in humans have shown thalamic activation in response to tactile and mechanical stimuli, but not to thermal stimuli [ 42 ]. Alternatively, it is possible that the VPL/VPM does contribute to heat pain transmission, but that the key signaling molecules altered in our CPSP model, such as PIK3CG and NLRP3, do not mediate this modality. Supporting this possibility, transcriptomic analysis revealed no significant changes in the expression of known heat pain–related molecules, including Transient Receptor Potential Vanilloid 1 [ 43 ] and Regulators of G protein Signaling family members [ 44 ].

Transcriptomic sequencing revealed significant alterations in genes related to apoptosis and transcriptional regulation. Previous studies have established that both apoptosis [ 45 ] and transcription factors [ 46 ] are critically involved in stroke pathophysiology. In the current study, Fig. 4 C highlights the central positioning of BTK in the PPI network. BTK is a known regulator of radiation-induced apoptosis [ 47 ] and promotes cell death through p53 phosphorylation [ 48 ]. These findings suggest that exploring the role of BTK in CPSP through apoptotic mechanisms may offer valuable insights. GO analysis (Supplementary Fig. 4 E) identified significant upregulation of Mmp12 mRNA, which is associated with collagen binding. In addition, both Supplementary Fig. 8 and Fig. 4 C underscore the key role of MMP9 within the PPI network, consistent with previous findings in CPSP brain tissue [ 49 ]. Matrix metalloproteinases, a family of zinc-dependent endopeptidases, are known to regulate tissue repair [ 50 ], inflammation [ 51 ], embryonic development [ 52 ], angiogenesis [ 53 ], and cell migration [ 54 ]. Notably, prior studies have demonstrated that reducing MMP2 and MMP9 levels can alleviate mechanical allodynia in CPSP models [ 49 ], highlighting MMPs as promising therapeutic targets in post-stroke pain. These genes warrant further experimental validation in future studies.

Numerous studies have investigated Danshen for its analgesic properties. For example, compound Danshen formulations have been shown to relieve both stable [ 55 ] and unstable angina [ 56 ]. Tanshinone has demonstrated efficacy in models of chemotherapy-induced neuropathic pain [ 57 ] and cancer-related bone pain [ 58 ]. Additionally, cryptotanshinone alleviates chronic constriction injury-induced pain through inhibition of the PI3K/Akt signaling pathway [ 59 ]. However, the pain-relieving properties of several Danshen-derived components identified in our study—namely MOL007155, tanshinaldehyde, Danshenol B, and sclareol—have not been previously reported. Most existing research on Danshen’s analgesic effects has focused on compound formulations or crude extracts, with limited investigation into the roles of individual active constituents in pain management. These findings indicate substantial potential for further exploration of Danshen’s individual components in pain therapy.

This study identified a novel analgesic property of Danshenol B. To date, no prior research has reported the use of Danshenol B in the context of pain management. In our study, analgesic effects were observed in mice administered 50 mg/kg of Danshenol B via oral gavage, whereas lower doses of 5 mg/kg and 10 mg/kg did not alleviate pain symptoms. Although higher doses were not tested, it is conceivable that they may further enhance the alleviation of CPSP, potentially accelerating recovery beyond the observed reductions in mechanical allodynia and cold hyperalgesia after 14 days of treatment. Danshenol B is a diterpene compound known to be a potent inhibitor of AR [ 18 ]. AR inhibitors are widely recognized for their therapeutic role in managing diabetic peripheral neuropathy [ 19 , 60 ]. Therefore, we speculate that the analgesic effects of Danshenol B may be attributed, at least in part, to its AR-inhibitory activity.

We identified Danshenol B as a promising therapeutic compound for the treatment of CPSP. Although the candidate compounds were initially selected based on the TCMSP criteria of oral bioavailability ≥ 30% and drug-likeness ≥ 0.183, Danshenol B demonstrated favorable pharmacokinetic properties. Specifically, it possesses an oral bioavailability of 57.95% and a blood-brain barrier permeability score of 0.1, indicating limited central nervous system penetration. Its transport rate in human colon cancer cells was 0.53 nm/s, suggesting good intestinal absorption. Moreover, its fractional negative accessible surface area (63.59%) implies moderate water solubility, while a topological polar surface area of 0.29 supports adequate membrane permeability. The low number of rotatable bonds (number = 2) further suggests a high degree of molecular rigidity and metabolic stability. With a predicted in vivo half-life of 4.27 h, Danshenol B appears suitable for administration 2–3 times daily. Collectively, these parameters support its potential as a drug candidate. In preclinical studies, Danshenol B produced dose-dependent analgesic effects in the CPSP mouse model, with the 50 mg/kg dose demonstrating significant efficacy via oral gavage. However, higher doses were not investigated, and the optimal therapeutic range remains to be determined. Additionally, while no overt toxicity was observed at the tested dose, the possibility of dose-related toxicity warrants further examination. Future studies should aim to define the compound’s safety profile, therapeutic window, and pharmacodynamic characteristics. Moreover, employing drug delivery strategies such as polymer-based carriers may enhance central nervous system penetration and reduce systemic toxicity, thereby improving the clinical translational potential of Danshenol B.

In this study, we demonstrate for the first time that PIK3CG plays a critical role in the pathogenesis of CPSP and that Danshenol B exerts analgesic effects by targeting the PIK3CG/NLRP3 signaling pathway. While previous studies have implicated the PIK3CG/NLRP3 axis in pain regulation in conditions such as arthritis [ 61 ] and Complete Freund’s Adjuvant-induced neuropathic pain [ 62 ], its involvement in CPSP has not been previously reported. Moreover, although network pharmacology analyses have suggested that Danshen may exert anti-tumor effects via the PIK3CG–Akt pathway in ovarian cancer [ 63 ], no prior studies have demonstrated that Danshenol B directly regulates PIK3CG or modulates pain through this signaling axis. Our findings provide the first evidence that the PIK3CG/NLRP3 signaling cascade is activated in CPSP and that its inhibition leads to significant analgesic outcomes. Collectively, these results establish a previously unrecognized role for the PIK3CG/NLRP3 signaling axis in CPSP and provide a compelling rationale for targeting PIK3CG in future therapeutic strategies.

Although this study provides compelling preclinical evidence for the involvement of the PIK3CG/NLRP3 signaling pathway in CPSP, its clinical relevance requires further investigation. Notably, elevated PIK3CG levels have been detected in the peripheral blood of stroke patients and are correlated with poor prognosis, suggesting that PIK3CG may serve as a potential therapeutic target for improving post-stroke outcomes [ 64 ]. In parallel, NLRP3 has been implicated in a wide range of human disorders, including Alzheimer’s disease, Parkinson’s disease, and stroke [ 28 ]. However, to date, no studies have directly demonstrated the involvement of NLRP3 in human CPSP. Despite this, growing evidence supports a critical role for NLRP3 in human neuropathic pain conditions [ 65 , 66 ]. Moreover, inhibition of NLRP3 has been shown to alleviate CPSP in rodent models, further reinforcing its potential as a therapeutic target [ 6 ]. These observations imply that the PIK3CG/NLRP3 axis may similarly contribute to CPSP pathophysiology in humans. Therefore, our findings provide a mechanistic basis for future clinical studies evaluating this signaling pathway as a potential therapeutic target for post-stroke pain management.

While transcriptomic profiling is a powerful approach for identifying gene expression changes following pharmacological intervention, it may not fully reflect protein-level regulatory mechanisms. In this study, we deliberately chose not to perform RNA sequencing of VPL/VPM tissues from the Coll + Danshenol B and Coll + PBS groups, as our mechanistic focus centered on the regulation of PIK3CG at the protein level. To support this decision, we performed RT-qPCR to examine the mRNA expression of several key genes identified through network pharmacology analysis, including Btk , Casp1 , Mmp9 , and Pik3cg (Supplementary Fig. 10 ). Notably, none of these genes exhibited significant changes in transcript levels following Danshenol B treatment. This finding suggests that Danshenol B may not exert its effects through transcriptional modulation but rather through post-transcriptional mechanisms, such as altered protein degradation or post-translational modifications. Supporting this notion, our DARTS assay demonstrated that Danshenol B directly binds to PIK3CG and enhances its resistance to proteolytic degradation. Collectively, these results indicate that PIK3CG regulation by Danshenol B likely occurs at the protein level, rather than through changes in mRNA abundance, underscoring the importance of integrating bioinformatic predictions with biochemical validation to elucidate drug mechanisms.

This study has several limitations. First, although molecular docking revealed strong binding affinities, we did not validate whether other active components of Danshen, such as tanshinaldehyde, sclareol, and MOL00715, possess therapeutic potential for CPSP. Second, despite the known AR-inhibitory effect of Danshenol B, we did not investigate whether this mechanism contributes to its analgesic effects in CPSP. Lastly, although Danshenol B exhibited a favorable binding energy with PIK3CG (–9.127 kcal/mol), it did not represent the strongest interaction identified in our docking analysis. The most favorable binding was observed between BTK and MOL007155 (-9.619 kcal/mol). However, due to the unavailability of commercial sources for MOL007155, we synthesized the compound in–house to further investigate its potential interaction with BTK in the context of CPSP.

Conclusions

Our findings revealed that Danshenol B, a component of Danshen, alleviates CPSP by inhibiting the PIK3CG/NLRP3 pathway. These results suggest a potential new clinical application for Danshen extracts in the management of CPSP.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Supplementary Material 2

Supplementary Material 3

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Panyang Li and Linna Yu contributed equally to this work.

Acknowledgements

Not applicable.

Author contributions

PL: Conceptualization, Data curation, Funding acquisition, Project administration, Supervision, Visualization. LY: Data curation, Resources, Validation, Visualization, Writing–original draft. JY: Formal analysis, Software, Validation. TZ: Methodology. SZ: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing–original draft, Writing–review and editing. All authors read and approved the final manuscript.

Funding

This research was funded by the Henan Province Key Research and Development and Promotion Special Project (Science and Technology Tackling) (grant number: 242102310060 to Panyang Li).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. The RNA-seq data were uploaded with Gene Expression Omnibus accession number GSE281041 .

Declarations

Ethics approval and consent to participate

All animal care procedures were approved by the Zhengzhou University Animal Care and Use Committee and complied with the International Association for the Study of Pain guidelines. Ethical approval was obtained from the Zhengzhou University Ethics Committee (Approval No. ZZUIRB2024-191).

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work, the authors used ChatGPT in order to enhance the fluency of the writing and the logical flow of the article. After using this tool, the authors reviewed and edited the content as needed and takes full responsibility for the content of the published article.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Abbreviations

Central post-stroke pain

Salvia miltiorrhiza

Ventral posterolateral thalamic nucleus and Ventral posteromedial thalamic nucleus

Differentially expressed genes

Traditional Chinese Medicine

Phosphatidylinositol-4,5-Bisphosphate 3-Kinase Catalytic Subunit Gamma

NLR Family Pyrin Domain Containing 3

Modified Neurological Severity Score

Paw withdrawal frequency

Paw withdrawal latency

EuKaryotic Orthologous Groups

Gene Ontology

Kyoto Encyclopedia of Genes and Genomes

Gene Set EnrichmentAnalysis

Differentially expressed

Biological Processes

Molecular Functions

Cellular Components

Protein-protein interaction

Collagenase IV

Tanshinaldehyde and (6 S)-6-(hydroxymethyl)-1,6-dimethyl-8,9-dihydro-7 H-naphtho[8,7-g]benzofuran-10,11-dione

Overexpression

Negative control

Green fluorescent protein

Matrix Metalloproteinase

Aldose Reductase

References

1. Klit H Finnerup NB Jensen TS Central post-stroke pain: clinical characteristics, pathophysiology, and management Lancet Neurol 2009 8 857 68 10.1016/S1474-4422(09)70176-0 19679277 Klit H, Finnerup NB, Jensen TS. Central post-stroke pain: clinical characteristics, pathophysiology, and management. Lancet Neurol. 2009;8:857–68. 19679277 10.1016/S1474-4422(09)70176-0

2. Mohanan AT, Nithya S, Nomier Y, Hassan DA, Jali AM, Qadri M, Machanchery S. Stroke-Induced central pain: overview of the mechanisms, management, and emerging targets of central Post-Stroke pain. Pharmaceuticals (Basel) 2023, 16. 10.3390/ph16081103 PMC10459894 37631018

3. Hiraga SI, Itokazu T, Hoshiko M, Takaya H, Nishibe M, Yamashita T. Microglial depletion under thalamic hemorrhage ameliorates mechanical allodynia and suppresses aberrant axonal sprouting. JCI Insight 2020, 5. 10.1172/jci.insight.131801 PMC7098795 32051342

4. Edwards SA Ioannou A Carin-Levy G Cowey E Brady M Morton S Sande TA Mead G Quinn TJ Properties of pain assessment tools for use in people living with stroke: systematic review Front Neurol 2020 11 792 10.3389/fneur.2020.00792 32849238 PMC7431893 Edwards SA, Ioannou A, Carin-Levy G, Cowey E, Brady M, Morton S, Sande TA, Mead G, Quinn TJ. Properties of pain assessment tools for use in people living with stroke: systematic review. Front Neurol. 2020;11:792. 32849238 10.3389/fneur.2020.00792 PMC7431893

5. JS K Pharmacological management of central post-stroke pain: a practical guide CNS Drugs 2014 28 787 97 10.1007/s40263-014-0194-y 25112542 JS K. Pharmacological management of central post-stroke pain: a practical guide. CNS Drugs. 2014;28:787–97. 25112542 10.1007/s40263-014-0194-y

6. Shi ZM Jing JJ Xue ZJ Chen WJ Tang YB Chen DJ Qi XY Huang L Zou YQ Wu XZ Yang F Stellate ganglion block ameliorated central post-stroke pain with comorbid anxiety and depression through inhibiting HIF-1alpha/NLRP3 signaling following thalamic hemorrhagic stroke J Neuroinflammation 2023 20 82 10.1186/s12974-023-02765-2 36944982 PMC10031944 Shi ZM, Jing JJ, Xue ZJ, Chen WJ, Tang YB, Chen DJ, Qi XY, Huang L, Zou YQ, Wu XZ, Yang F. Stellate ganglion block ameliorated central post-stroke pain with comorbid anxiety and depression through inhibiting HIF-1alpha/NLRP3 signaling following thalamic hemorrhagic stroke. J Neuroinflammation. 2023;20:82. 36944982 10.1186/s12974-023-02765-2 PMC10031944

7. Gritsch S, Bali KK, Kuner R, Vardeh D. Functional characterization of a mouse model for central post-stroke pain. Mol Pain 2016, 12. 10.1177/1744806916629049 PMC4956143 27030713

8. XY LZ FZ L XT H HB Z WK TYSL XJ Z GH L Effect of electroacupuncture on relieving central post-stroke pain by inhibiting autophagy in the hippocampus Brain Res 2020 1733 146680 10.1016/j.brainres.2020.146680 31987731 XY LZ, FZ L, XT H, HB Z, WK TYSL, XJ Z, GH L. Effect of electroacupuncture on relieving central post-stroke pain by inhibiting autophagy in the hippocampus. Brain Res. 2020;1733:146680. 31987731 10.1016/j.brainres.2020.146680

9. Mastroiacovo F Busceti CL Biagioni F Moyanova SG Meisler MH Battaglia G Caricasole A Bruno V Nicoletti F Induction of the Wnt antagonist, Dickkopf-1, contributes to the development of neuronal death in models of brain focal ischemia J Cereb Blood Flow Metab 2009 29 264 76 10.1038/jcbfm.2008.111 18827832 Mastroiacovo F, Busceti CL, Biagioni F, Moyanova SG, Meisler MH, Battaglia G, Caricasole A, Bruno V, Nicoletti F. Induction of the Wnt antagonist, Dickkopf-1, contributes to the development of neuronal death in models of brain focal ischemia. J Cereb Blood Flow Metab. 2009;29:264–76. 18827832 10.1038/jcbfm.2008.111

10. Deng L Zhang J Chen S Wu Y Fan X Zuo T Hu Q Jiang L Yang S Dong Z miR-671-5p upregulation attenuates Blood-Brain barrier disruption in the ischemia stroke model via the NF-small ka, CyrillicB/MMP-9 signaling pathway Mol Neurobiol 2023 60 3824 38 10.1007/s12035-023-03318-7 36949221 Deng L, Zhang J, Chen S, Wu Y, Fan X, Zuo T, Hu Q, Jiang L, Yang S, Dong Z. miR-671-5p upregulation attenuates Blood-Brain barrier disruption in the ischemia stroke model via the NF-small ka, CyrillicB/MMP-9 signaling pathway. Mol Neurobiol. 2023;60:3824–38. 36949221 10.1007/s12035-023-03318-7

11. Anderson CD Biffi A Nalls MA Devan WJ Schwab K Ayres AM Valant V Ross OA Rost NS Saxena R Common variants within oxidative phosphorylation genes influence risk of ischemic stroke and intracerebral hemorrhage Stroke 2013 44 612 9 10.1161/STROKEAHA.112.672089 23362085 PMC3582722 Anderson CD, Biffi A, Nalls MA, Devan WJ, Schwab K, Ayres AM, Valant V, Ross OA, Rost NS, Saxena R, et al. Common variants within oxidative phosphorylation genes influence risk of ischemic stroke and intracerebral hemorrhage. Stroke. 2013;44:612–9. 23362085 10.1161/STROKEAHA.112.672089 PMC3582722

12. Hosomi K Yamamoto T Agari T Takeshita S Tanei T Imoto H Mori N Oshino S Kurisu K Kishima H Saitoh Y Benefit of spinal cord stimulation for patients with central poststroke pain: a retrospective multicenter study J Neurosurg 2022 136 601 12 10.3171/2020.11.JNS202999 34214987 Hosomi K, Yamamoto T, Agari T, Takeshita S, Tanei T, Imoto H, Mori N, Oshino S, Kurisu K, Kishima H, Saitoh Y. Benefit of spinal cord stimulation for patients with central poststroke pain: a retrospective multicenter study. J Neurosurg. 2022;136:601–12. 34214987 10.3171/2020.11.JNS202999

13. Hamani C Fonoff ET Parravano DC Silva VA Galhardoni R Monaco BA Navarro J Yeng LT Teixeira MJ de Andrade DC Motor cortex stimulation for chronic neuropathic pain: results of a double-blind randomized study Brain 2021 144 2994 3004 10.1093/brain/awab189 34373901 Hamani C, Fonoff ET, Parravano DC, Silva VA, Galhardoni R, Monaco BA, Navarro J, Yeng LT, Teixeira MJ, de Andrade DC. Motor cortex stimulation for chronic neuropathic pain: results of a double-blind randomized study. Brain. 2021;144:2994–3004. 34373901 10.1093/brain/awab189

14. Sheng R Chen C Chen H Yu P Repetitive transcranial magnetic stimulation for stroke rehabilitation: insights into the molecular and cellular mechanisms of neuroinflammation Front Immunol 2023 14 1197422 10.3389/fimmu.2023.1197422 37283739 PMC10239808 Sheng R, Chen C, Chen H, Yu P. Repetitive transcranial magnetic stimulation for stroke rehabilitation: insights into the molecular and cellular mechanisms of neuroinflammation. Front Immunol. 2023;14:1197422. 37283739 10.3389/fimmu.2023.1197422 PMC10239808

15. XD ME Cao YF Che YY Li J Shang ZP Zhao WJ Qiao YJ Zhang JY Danshen: a phytochemical and Pharmacological overview Chin J Nat Med 2019 17 59 80 30704625 10.1016/S1875-5364(19)30010-X XD ME, Cao YF, Che YY, Li J, Shang ZP, Zhao WJ, Qiao YJ, Zhang JY. Danshen: a phytochemical and Pharmacological overview. Chin J Nat Med. 2019;17:59–80. 30704625 10.1016/S1875-5364(19)30010-X

16. Collaboration R Effects of antiplatelet therapy after stroke due to intracerebral haemorrhage (RESTART): a randomised, open-label trial Lancet 2019 393 2613 23 10.1016/S0140-6736(19)30840-2 31128924 PMC6617509 Collaboration R. Effects of antiplatelet therapy after stroke due to intracerebral haemorrhage (RESTART): a randomised, open-label trial. Lancet. 2019;393:2613–23. 31128924 10.1016/S0140-6736(19)30840-2 PMC6617509

17. Sun Y Yang J A bioinformatics investigation into the Pharmacological mechanisms of the effect of Fufang Danshen on pain based on methodologies of network Pharmacology Sci Rep 2019 9 5913 10.1038/s41598-019-40694-4 30976033 PMC6459854 Sun Y, Yang J. A bioinformatics investigation into the Pharmacological mechanisms of the effect of Fufang Danshen on pain based on methodologies of network Pharmacology. Sci Rep. 2019;9:5913. 30976033 10.1038/s41598-019-40694-4 PMC6459854

18. Tezuka Y Kasimu R Basnet P Namba T Kadota S Aldose reductase inhibitory constituents of the root of Salvia miltiorhiza bunge Chem Pharm Bull (Tokyo) 1997 45 1306 11 10.1248/cpb.45.1306 9377780 Tezuka Y, Kasimu R, Basnet P, Namba T, Kadota S. Aldose reductase inhibitory constituents of the root of Salvia miltiorhiza bunge. Chem Pharm Bull (Tokyo). 1997;45:1306–11. 9377780 10.1248/cpb.45.1306

19. Chalk C, Benstead TJ, Moore F. Aldose reductase inhibitors for the treatment of diabetic polyneuropathy. Cochrane Database Syst Rev 2007, 2007:CD004572. 10.1002/14651858.CD004572.pub2 PMC8406996 17943821

20. Kaneda MM Messer KS Ralainirina N Li H Leem CJ Gorjestani S Woo G Nguyen AV Figueiredo CC Foubert P PI3Kgamma is a molecular switch that controls immune suppression Nature 2016 539 437 42 10.1038/nature19834 27642729 PMC5479689 Kaneda MM, Messer KS, Ralainirina N, Li H, Leem CJ, Gorjestani S, Woo G, Nguyen AV, Figueiredo CC, Foubert P, et al. PI3Kgamma is a molecular switch that controls immune suppression. Nature. 2016;539:437–42. 27642729 10.1038/nature19834 PMC5479689

21. Luo Q Raulston EG Prado MA Wu X Gritsman K Whalen KS Yan K Booth CAG Xu R van Galen P Targetable leukaemia dependency on noncanonical PI3Kgamma signalling Nature 2024 630 198 205 10.1038/s41586-024-07410-3 38720074 PMC13270182 Luo Q, Raulston EG, Prado MA, Wu X, Gritsman K, Whalen KS, Yan K, Booth CAG, Xu R, van Galen P, et al. Targetable leukaemia dependency on noncanonical PI3Kgamma signalling. Nature. 2024;630:198–205. 38720074 10.1038/s41586-024-07410-3 PMC13270182

22. Piddock RE Loughran N Marlein CR Robinson SD Edwards DR Yu S Pillinger GE Zhou Z Zaitseva L Auger MJ PI3Kdelta and PI3Kgamma isoforms have distinct functions in regulating pro-tumoural signalling in the multiple myeloma microenvironment Blood Cancer J 2017 7 e539 10.1038/bcj.2017.16 28282033 PMC5380901 Piddock RE, Loughran N, Marlein CR, Robinson SD, Edwards DR, Yu S, Pillinger GE, Zhou Z, Zaitseva L, Auger MJ, et al. PI3Kdelta and PI3Kgamma isoforms have distinct functions in regulating pro-tumoural signalling in the multiple myeloma microenvironment. Blood Cancer J. 2017;7:e539. 28282033 10.1038/bcj.2017.16 PMC5380901

23. Jin R Yu S Song Z Quillin JW Deasis DP Penninger JM Nanda A Granger DN Li G Phosphoinositide 3-kinase-gamma expression is upregulated in brain microglia and contributes to ischemia-induced microglial activation in acute experimental stroke Biochem Biophys Res Commun 2010 399 458 64 10.1016/j.bbrc.2010.07.116 20678469 PMC3169085 Jin R, Yu S, Song Z, Quillin JW, Deasis DP, Penninger JM, Nanda A, Granger DN, Li G. Phosphoinositide 3-kinase-gamma expression is upregulated in brain microglia and contributes to ischemia-induced microglial activation in acute experimental stroke. Biochem Biophys Res Commun. 2010;399:458–64. 20678469 10.1016/j.bbrc.2010.07.116 PMC3169085

24. Shang S Liu L Wu X Fan F Hu E Wang L Ding Y Zhang Y Lu X Inhibition of PI3Kgamma by AS605240 protects tMCAO mice by attenuating Pro-Inflammatory signaling and cytokine release in reactive astrocytes Neuroscience 2019 415 107 20 10.1016/j.neuroscience.2019.06.001 31195053 Shang S, Liu L, Wu X, Fan F, Hu E, Wang L, Ding Y, Zhang Y, Lu X. Inhibition of PI3Kgamma by AS605240 protects tMCAO mice by attenuating Pro-Inflammatory signaling and cytokine release in reactive astrocytes. Neuroscience. 2019;415:107–20. 31195053 10.1016/j.neuroscience.2019.06.001

25. Jin R Song Z Yu S Piazza A Nanda A Penninger JM Granger DN Li G Phosphatidylinositol-3-kinase gamma plays a central role in blood-brain barrier dysfunction in acute experimental stroke Stroke 2011 42 2033 44 10.1161/STROKEAHA.110.601369 21546487 PMC3129812 Jin R, Song Z, Yu S, Piazza A, Nanda A, Penninger JM, Granger DN, Li G. Phosphatidylinositol-3-kinase gamma plays a central role in blood-brain barrier dysfunction in acute experimental stroke. Stroke. 2011;42:2033–44. 21546487 10.1161/STROKEAHA.110.601369 PMC3129812

26. Guan X Fu Q Xiong B Song Z Shu B Bu H Xu B Manyande A Cao F Tian Y Activation of PI3Kgamma/Akt pathway mediates bone cancer pain in rats J Neurochem 2015 134 590 600 10.1111/jnc.13139 25919859 Guan X, Fu Q, Xiong B, Song Z, Shu B, Bu H, Xu B, Manyande A, Cao F, Tian Y. Activation of PI3Kgamma/Akt pathway mediates bone cancer pain in rats. J Neurochem. 2015;134:590–600. 25919859 10.1111/jnc.13139

27. Fu J Wu H Structural mechanisms of NLRP3 inflammasome assembly and activation Annu Rev Immunol 2023 41 301 16 10.1146/annurev-immunol-081022-021207 36750315 PMC10159982 Fu J, Wu H. Structural mechanisms of NLRP3 inflammasome assembly and activation. Annu Rev Immunol. 2023;41:301–16. 36750315 10.1146/annurev-immunol-081022-021207 PMC10159982

28. Coll RC Schroder K Pelegrin P NLRP3 and pyroptosis blockers for treating inflammatory diseases Trends Pharmacol Sci 2022 43 653 68 10.1016/j.tips.2022.04.003 35513901 Coll RC, Schroder K, Pelegrin P. NLRP3 and pyroptosis blockers for treating inflammatory diseases. Trends Pharmacol Sci. 2022;43:653–68. 35513901 10.1016/j.tips.2022.04.003

29. Xia CY Guo YX Lian WW Yan Y Ma BZ Cheng YC Xu JK He J Zhang WK The NLRP3 inflammasome in depression: potential mechanisms and therapies Pharmacol Res 2023 187 106625 10.1016/j.phrs.2022.106625 36563870 Xia CY, Guo YX, Lian WW, Yan Y, Ma BZ, Cheng YC, Xu JK, He J, Zhang WK. The NLRP3 inflammasome in depression: potential mechanisms and therapies. Pharmacol Res. 2023;187:106625. 36563870 10.1016/j.phrs.2022.106625

30. Li JM Deng HS Yao YD Wang WT Hu JQ Dong Y Wang PX Liu L Liu ZQ Xie Y Sinomenine ameliorates collagen-induced arthritis in mice by targeting GBP5 and regulating the P2 × 7 receptor to suppress NLRP3-related signaling pathways Acta Pharmacol Sin 2023 44 2504 24 10.1038/s41401-023-01124-4 37482570 PMC10692212 Li JM, Deng HS, Yao YD, Wang WT, Hu JQ, Dong Y, Wang PX, Liu L, Liu ZQ, Xie Y, et al. Sinomenine ameliorates collagen-induced arthritis in mice by targeting GBP5 and regulating the P2X7 receptor to suppress NLRP3-related signaling pathways. Acta Pharmacol Sin. 2023;44:2504–24. 37482570 10.1038/s41401-023-01124-4 PMC10692212

31. Takahashi M NLRP3 inflammasome as a key driver of vascular disease Cardiovasc Res 2022 118 372 85 10.1093/cvr/cvab010 33483732 Takahashi M. NLRP3 inflammasome as a key driver of vascular disease. Cardiovasc Res. 2022;118:372–85. 33483732 10.1093/cvr/cvab010

32. Li W Shen N Kong L Huang H Wang X Zhang Y Wang G Xu P Hu W STING mediates microglial pyroptosis via interaction with NLRP3 in cerebral ischaemic stroke Stroke Vasc Neurol 2024 9 153 64 10.1136/svn-2023-002320 37402504 PMC11103158 Li W, Shen N, Kong L, Huang H, Wang X, Zhang Y, Wang G, Xu P, Hu W. STING mediates microglial pyroptosis via interaction with NLRP3 in cerebral ischaemic stroke. Stroke Vasc Neurol. 2024;9:153–64. 37402504 10.1136/svn-2023-002320 PMC11103158

33. Tao YW Yang L Chen SY Zhang Y Zeng Y Wu JS Meng XL Pivotal regulatory roles of traditional Chinese medicine in ischemic stroke via Inhibition of NLRP3 inflammasome J Ethnopharmacol 2022 294 115316 10.1016/j.jep.2022.115316 35513214 Tao YW, Yang L, Chen SY, Zhang Y, Zeng Y, Wu JS, Meng XL. Pivotal regulatory roles of traditional Chinese medicine in ischemic stroke via Inhibition of NLRP3 inflammasome. J Ethnopharmacol. 2022;294:115316. 35513214 10.1016/j.jep.2022.115316

34. Schallert T Kozlowski DA Humm JL Cocke RR Use-dependent structural events in recovery of function Adv Neurol 1997 73 229 38 8959217 Schallert T, Kozlowski DA, Humm JL, Cocke RR. Use-dependent structural events in recovery of function. Adv Neurol. 1997;73:229–38. 8959217

35. Su S Yudin Y Kim N Tao YX Rohacs T TRPM3 channels play roles in heat hypersensitivity and spontaneous pain after nerve injury J Neurosci 2021 41 2457 74 10.1523/JNEUROSCI.1551-20.2020 33478988 PMC7984590 Su S, Yudin Y, Kim N, Tao YX, Rohacs T. TRPM3 channels play roles in heat hypersensitivity and spontaneous pain after nerve injury. J Neurosci. 2021;41:2457–74. 33478988 10.1523/JNEUROSCI.1551-20.2020 PMC7984590

36. Cunningham CB Benowitz KM Moore AJ The updated genome of the burying beetle Nicrophorus vespilloides, a model species for evolutionary and genetic studies of parental care Ecol Evol 2024 14 e70601 10.1002/ece3.70601 39659731 PMC11628744 Cunningham CB, Benowitz KM, Moore AJ. The updated genome of the burying beetle Nicrophorus vespilloides, a model species for evolutionary and genetic studies of parental care. Ecol Evol. 2024;14:e70601. 39659731 10.1002/ece3.70601 PMC11628744

37. Love MI Huber W Anders S Moderated Estimation of fold change and dispersion for RNA-seq data with DESeq2 Genome Biol 2014 15 550 10.1186/s13059-014-0550-8 25516281 PMC4302049 Love MI, Huber W, Anders S. Moderated Estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15:550. 25516281 10.1186/s13059-014-0550-8 PMC4302049

38. Liao Y Smyth GK Shi W FeatureCounts: an efficient general purpose program for assigning sequence reads to genomic features Bioinformatics 2014 30 923 30 10.1093/bioinformatics/btt656 24227677 Liao Y, Smyth GK, Shi W. FeatureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics. 2014;30:923–30. 24227677 10.1093/bioinformatics/btt656

39. Cozzini P Fornabaio M Marabotti A Abraham DJ Kellogg GE Mozzarelli A Simple, intuitive calculations of free energy of binding for protein-ligand complexes. 1. Models without explicit constrained water J Med Chem 2002 45 2469 83 10.1021/jm0200299 12036355 Cozzini P, Fornabaio M, Marabotti A, Abraham DJ, Kellogg GE, Mozzarelli A. Simple, intuitive calculations of free energy of binding for protein-ligand complexes. 1. Models without explicit constrained water. J Med Chem. 2002;45:2469–83. 12036355 10.1021/jm0200299

40. Lu C Liu J Escames G Yang Y Wu X Liu Q Chen J Song Y Wang Z Deng C PIK3CG regulates NLRP3/GSDMD-Mediated pyroptosis in septic myocardial injury Inflammation 2023 46 2416 32 10.1007/s10753-023-01889-0 37676465 Lu C, Liu J, Escames G, Yang Y, Wu X, Liu Q, Chen J, Song Y, Wang Z, Deng C, et al. PIK3CG regulates NLRP3/GSDMD-Mediated pyroptosis in septic myocardial injury. Inflammation. 2023;46:2416–32. 37676465 10.1007/s10753-023-01889-0

41. Zhang H Qian YL Li C Liu D Wang L Wang XY Liu MJ Liu H Zhang S Guo XY Brain-Derived neurotrophic factor in the mesolimbic reward circuitry mediates nociception in chronic neuropathic pain Biol Psychiatry 2017 82 608 18 10.1016/j.biopsych.2017.02.1180 28390647 PMC5788809 Zhang H, Qian YL, Li C, Liu D, Wang L, Wang XY, Liu MJ, Liu H, Zhang S, Guo XY, et al. Brain-Derived neurotrophic factor in the mesolimbic reward circuitry mediates nociception in chronic neuropathic pain. Biol Psychiatry. 2017;82:608–18. 28390647 10.1016/j.biopsych.2017.02.1180 PMC5788809

42. Habig K Kramer HH Lautenschlager G Walter B Best C Processing of sensory, painful and vestibular stimuli in the thalamus Brain Struct Funct 2023 228 433 47 10.1007/s00429-022-02582-y 36239796 PMC9944400 Habig K, Kramer HH, Lautenschlager G, Walter B, Best C. Processing of sensory, painful and vestibular stimuli in the thalamus. Brain Struct Funct. 2023;228:433–47. 36239796 10.1007/s00429-022-02582-y PMC9944400

43. Iftinca M Defaye M Altier C TRPV1-Targeted drugs in development for human pain conditions Drugs 2021 81 7 27 10.1007/s40265-020-01429-2 33165872 Iftinca M, Defaye M, Altier C. TRPV1-Targeted drugs in development for human pain conditions. Drugs. 2021;81:7–27. 33165872 10.1007/s40265-020-01429-2

44. Bosier B Doyen PJ Brolet A Muccioli GG Ahmed E Desmet N Hermans E Deumens R Inhibition of the regulator of G protein signalling RGS4 in the spinal cord decreases neuropathic hyperalgesia and restores cannabinoid CB1 receptor signalling Br J Pharmacol 2015 172 5333 46 10.1111/bph.13324 26478461 PMC5341217 Bosier B, Doyen PJ, Brolet A, Muccioli GG, Ahmed E, Desmet N, Hermans E, Deumens R. Inhibition of the regulator of G protein signalling RGS4 in the spinal cord decreases neuropathic hyperalgesia and restores cannabinoid CB1 receptor signalling. Br J Pharmacol. 2015;172:5333–46. 26478461 10.1111/bph.13324 PMC5341217

45. Chen X Li Z Zhang B Liu T Yao W Wan L Zhang C Zhang Y Antinociception role of 14,15-epoxyeicosatrienoic acid in a central post-stroke pain model in rats mediated by anti-inflammation and anti-apoptosis effect Neurochem Int 2022 154 105291 10.1016/j.neuint.2022.105291 35074479 Chen X, Li Z, Zhang B, Liu T, Yao W, Wan L, Zhang C, Zhang Y. Antinociception role of 14,15-epoxyeicosatrienoic acid in a central post-stroke pain model in rats mediated by anti-inflammation and anti-apoptosis effect. Neurochem Int. 2022;154:105291. 35074479 10.1016/j.neuint.2022.105291

46. Yang F Luo WJ Sun W Wang Y Wang JL Yang F Li CL Wei N Wang XL Guan SM Chen J SDF1-CXCR4 signaling maintains central Post-Stroke pain through mediation of Glial-Neuronal interactions Front Mol Neurosci 2017 10 226 10.3389/fnmol.2017.00226 28785202 PMC5519565 Yang F, Luo WJ, Sun W, Wang Y, Wang JL, Yang F, Li CL, Wei N, Wang XL, Guan SM, Chen J. SDF1-CXCR4 signaling maintains central Post-Stroke pain through mediation of Glial-Neuronal interactions. Front Mol Neurosci. 2017;10:226. 28785202 10.3389/fnmol.2017.00226 PMC5519565

47. Uckun FM Waddick KG Mahajan S Jun X Takata M Bolen J Kurosaki T BTK as a mediator of radiation-induced apoptosis in DT-40 lymphoma B cells Science 1996 273 1096 100 10.1126/science.273.5278.1096 8688094 Uckun FM, Waddick KG, Mahajan S, Jun X, Takata M, Bolen J, Kurosaki T. BTK as a mediator of radiation-induced apoptosis in DT-40 lymphoma B cells. Science. 1996;273:1096–100. 8688094 10.1126/science.273.5278.1096

48. Althubiti M Rada M Samuel J Escorsa JM Najeeb H Lee KG Lam KP Jones GD Barlev NA Macip S BTK modulates p53 activity to enhance apoptotic and senescent responses Cancer Res 2016 76 5405 14 10.1158/0008-5472.CAN-16-0690 27630139 Althubiti M, Rada M, Samuel J, Escorsa JM, Najeeb H, Lee KG, Lam KP, Jones GD, Barlev NA, Macip S. BTK modulates p53 activity to enhance apoptotic and senescent responses. Cancer Res. 2016;76:5405–14. 27630139 10.1158/0008-5472.CAN-16-0690

49. Bai Q, Han Y, Khan S, Wu T, Yang Y, Wang Y, Tang H, Li Q, Jiang W, editors. A Novel Endoplasmic Reticulum-Targeted Metal-Organic Framework-Confined Ruthenium (Ru) Nanozyme Regulation of Oxidative Stress for Central Post-Stroke Pain. Adv Healthc Mater 2024, 13:e2302526. 10.1002/adhm.202302526 37823717

50. Chang M Matrix metalloproteinase profiling and their roles in disease RSC Adv 2023 13 6304 16 10.1039/D2RA07005G 36825288 PMC9942564 Chang M. Matrix metalloproteinase profiling and their roles in disease. RSC Adv. 2023;13:6304–16. 36825288 10.1039/d2ra07005g PMC9942564

51. Lee HS, Kim WJ. The role of matrix metalloproteinase in inflammation with a focus on infectious diseases. Int J Mol Sci 2022, 23. 10.3390/ijms231810546 PMC9500641 36142454

52. Zhang L Li Y Guan CY Tian S Lv XD Li JH Ma X Xia HF Therapeutic effect of human umbilical cord-derived mesenchymal stem cells on injured rat endometrium during its chronic phase Stem Cell Res Ther 2018 9 36 10.1186/s13287-018-0777-5 29433563 PMC5810045 Zhang L, Li Y, Guan CY, Tian S, Lv XD, Li JH, Ma X, Xia HF. Therapeutic effect of human umbilical cord-derived mesenchymal stem cells on injured rat endometrium during its chronic phase. Stem Cell Res Ther. 2018;9:36. 29433563 10.1186/s13287-018-0777-5 PMC5810045

53. Fields GB Mechanisms of action of novel drugs targeting Angiogenesis-Promoting matrix metalloproteinases Front Immunol 2019 10 1278 10.3389/fimmu.2019.01278 31214203 PMC6558196 Fields GB. Mechanisms of action of novel drugs targeting Angiogenesis-Promoting matrix metalloproteinases. Front Immunol. 2019;10:1278. 31214203 10.3389/fimmu.2019.01278 PMC6558196

54. Kurt-Celep I Nihan Kilinc A Griffin M Telci D Nitrosylation of tissue transglutaminase enhances fibroblast migration and regulates MMP activation Matrix Biol 2022 105 1 16 10.1016/j.matbio.2021.10.005 34763097 Kurt-Celep I, Nihan Kilinc A, Griffin M, Telci D. Nitrosylation of tissue transglutaminase enhances fibroblast migration and regulates MMP activation. Matrix Biol. 2022;105:1–16. 34763097 10.1016/j.matbio.2021.10.005

55. Zhang GX Zhang YY Zhang XX Wang PQ Liu J Liu Q Wang Z Different network Pharmacology mechanisms of Danshen-based Fangjis in the treatment of stable angina Acta Pharmacol Sin 2018 39 952 60 10.1038/aps.2017.191 29417948 PMC6256275 Zhang GX, Zhang YY, Zhang XX, Wang PQ, Liu J, Liu Q, Wang Z. Different network Pharmacology mechanisms of Danshen-based Fangjis in the treatment of stable angina. Acta Pharmacol Sin. 2018;39:952–60. 29417948 10.1038/aps.2017.191 PMC6256275

56. Wu JR Liu S Zhang XM Zhang B Danshen injection as adjuvant treatment for unstable angina pectoris: A systematic review and meta-analysis Chin J Integr Med 2017 23 306 11 10.1007/s11655-016-2272-0 27921196 Wu JR, Liu S, Zhang XM, Zhang B. Danshen injection as adjuvant treatment for unstable angina pectoris: A systematic review and meta-analysis. Chin J Integr Med. 2017;23:306–11. 27921196 10.1007/s11655-016-2272-0

57. Di Cesare Mannelli L Piccolo M Maione F Ferraro MG Irace C De Feo V Ghelardini C Mascolo N Tanshinones from Salvia miltiorrhiza bunge revert chemotherapy-induced neuropathic pain and reduce glioblastoma cells malignancy Biomed Pharmacother 2018 105 1042 9 10.1016/j.biopha.2018.06.047 30021339 Di Cesare Mannelli L, Piccolo M, Maione F, Ferraro MG, Irace C, De Feo V, Ghelardini C, Mascolo N. Tanshinones from Salvia miltiorrhiza bunge revert chemotherapy-induced neuropathic pain and reduce glioblastoma cells malignancy. Biomed Pharmacother. 2018;105:1042–9. 30021339 10.1016/j.biopha.2018.06.047

58. Hao W Chen L Wu LF Yang F Niu JX Kaye AD Xu SY Tanshinone IIA exerts an antinociceptive effect in rats with Cancer-induced bone pain Pain Physician 2016 19 465 76 27676663 Hao W, Chen L, Wu LF, Yang F, Niu JX, Kaye AD, Xu SY. Tanshinone IIA exerts an antinociceptive effect in rats with Cancer-induced bone pain. Pain Physician. 2016;19:465–76. 27676663

59. Zhang W Suo M Yu G Zhang M Antinociceptive and anti-inflammatory effects of Cryptotanshinone through PI3K/Akt signaling pathway in a rat model of neuropathic pain Chem Biol Interact 2019 305 127 33 10.1016/j.cbi.2019.03.016 30922766 Zhang W, Suo M, Yu G, Zhang M. Antinociceptive and anti-inflammatory effects of Cryptotanshinone through PI3K/Akt signaling pathway in a rat model of neuropathic pain. Chem Biol Interact. 2019;305:127–33. 30922766 10.1016/j.cbi.2019.03.016

60. Airey M, Bennett C, Nicolucci A, Williams R. Aldose reductase inhibitors for the prevention and treatment of diabetic peripheral neuropathy. Cochrane Database Syst Rev 2000, 1996:CD002182. 10.1002/14651858.CD002182 PMC10734249 10796870

61. Xu B Xu Y Kong J Liu Y Zhang L Shen F Wang J Shen X Chen H Chrysin mitigated neuropathic pain and peripheral sensitization in knee osteoarthritis rats by repressing the RAGE/PI3K/AKT pathway regulated by HMGB1 Cytokine 2024 180 156635 10.1016/j.cyto.2024.156635 38749277 Xu B, Xu Y, Kong J, Liu Y, Zhang L, Shen F, Wang J, Shen X, Chen H. Chrysin mitigated neuropathic pain and peripheral sensitization in knee osteoarthritis rats by repressing the RAGE/PI3K/AKT pathway regulated by HMGB1. Cytokine. 2024;180:156635. 38749277 10.1016/j.cyto.2024.156635

62. Zhou M Tan W Hasimu H Liu J Gu Z Zhao J Euphorbium total triterpenes improve freund’s complete adjuvant-induced arthritis through PI3K/AKT/Bax and NF-kappaB/NLRP3 signaling pathways J Ethnopharmacol 2023 306 116146 10.1016/j.jep.2023.116146 36610673 Zhou M, Tan W, Hasimu H, Liu J, Gu Z, Zhao J. Euphorbium total triterpenes improve freund’s complete adjuvant-induced arthritis through PI3K/AKT/Bax and NF-kappaB/NLRP3 signaling pathways. J Ethnopharmacol. 2023;306:116146. 36610673 10.1016/j.jep.2023.116146

63. Xu X, Zhang Z, Liu L, Che C, Li W. Exploring the Antiovarian Cancer Mechanisms of Salvia Miltiorrhiza Bunge by Network Pharmacological Analysis and Molecular Docking. Comput Math Methods Med 2022, 2022:7895246. 10.1155/2022/7895246 PMC9726254 36483919

64. Amini H Knepp B Rodriguez F Jickling GC Hull H Carmona-Mora P Bushnell C Ander BP Sharp FR Stamova B Early peripheral blood gene expression associated with good and poor 90-day ischemic stroke outcomes J Neuroinflammation 2023 20 13 10.1186/s12974-022-02680-y 36691064 PMC9869610 Amini H, Knepp B, Rodriguez F, Jickling GC, Hull H, Carmona-Mora P, Bushnell C, Ander BP, Sharp FR, Stamova B. Early peripheral blood gene expression associated with good and poor 90-day ischemic stroke outcomes. J Neuroinflammation. 2023;20:13. 36691064 10.1186/s12974-022-02680-y PMC9869610

65. Chen R Yin C Fang J Liu B The NLRP3 inflammasome: an emerging therapeutic target for chronic pain J Neuroinflammation 2021 18 84 10.1186/s12974-021-02131-0 33785039 PMC8008529 Chen R, Yin C, Fang J, Liu B. The NLRP3 inflammasome: an emerging therapeutic target for chronic pain. J Neuroinflammation. 2021;18:84. 33785039 10.1186/s12974-021-02131-0 PMC8008529

66. Starobova H Nadar EI Vetter I The NLRP3 inflammasome: role and therapeutic potential in pain treatment Front Physiol 2020 11 1016 10.3389/fphys.2020.01016 32973552 PMC7468416 Starobova H, Nadar EI, Vetter I. The NLRP3 inflammasome: role and therapeutic potential in pain treatment. Front Physiol. 2020;11:1016. 32973552 10.3389/fphys.2020.01016 PMC7468416