Caffeic acid phenethyl ester attenuates inflammatory pain through promoting spinal microglial M1-to-M2 polarization by suppressing the PI3K/Akt/NF-κB pathway and attenuating peripheral inflammation

Abstract

Inflammatory pain is a major global health challenge, significantly affecting quality of life and emotional well-being. Current treatment options are limited and often accompanied by adverse effects. Caffeic acid phenethyl ester (CAPE), a natural compound with notable anti-inflammatory properties, has not yet been fully elucidated for its efficacy in inflammatory pain. This work examined the role of CAPE in modulating inflammatory pain. Inflammatory pain was induced in mice by administration of Complete Freund’s Adjuvant (CFA), and pain relief was assessed through mechanical and thermal sensitivity tests. Combined with network pharmacology and molecular docking analysis, the PI3K/Akt/NF-κB pathway was identified as a potential therapeutic target. Further validation was performed using Western blot, immunofluorescence, qRT-PCR, toe thickness measurement, and H&E staining of the plantar skin sections. CAPE administration produced significant reductions in CFA-induced pain and anxiety-like behaviors. Intraperitoneal administration of CAPE significantly suppressed the phosphorylation of PI3K, Akt, and NF-κB in microglia, reduced the expression of M1 microglial marker CD86 and pro-inflammatory cytokines (TNF-α, IL-1β, IL-6), and increased the expression of M2 marker CD206 and anti-inflammatory cytokines (IL-4, IL-10). Additionally, CAPE reduced paw edema and inflammatory factor levels in toe tissue. In vitro experiments further confirmed that CAPE induced the polarization of microglia from the M1 to M2 phenotype. Our results demonstrate that CAPE facilitates the transition of microglia to the M2 phenotype mediated by the PI3K/Akt/NF-κB pathway, which attenuates peripheral inflammation and subsequently diminishes inflammation-induced hypersensitivity. These results offer novel perspectives on the possible therapeutic applications of CAPE in the management of inflammatory pain.

Introduction

Chronic pain, induced by various forms of inflammation and injury, is a common and persistent clinical health issue. Approximately one-third of the global population suffers from pain, leading to billions of dollars in healthcare costs each year. ¹ Beyond physical discomfort, chronic pain often leads to psychological distress, including anxiety and depression, highlighting the urgent need for effective treatment options for inflammatory pain. While multiple therapeutic options exist, including anti-inflammatory drugs and opioids, their pain-relieving effects remain suboptimal. Furthermore, these drugs are often associated with high toxicity and severe side effects, limiting their long-term use. ² Thus, there is a pressing need to develop safer, novel analgesics.

In recent years, natural compounds have attracted increasing interest due to their favorable safety profiles and multi-target regulatory capacity in modulating diverse inflammatory processes and pain responses. Propolis is a natural substance formed by bees mixing plant components with wax and enzymes from their saliva, possessing broad biological activities and therapeutic potential. ³ CAPE is the core active ingredient of propolis, and studies have shown that it can reduce inflammation, ⁴ inhibit oxidative stress,^5,6 protect neural cells from apoptosis, ⁷ and exert antitumor effects.^8,9 Research by Wang et al. ¹⁰ found that CAPE alleviates postoperative cognitive dysfunction (POCD) by regulating microglial polarization and reducing oxidative stress. Furthermore, studies have demonstrated that prolonged administration of CAPE can alleviate neuropathic pain.^11,12 However, the specific mechanisms through which CAPE manages chronic inflammatory pain remain to be further explored.

Within the central nervous system (CNS), microglia function as resident immune cells essential for neurodevelopment and homeostatic balance, and the inflammatory response following spinal cord injury. ¹³ Similar to peripheral macrophages, microglia are highly plastic and can be rapidly activated in response to external stimuli. ¹⁴ Emerging evidence suggests that spinal microglia activation contributes significantly to the pathogenesis of diverse pain conditions. Activated microglia can polarize to the M1 phenotype, with common markers including CD86, CD16, and CD32, and release pro-inflammatory cytokines, which directly or indirectly cause pain, exacerbate inflammation, and contribute to tissue damage. In contrast, M2 microglia, marked by CD206, CD163, etc., produce anti-inflammatory cytokines, reduce inflammation, and promote tissue repair. ¹⁵ Therefore, early intervention targeting M1 polarization of microglia holds significant therapeutic potential. ¹⁵ However, it remains unclear whether CAPE is involved in regulating the polarization of activated microglia in inflammatory pain.

The phosphoinositide 3-kinase (PI3K) family is classified into three types (I, II, III) based on lipid substrate preference and structural characteristics. Currently, research on class I PI3K is the most extensive. This subclass is widely present in mammalian cells and can be activated by various membrane signals, including receptor tyrosine kinases, G protein-coupled receptors, and members of the Ras protein family. ¹⁶ The study of the PI3K signaling pathway (especially class I) mainly focuses on protein kinase B (AKT) and its downstream targets, which are critically involved in the pathogenesis of numerous diseases, including abdominal pain caused by ulcerative colitis ¹⁷ and neuropathic pain. ¹⁸ NF-κB is a transcription factor found in the cytoplasm of all cells, which, upon activation by stimuli such as endotoxins, inflammatory mediators, carcinogens, pathogens, nicotine, and tumor promoters, translocates to the nucleus and contributes essentially to inflammatory and pain processes.^19–21 Studies have shown that in inflammatory diseases, the PI3K/AKT and NF-κB signaling pathways are frequently activated concurrently.^22–24 Furthermore, the NF-κB pathway is commonly regulated by PI3K-mediated phosphorylation.^24,25

In this study, an inflammatory pain model was established in mice by CFA induction. CAPE was administered intraperitoneally to examine its effects on mechanical allodynia and thermal hyperalgesia, as well as to explore its mechanisms at both spinal and peripheral levels.

Materials and methods

Animals

Male C57BL/6 mice (6–8 weeks, 18–22 g) were obtained from the Vital River Laboratory Animal Technology Company, Beijing, China. Mice were housed at 22 ± 2°C with 45%–55% humidity. They were kept five per cage under a 12-h light/dark cycle. Food and water were provided ad libitum. All animals were acclimated for 5–7 days before experiments. The experimental protocol was approved by the Animal Ethics Committee of the First Hospital of Jiaxing (approval No. JXYY2025-006) and adhered to the national regulations on laboratory animal welfare and ethics in China, including principles of animal protection and welfare. All methods were performed in accordance with the ARRIVE guidelines. At the end of the experiment, all mice were deeply anesthetized with isoflurane inhalation and subsequently euthanized by cervical dislocation.

Drugs and drug administration

CAPE (≥99% purity; MedChemExpress, HY-N0274) was dissolved in 10% DMSO (Sigma-Aldrich), 40% PEG300 (MedChemExpress, HY-Y0873), 5% Tween-80 (MedChemExpress, HY-Y1891), and 45% saline. CAPE (3, 9, or 27 mg/kg) was injected intraperitoneally for five consecutive days starting on day 3 after CFA surgery. Another group received a single injection of CAPE on day 3. For subsequent experiments, mice were randomly assigned to five groups: Sham, CFA+Vehicle, CFA+CAPE (27 mg/kg), CFA+LY294002 (5 µg/5 µL), and CFA+CAPE (27 mg/kg) + LY294002 (5 µg/5 µL). LY294002 (MedChemExpress, HY-10108) was given by intraspinal injection.

Cell culture

Microglia BV2 cells were procured from Procell (Wuhan, China). Cells were cultured in Dulbecco’s Modified Eagle Medium (Gibco) with 10% fetal bovine serum (FBS; Gibco, Thermo Fisher Scientific) and 1% penicillin–streptomycin. Cultures were maintained in a humidified incubator at 37°C with 5% CO₂.

CCK-8 assay

The CCK-8 assay (Dojindo, Japan) was used to evaluate the effect of CAPE on LPS-treated microglia. BV2 cells were treated with CAPE (0, 20, 40, 80, or 160 µM), alone or with 1 µg/mL LPS, for 0, 12, or 24 h. After treatment, 10 µL of CCK-8 reagent was added to each well. Plates were incubated at 37°C for 1 h. The OD450 was measured using a Multiskan GO microplate reader (Thermo Fisher Scientific, USA).

CFA-induced pain model

Inflammatory pain was induced by subcutaneous injection of 30 µL CFA into the plantar surface of the left hind paw. Sham group were given an equal volume of saline.

Mechanical allodynia test

Mechanical sensitivity was evaluated with von Frey filaments (BME-404, Institute of Biological Sciences, Chinese Academy of Medical Sciences, Yunnan, China). The 50% paw withdrawal threshold (PWT) was calculated using Dixon’s up-and-down method. For testing, animals were placed in acrylic chambers on a metal mesh floor and were given 30 min to adapt. Filaments of increasing or decreasing force were applied perpendicularly to the plantar surface of the left hind paw until bending occurred. A withdrawal or licking response was considered positive. Depending on the response, the subsequent filament was either of higher or lower strength. All assessments were conducted at 10:00 AM by investigators blinded to treatment groups.

Thermal hyperalgesia test

Thermal nociception was assessed using a hot plate (ZS Dichuang, Beijing, China). The plate was maintained at 52°C. The latency to licking, shaking, or jumping was recorded as an index of pain.

CATWALK automated gait analysis

The CatWalk XT 10.0 system (Noldus Information Technology, Netherlands) was used to analyze gait. The walkway consisted of a glass plate illuminated by a red background light above and a green light reflected from below. When animals crossed the walkway, a camera beneath the glass recorded the fluorescent signal from paw contact. Parameters including maximum contact area and maximum contact mean intensity were analyzed with system software. ²⁶

Open field test (OFT)

The OFT was performed as described previously. ²⁷ The apparatus was a white opaque box divided into central and peripheral areas. Each mouse was placed in the box alone and allowed to explore freely for 10 min. Locomotor activity was recorded with a top-mounted camera and analyzed with Jiliang software (Shanghai, China). After each trial, feces were removed. The box was wiped with 75% ethanol and dried with a cloth. It was then left to air-dry before testing the next animal.

Light-dark shuttle test (LDT)

The light-dark shuttle box (Zhongshi Technology, China) was used for assessing anxiety-like behavior. The box measured 40 cm × 20 cm × 28 cm, was split into two equal areas, designated as light and dark chambers, connected by a small passage. Prior to testing, mice were allowed to acclimate to the experimental environment for 30 min. Each mouse was tested individually. After placement in the light chamber, for a duration of 10 min, mice were allowed unrestricted movement. The duration in the light chamber and the frequency of crossings between chambers were recorded.

Intrathecal injection

A single intrathecal injection was administered to awake mice. ²⁸ A 30-G needle was inserted vertically into the L3-L4 intervertebral space to reach the subarachnoid space. The needle was advanced caudally by 0.5 cm. After confirming correct placement, 5 µL of solution was injected at a constant rate over 10 s. No anesthesia was used to avoid interference with lidocaine effects or motor function assessment. No motor dysfunction was observed immediately after injection.

H&E staining

The left hind paw of mice was fixed overnight in 4% paraformaldehyde solution (BL539A, Biosharp, Hefei, China). Samples were washed in distilled water for 3–5 h and embedded in a wax-resin mixture. Sections of 5 µm thickness were prepared. They were processed according to Hua et al., ²⁹ dewaxed, rehydrated, and stained with H&E. Sections were examined under a light microscope (Olympus BX51, Japan).

Western blot

Lumbar spinal cord tissue was collected after isoflurane anesthesia and lysed in RIPA buffer with protease and phosphatase inhibitors. Lysates were centrifuged at 12,000 rpm for 10 min at 4°C, and protein concentrations were determined using a BCA assay. Equal amounts of protein were separated by SDS-PAGE, transferred to 0.45 µm PVDF membranes, blocked, and incubated overnight at 4°C with primary antibodies: PI3K (1:1000, Affinity, AF6241), p-PI3K (1:1000, Affinity, AF3242), AKT (1:2000, Affinity, AF6261), p-AKT (1:1000, CST, 4060), NF-κB (1:1000, Affinity, AF5006), p-NF-κB (1:1000, Affinity, AF2006), β-actin (1:50,000, Abclonal, AC026), GAPDH (1:50,000, Abclonal, A19056), CD86 (1:1000, CST, #19589), CD206 (1:1000, CST, #24595). Membranes were washed and incubated with HRP-conjugated goat anti-rabbit IgG (1:20,000, Servicebio, GB23303) for 2 h at room temperature. Blots were visualized with ECL reagent (Epizyme Biotech), and band intensity was quantified using ImageJ.

Immunofluorescence

After perfusion, lumbar spinal cord tissue was fixed in 4% paraformaldehyde at 4°C for 24 h and then dehydrated in 15%–30% sucrose at 4°C for 48 h. Sections were permeabilized with 0.2% Triton X-100 for 15 min and blocked with 5% BSA for 1 h at room temperature. Primary antibodies were incubated overnight at 4°C: Iba-1 (1:500, Abcam, ab5076), CD86 (1:1000, CST, #19589), and CD206 (1:1000, CST, #24595). The following day, sections were treated with Alexa Fluor-488 or Alexa Fluor-594 secondary antibodies for 1 h at room temperature in the dark. Images were captured with a fluorescence microscope (CKX41SF, Olympus, Japan).

Real-time quantitative PCR (RT-qPCR)

Total RNA was extracted from lumbar spinal cord, plantar skin, and cultured cells using Trizol reagent (Invitrogen, USA). RNA concentration was measured with a NanoDrop 2000 (Thermo). cDNA was synthesized from 1000 ng RNA using a reverse transcription kit (Takara, Japan). qPCR was performed with SYBR Green, and GAPDH served as the internal control. Relative expression was calculated by the 2^−ΔΔCT method. Primers were designed and synthesized by Sangon Biotech (Shanghai, China), with sequences listed in Table 1.

Table 1.

Primers used for RT-qPCR.

Target genes	Forward primer sequence (5′-3′)	Reverse primer sequence (5′-3′)
CD86	ACGGAGTCAATGAAGATTTCCT	GATTCGGCTTCTTGTGACATAC
CD206	CCTATGAAAATTGGGCTTACGG	CTGACAAATCCAGTTGTTGAGG
TNF-α	CGTCAGCCGATTTGCTATCT	CGGACTCCGCAAAGTCTAAG
IL-1β	TGACGGACCCCAAAAGATGA	TCTCCACAGCCACAATGAGT
IL-6	AGACTTCCATCCAGTTGCCT	AGTGCATCATCGTTGTTCATACA
IL-4	CATGGGAAAACTCCATGCTT	TGGACTCATTCATGGTGCAG
IL-10	GCTCCTAGAGCTGCGGACT	TCATTTCCGATAAGGCTTGG
GAPDH	ACCACCATGGAGAAGGCTGG	CTCAGTGTAGCCCAGGATGC

Network pharmacology and molecular docking analysis

CAPE’s SMILES structure was acquired from PubChem. Targets were predicted with the SwissTargetPrediction platform and supplemented with Bionet and CTD databases. Results were merged and duplicates were removed to generate potential targets of CAPE. Inflammatory pain-related targets were collected from GeneCards, NCBI, and OMIM databases using “inflammatory pain” as the keyword. Results were merged and duplicates were removed to establish disease-related targets. The intersection of drug and disease targets was identified with the jvenn tool, and potential CAPE targets for inflammatory pain were obtained. A drug-component-target network was constructed with Cytoscape (v3.10.1), and nodes with higher betweenness centrality, closeness centrality, and degree values were selected as core targets. Common targets were uploaded to the STRING database with a confidence threshold of >0.4, and free nodes were removed. The protein-protein interaction (PPI) network was visualized and analyzed with Cytoscape (v3.10.1). The DAVID database was employed for GO annotation and KEGG enrichment analysis of the targets. Significant pathways and their targets were submitted to the KEGG database, and pathway–target diagrams were generated with the MicroBioinformatics platform (www.bioinformatics.com.cn). Crystal structures of target proteins were obtained with Chem3D 19.0. Docking between CAPE and core targets was performed with AutoDockTools 1.5.7, and binding energy was calculated to evaluate affinity. Docking results were visualized with PyMOL 2.5.4.

Statistical analysis

All analyses were performed using GraphPad Prism 10 (GraphPad Software, San Diego, CA). Data are expressed as mean ± SEM. Two-group comparisons were conducted with t-tests, while multiple groups were analyzed by one-way ANOVA followed by Tukey’s post hoc test. Two-way repeated-measures ANOVA with Bonferroni correction was applied for experiments with two factors. Significance was considered at p < 0.05.

Results

CAPE alleviates CFA-induced mechanical and thermal hyperalgesia in mice

A CFA-induced inflammatory pain model was established in mice. Compared with the Sham group, CFA-treated mice showed reduced PWT and paw withdrawal latency (PWL) from days 1 to 9 after surgery (Figure 1(a) and (b)). Swelling of the left paw was observed, and plantar thickness increased over time (Figure 1(c)–(f)). H&E staining revealed pronounced infiltration of inflammatory cells. Inflammation scores were significantly elevated on day 3 (Figure 1(g)–(j)). These results confirmed successful establishment of the model. To examine the effect of single-dose CAPE, based on prior studies, ¹¹ mice received intraperitoneal injections of CAPE (3, 9, 27 mg/kg) on day 3 after CFA surgery. Behavioral evaluations were performed 1 h before injection (baseline) and subsequently at 1, 2, 3, 4, 5, and 6 h after injection (Figure 1(k)). CAPE at 9 and 27 mg/kg significantly increased PWT and PWL. Effects lasted 4 and 5 h, respectively, and were dose-dependent. CAPE at 3 mg/kg had no significant effect (Figure 1(l) and (m)). To further evaluate continuous administration, mice received intraperitoneal CAPE (3, 9, 27 mg/kg) once daily for 5 days, starting on day 3 after CFA induction (Figure 1(n)). Behavioral testing was conducted 1 h after each injection. CAPE at 9 and 27 mg/kg dose-dependently increased PWT and PWL. Effects persisted for 1–2 days after discontinuation. CAPE at 3 mg/kg again showed no significant effect (Figure 1(o) and (p)). Together, the results indicate that CAPE can alleviate hyperalgesia, while repeated dosing leads to prolonged analgesia.

Figure 1.

CAPE alleviates CFA-induced inflammatory pain in mice. (a–b) From postoperative days 1 to 9, ipsilateral hind paw PWT and PWL were significantly lower in CFA-treated mice than in Sham mice (n = 8). ***p < 0.001 versus Sham group, two-way ANOVA. (c–f) Representative images of hind paw appearance and plantar thickness in Sham and CFA mice on days 3 and 7 (n = 8). (g–j) H&E staining of plantar skin and inflammation scores (n = 4). Scale bar = 50 µm. (k) Timeline of single intraperitoneal (i.p.) CAPE administration. (l, m) Effects of single i.p. CAPE at different doses (3, 9, 27 mg/kg) on PWT and PWL (n = 8). (n) Timeline of repeated i.p. CAPE administration. (o, p) Effects of repeated i.p. CAPE at different doses (3, 9, 27 mg/kg) on PWT and PWL (n = 8). (l, m; o, p) *p < 0.05, **p < 0.01, ***p < 0.001 versus CFA + Vehicle group, two-way ANOVA; (f, j) ***p < 0.001 versus Sham group, one-way ANOVA.

CAPE attenuates pain-related and anxiety-like behaviors in CFA-treated mice

The CatWalk gait analysis system was used to evaluate mice treated with CAPE (27 mg/kg, i.p., once daily for 5 days). Pain-like behavior induced by CFA was assessed by measuring maximum contact intensity and maximum contact area of both hind paws. On day 7, both parameters were significantly reduced in CFA+Vehicle mice compared with Sham (Figure 2(a)–(c)), indicating successful induction of pain-like behavior. CAPE treatment significantly improved both measures. Exploratory behavior was assessed using the OFT. After CFA modeling, CFA+Vehicle mice spent less time in the central zone and traveled shorter distances compared with Sham mice, indicating anxiety-like behavior. CAPE treatment reversed these abnormalities (Figure 2(d)–(f)). Anxiety-like behavior was also evaluated using the LDT. CFA+Vehicle mice showed reduced distance traveled, dwell time, and number of transitions into the light chamber compared with Sham mice. CAPE treatment significantly improved these measures (Figure 2(g)–(j)). Together, these results demonstrate that CAPE alleviates CFA-induced pain-like and anxiety-like behaviors in mice.

Figure 2.

CAPE ameliorates CFA-induced pain-like and anxiety-like behaviors in mice. (a) Representative CatWalk gait footprints. (b, c) Comparison of gait parameters among Sham, CFA+Vehicle, and CFA+CAPE (27 mg/kg) group, including maximum contact area (b) and maximum contact intensity (c) (n = 6). Data are expressed as left hind limb (LH) to right hind limb (RH) ratios. (d) Representative movement trajectories in the OFT. (e, f) Distance traveled (e) and dwell time in the central zone (f) (n = 8). (g) Representative trajectories in the LDT. (h–j) Time spent in the light chamber (h) number of transitions (i) and distance traveled in the light chamber (j) (n = 8). *p < 0.05, **p < 0.01, ***p < 0.001 versus CFA+Vehicle group, one-way ANOVA.

CAPE regulates microglial activation and polarization

Prior research indicated that modulating microglial activation and promoting M1 to M2 polarization can reduce inflammatory responses. ³⁰ In this study, immunofluorescence staining for Iba-1⁺ microglia revealed increased cell numbers in CFA+Vehicle mice. CAPE treatment reduced this increase (Figure 3(a) and (b)). Morphological analysis showed reduced number of endpoints (Figure 3(c)) and total process length (Figure 3(d)) in CFA mice. CAPE reversed these changes. These observations support the possibility that CAPE fosters the M1-to-M2 transition.

Figure 3.

CAPE promotes M1-to-M2 polarization of spinal microglia in CFA-treated mice. (a) Representative immunofluorescence images of Iba-1+ microglia and skeletonized images for morphological analysis (Scale bars, 200 µm (left) and 20 µm (right)). (b) Quantification of Iba-1⁺ cell numbers (n = 4). (c, d) Quantification of endpoint number (c) and total process length (d) (n = 10). (e–g) Immunofluorescence staining and quantification of CD86 and CD206 in spinal dorsal horn microglia (n = 4). Scale bars, 200 µm. (h–j) Western blot analysis of CD86 and CD206 protein expression (n = 4). (k–q) RT-qPCR analysis of CD86, CD206, IL-1β, IL-4, IL-6, IL-10, and TNF-α mRNA expression (n = 4–6). *p < 0.05, **p < 0.01, ***p < 0.001 versus CFA + Vehicle group, one-way ANOVA.

To confirm this, immunofluorescence staining was performed for the M1 marker CD86 and the M2 marker CD206 (Figure 3(e) and (f)). CFA+Vehicle mice showed higher CD86 fluorescence intensity but unchanged CD206 compared with Sham. CAPE treatment decreased CD86 and increased CD206 (Figure 3(g)).

Western blotting confirmed these results. CD86 protein was increased in CFA mice, whereas CD206 was unchanged. CAPE downregulated CD86 and upregulated CD206 (Figure 3(h)–(j)). qPCR analysis showed that CD86 and the proinflammatory cytokines were elevated in CFA mice. CAPE reduced their expression. In contrast, CD206 and the anti-inflammatory cytokines were decreased in CFA mice but increased with CAPE treatment (Figure 3(k)–(q)). Together, these findings indicate that CAPE suppresses microglial hyperactivation and promotes polarization from M1 to M2, thereby alleviating inflammatory pain.

Network pharmacology and molecular docking reveal potential targets of CAPE

Network pharmacology analysis and molecular docking were performed to explore the mechanism of CAPE in inflammatory pain. The chemical structure of CAPE is shown in Figure 4(a). Venn diagram analysis identified 390 common targets between CAPE and inflammatory pain (Figure 4(b)). A protein–protein interaction (PPI) network was constructed with 70 nodes and 1888 edges. Core targets included NFKB1, Akt1, CASP3, and IL-6 (Figure 4(c)). GO enrichment analysis identified 1109 significant terms, including 817 biological processes (BP), 107 cellular components (CC), and 185 molecular functions (MF). Key BPs included negative regulation of apoptosis, positive regulation of phosphorylation, and inflammatory response. CCs were mainly cytoplasm, extracellular region, and exosomes. MFs included homodimerization, enzyme binding, and ATP binding (Figure 4(d)). KEGG enrichment analysis revealed 187 pathways, including pathways in cancer, AGE-RAGE signaling, and PI3K–Akt signaling (Figure 4(e)). Prior research indicated that the PI3K/Akt/NF-κB pathway regulates microglial polarization and inflammatory response. ³⁰ NF-κB also regulates inflammatory mediators.^31–33 Based on enrichment results and literature, CAPE was predicted to promote M2 polarization of microglia through PI3K/Akt/NF-κB signaling. Molecular docking confirmed strong binding between CAPE and PI3K, Akt, and NF-κB1. Binding energies are shown in Table 2, and docking conformations are presented in Figure 4(f) to (h). Together, these findings suggest that CAPE alleviates inflammatory pain by acting on the PI3K/Akt/NF-κB signaling pathway.

Figure 4.

Predicted targets and pathways of CAPE in inflammatory pain revealed by network pharmacology. (a) Chemical structure of CAPE. (b) Venn diagram showing 390 common targets between CAPE and inflammatory pain. (c) PPI network of 70 core targets. (d) Top 10 enriched GO terms in BP, CC, and MF. (e) Bubble chart of the top 20 enriched KEGG pathways. (f–h) Molecular docking of CAPE with PIK3CG (f), AKT1 (g), and NF-κB1 (h).

Table 2.

CAPE’s actives docked to key targets scored.

Compounds	Targets	Binding energy (Kcal/mol)	Uniprot ID
CAPE	PIK3CG	−6.1	P48736
CAPE	AKT1	−5.7	P31749
CAPE	NF-κB1	−6.4	P19838

CAPE promotes M2 polarization by inhibiting the PI3K/Akt/NF-κB pathway in the spinal dorsal horn of CFA mice

To confirm these findings, the phosphorylation level of this molecule was initially assessed. Western blotting showed that phosphorylation of PI3K, Akt, and NF-κB was significantly increased in CFA +Vehicle mice, while total protein levels remained unchanged (Figure 5(a)). CAPE treatment (27 mg/kg, i.p., once daily for 5 days) significantly reduced these phosphorylation levels (Figure 5(b)–(d)). To further confirm the role of PI3K/Akt/NF-κB signaling in CAPE-mediated analgesia, the PI3K inhibitor LY294002 (5 µg/5 µL, i.t., once daily for 5 days) was administered in CFA mice. ³⁴ PI3K inhibition significantly increased PWT and PWL, producing effects comparable to CAPE. Combined treatment with CAPE and LY294002 did not show additional analgesic effects (Figure 5(e) and (f)). Anxiety-like behavior was assessed using the OFT and LDT tests. OFT results showed no significant differences in central zone time or distance traveled across the three groups (Figure 5(g)–(i)). Similarly, in the LDT, distance traveled, time spent, and number of transitions into the light chamber were comparable among groups (Figure 5(j)–(m)). These findings indicate that CAPE and LY294002 did not affect exploratory or anxiety-like behavior under these conditions. These results indicate that CAPE alleviates CFA-induced hyperalgesia and anxiety-like behavior by inhibiting the spinal PI3K/Akt/NF-κB pathway. The effect was consistent with PI3K inhibition, and no synergistic effect was observed with combined treatment.

Figure 5.

CAPE alleviates CFA-induced inflammatory pain by modulating the PI3K/Akt/NF-κB pathway in the spinal dorsal horn. (a–d) Spinal cord expression of p-PI3K/PI3K, p-Akt/Akt, and p-NF-κB/NF-κB was evaluated after intraperitoneal injection (n = 4). (e, f) Effects of CAPE (i.p.) and the PI3K inhibitor LY294002 (i.t.) on CFA-induced mechanical (e) and thermal (f) hyperalgesia. Combined treatment did not produce synergistic effects (n = 8). (g–i) representative trajectories (g), Central zone locomotor distance (h) and dwell time (i) in the OFT (n = 8). (j–m) representative trajectories (j), Dwell time (k), Distance traveled (l) and number of transitions (m) in the LDT (n = 6). (b–d; h, i; k–m) *p < 0.05, **p < 0.01, ***p < 0.001 for pairwise comparisons among the CFA+Vehicle, CFA+LY294002, and CFA+LY294002+CAPE groups (one-way ANOVA); ns, not significant. (e, f) *p < 0.05, ***p < 0.001 versus CFA + Vehicle group, two-way ANOVA.

CAPE promotes M1 to M2 polarization of microglia in vitro

An in vitro inflammatory model was established by stimulating BV2 microglia with lipopolysaccharide (LPS, 1 µg/mL) for 6 h. Cell viability was first assessed with the CCK-8 assay. CAPE at 160 µM reduced viability at 12 and 24 h (Figure 6(a)). Therefore, 80 µM was chosen for subsequent experiments. RT-qPCR analysis showed that CAPE significantly reduced mRNA expression of the M1 marker CD86 and the proinflammatory cytokines TNF-α, IL-1β, and IL-6 in LPS-stimulated BV2 cells. CAPE also increased expression of the M2 marker CD206 and the anti-inflammatory cytokines IL-4 and IL-10 (Figure 6(b)–(h)). These results suggest that CAPE promotes M1-to-M2 polarization in vitro. To further investigate the mechanism, primary microglia were pretreated with the PI3K inhibitor LY294002 (10 µM) for 1 h before LPS stimulation. ³⁰ Both CAPE and CAPE+LY294002 suppressed LPS-induced PI3K, Akt, and NF-κB phosphorylation, showing no intergroup significance (Figure 6(i)–(k)). The effects were comparable, suggesting that CAPE may promote M2 polarization by inhibiting the PI3K/Akt/NF-κB pathway.

Figure 6.

Effect of CAPE on LPS-induced polarization of BV2 microglia. (a) CCK-8 assay of BV2 cell viability after CAPE treatment at different concentrations (n = 3). ***p < 0.001 versus Control group, one-way ANOVA. (b–h) RT-qPCR analysis of mRNA expression of M1/M2 markers and inflammatory cytokines in LPS-stimulated BV2 cells treated with CAPE (n = 4–6). (i–k) Expression of p-PI3K/PI3K, p-Akt/Akt, and p-NF-κB/NF-κB in LPS-stimulated BV2 cells after CAPE exposure (n = 4). *p < 0.05, **p < 0.01, ***p < 0.001, one-way ANOVA; ns, not significant.

CAPE alleviates CFA-induced plantar skin inflammation in mice

The effect of CAPE on plantar skin inflammation was further evaluated. On day 7 after modeling, plantar thickness was 2.21 ± 0.10 mm in Sham mice, 4.14 ± 0.13 mm in CFA+Vehicle mice, and 3.45 ± 0.13 mm in CFA+CAPE mice. These results indicated that CAPE reduced CFA-induced paw edema (Figure 7(a) and (b)). Plantar skin was collected 4 h after CAPE treatment for H&E staining and qPCR. H&E staining showed extensive lymphocyte infiltration and mild neutrophil infiltration in CFA+Vehicle mice compared with Sham. CAPE treatment significantly reduced these infiltrates H&E staining revealed that intraperitoneal administration of CAPE significantly reduced the number of inflammatory cells in the paw skin of CFA-treated mice and markedly lowered the inflammatory score (Figure 7(c) and (d)). RT-qPCR analysis further confirmed these findings. CFA increased mRNA expression of IL-6, IL-1β, and TNF-α, while CAPE treatment reduced their expression (Figure 7(e)–(g)). Together, these results demonstrate that CAPE alleviates CFA-induced paw edema and local inflammatory responses.

Figure 7.

CAPE attenuates CFA-induced plantar skin inflammation in mice. (a) Representative images of the left hind paw in Sham, CFA+Vehicle, and CFA + CAPE groups. (b) Plantar thickness changes over time (n = 8). ***p < 0.001 versus CFA+Vehicle, two-way ANOVA. (c, d) H&E staining of paw skin in different groups (n = 4). Scale bar = 50 µm. (e–g) RT-qPCR analysis of TNF-α, IL-1β, and IL-6 mRNA expression in plantar skin tissue (n = 4–6). (d–g) **p < 0.01, ***p < 0.001 versus CFA + Vehicle group, one-way ANOVA.

Discussion

This study demonstrates that intraperitoneal administration of CAPE alleviates mechanical pain, thermal pain, anxiety-like behaviors, and peripheral inflammation. Specifically, CAPE suppresses microglial polarization and M1 phenotype activation both in vivo and in vitro reduces the release of pro-inflammatory cytokines. Through network pharmacology and molecular docking analysis, followed by a series of experimental validations, we found that CAPE regulates microglial polarization from M1 to M2 by inhibiting the expression of p-PI3K, p-Akt, and p-NF-κB. Additionally, CAPE remarkably alleviates the release of peripheral inflammatory factors and paw edema, thereby improving inflammatory pain (Figure 8).

Figure 8.

Mechanistic scheme of CAPE in inflammatory pain relief.

The bioactive component of propolis, CAPE, has gained attention due to its potent antioxidant properties. Previous studies have shown that repeated use of CAPE alleviates neuropathic pain induced by CCI. ¹¹ In this study, a single injection of CAPE temporarily relieved CFA-induced hyperalgesia. Doses of 9 and 27 mg/kg exhibited analgesic effects within 1 h of administration, with durations of 4 and 5 h, respectively. Continuous intraperitoneal injection of CAPE from day 3 to day 7 post-CFA treatment sustained analgesic effects until days 8 and 9. Additionally, anxiety-like behaviors were significantly improved. These results suggest that CAPE has therapeutic potential for both neuropathic and inflammatory pain. Because the highest dose tested (27 mg/kg) already produced robust analgesia, we did not escalate beyond this level; however, a formal dose–response curve and safety assessment at higher doses are clearly warranted.

The pathogenesis of inflammatory pain is complex, with the activation of microglia in the CNS playing a key role in its initiation and progression. ³⁶ Once activated, microglia shift to the pro-inflammatory M1 phenotype, while the M2 phenotype also compensatorily increases to promote tissue repair. ³⁷ M1 microglia release inflammatory cytokines and chemokines (such as CCL2, CCL3), which exacerbate the inflammatory response, trigger neuronal sensitization, and lead to CFA-induced inflammatory pain. Recent in vivo studies have shown that CAPE can cross the blood-brain barrier in rats,^38,39 enhancing its effects in the CNS. In this study, intraperitoneal administration of CAPE inhibited microglial activation, with a marked decrease in the fluorescence intensity of CD86, while CD206 was enhanced. Inhibition of macrophage M1 polarization can reduce the release of inflammatory factors. ⁴⁰ Consistent with this, the protein expression and mRNA levels of CD86 in the CAPE group were markedly suppressed, and the mRNA levels of pro-inflammatory cytokines were also reduced. Additionally, CAPE promoted the protein expression and mRNA levels of CD206, while the mRNA levels of anti-inflammatory cytokines were increased. These results were also validated in vitro using cultured microglial cells. Therefore, this study confirms that CAPE modulates microglial polarization, promoting the shift from the M1 to the M2 phenotype.

PI3K is a class of signaling lipases, and Akt is an important messenger in the PI3K signaling pathway. It regulates downstream effector molecules such as NF-κB, mTOR, and GSK3β through a phosphorylation cascade. This cascade is controlled by lipid and protein phosphatases and regulates cellular biological processes such as growth, proliferation, and survival. ⁴¹ Its impact varies depending on disease models and tissue types. For example, in experimental models of aneurysmal subarachnoid hemorrhage, this signaling pathway is markedly suppressed, ⁴² while in osteoarthritis models, it is activated. ⁴³ Inhibition of this signaling pathway alleviates various types of pain, including osteoarthritis pain and bone cancer pain.^44,45 NF-κB, as a transcription factor, plays a key role in the complex regulation of inflammation and pain responses. It is involved in skin inflammation, ⁴⁶ and bone cancer pain. ⁴⁷ Its activity is closely related to the PI3K/Akt pathway, but the degree of association varies depending on the disease type. Studies show that activation of this pathway in dry eye disease and UV-induced skin aging models triggers downstream NF-κB activation, leading to inflammatory damage. ⁴⁸ However, in Parkinson’s disease and brain ischemia/reperfusion injury models, inhibition of the PI3K/Akt pathway enhances NF-κB activity, inducing inflammation and promoting M1 microglial polarization. ⁴⁹ Our study found that CAPE effectively alleviates CFA-induced inflammatory pain and associated negative emotional responses. We employed network pharmacology to identify the most highly-ranked signaling pathways. Further experiments confirmed that this pathway is involved in the development of inflammatory pain. In the model group, the expression of phosphorylated proteins was significantly upregulated, which was markedly suppressed following intraperitoneal administration of CAPE. When the PI3K inhibitor LY294002 was intrathecally administered to the model group, it not only downregulated the levels of phosphorylated PI3K but also significantly suppressed the levels of phosphorylated NF-κB. The combined use produced an effect comparable to that of either agent alone. Further studies revealed that the elevated effect of the M1 marker CD86 after treatment or inhibitor administration was reversed, while the M2 marker CD206 levels were further increased, and the release of anti-inflammatory cytokines was enhanced, indicating that microglia underwent a shift from the M1 to M2 phenotype. Additionally, these findings were further validated in microglial cell-based in vitro experiments. Therefore, our study strongly supports that CAPE alleviates inflammatory pain by inhibiting the PI3K/Akt/NF-κB signaling pathway in the spinal cord and promoting M2 polarization of microglia. In view of CAPE’s marked systemic anti-inflammatory properties, its ability to dampen spinal cord inflammation may partly reflect diminished delivery of peripheral inflammatory mediators – such as cytokines and chemokines – into the central nervous system. Consequently, whether the compound acts directly on spinal targets remains an open issue that merits further investigation.

Given the systemic effects of intraperitoneal injection, CAPE’s peripheral mechanisms involve suppression of local macrophage-driven inflammation and synergistic inflammatory mediators in CFA-induced tissue damage, thereby reducing pain hypersensitivity. ⁵⁰ In this study, CFA-treated mice exhibited significant paw edema, and H&E staining showed extensive infiltration of inflammatory cells and elevated expression of pro-inflammatory cytokines in the skin tissue. These symptoms were significantly improved after CAPE treatment. Research by Lim et al. ⁵¹ showed that CAPE alleviates skin inflammation by inhibiting NF-κB activation. Other studies have also confirmed that CAPE delays the progression of osteoarthritis by activating the NRF2/HO-1 pathway and inhibiting NF-κB signaling. ⁵² These findings are consistent with ours, further supporting the peripheral anti-inflammatory effects of CAPE and providing strong evidence for its potential in the treatment of inflammatory pain.

Centering on microglial polarization, the present study did not extensively explore CAPE’s actions on astrocytes; thus, direct evidence for its modulation of astrocytic activity remains lacking. Consequently, elucidating the microglia–astrocyte crosstalk network is urgently required to fully understand CAPE’s antinociceptive effects in the CNS.

Conclusion

Taken together, the results show that CAPE effectively alleviates CFA-induced inflammatory pain. Its mechanism involves the regulation of the PI3K/Akt/NF-κB pathway in spinal microglia, promoting the shift from the M1 to M2 phenotype. Additionally, CAPE alleviates peripheral pain by inhibiting tissue inflammation.

Supplemental Material

sj-docx-1-mpx-10.1177_17448069251410746 – Supplemental material for Caffeic acid phenethyl ester attenuates inflammatory pain through promoting spinal microglial M1-to-M2 polarization by suppressing the PI3K/Akt/NF-κB pathway and attenuating peripheral inflammation

Supplemental material, sj-docx-1-mpx-10.1177_17448069251410746 for Caffeic acid phenethyl ester attenuates inflammatory pain through promoting spinal microglial M1-to-M2 polarization by suppressing the PI3K/Akt/NF-κB pathway and attenuating peripheral inflammation by Dongjie Wang, Yuhua Li, Chaobo Ni, Longsheng Xu, Xiaogeng Huang, Shuyao Zhang, Guofeng Shen, Heng Zhang, Huadong Ni, Ming Yao, Xuewu Lin and Gang Liu in Molecular Pain