A programmable nanoplatform for inflammatory-state reprogramming in psoriasis treatment

Abstract

Psoriasis is a chronic, immune-mediated inflammatory skin disease. Although current treatments alleviate symptoms, they primarily focus on inflammation suppression through single-target approaches, which fail to comprehensively address the complex pathological imbalances. To overcome this limitation, we developed a programmable delivery system based on a nanocomposite material composed of Tannic acid (TA), Cerium (Ce), and Luteolin (Lut), integrated into an in situ gelling hydrogel platform (TA-Ce-Lut-Gel), enabling multi-dimensional treatment strategy for psoriasis by precisely modulating oxidative stress, immune responses, and skin microbiota. The TA-Ce-Lut metal-polyphenol nanoparticles, constructed through coordination bonds, exhibited a size of 177.30 ± 0.55 nm and high encapsulation efficiency of 72.89 ± 5.32%. Upon incorporation into the hydrogel, this system provided a localized and sustained drug depot. In an imiquimod-induced mouse model of psoriasis (both ear and back), TA-Ce-Lut-Gel significantly reduced skin inflammation, improved skin thickness, erythema, and scaling, showing superior efficacy compared to traditional anti-inflammatory treatments. This system alleviated ROS levels, regulated macrophage M2 polarization, and reshaped the immune response, while also modulating the TRAF6/NF-κB/STAT3 signaling pathway to significantly inhibit pro-inflammatory cytokine secretion. Additionally, combined with NIR irradiation, TA-Ce-Lut-Gel enhanced transdermal drug delivery and antioxidant activity, further enhancing therapeutic effects and promoting skin microbiota recovery. This study presents an effective platform for localized precise treatment of psoriasis, overcoming the limitations of traditional single-target anti-inflammatory therapies and offering a potential strategy for the treatment of other immune-mediated diseases, with broad clinical application prospects.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12951-026-04842-z.

Introduction

Psoriasis is a chronic immune-mediated inflammatory skin disorder [1, 2]. Its clinical persistence is not due to the continuous activation of a single inflammatory factor, but rather the result of multiple pathological processes forming a self-sustaining inflammatory loop [3]. Abnormal proliferation and differentiation of keratinocytes, amplification of the immune-inflammatory axis centered on IL-17 A, oxidative stress, mitochondrial dysfunction, and skin barrier disruption interact with dysbiosis to lock the local inflammatory state, leading to the recurrence and difficulty in curing psoriasis [4, 5].

In this pathological loop, keratinocytes are not merely passive targets of inflammation. They actively participate in and exacerbate the immune response through abnormal proliferation, migration, and secretion of inflammatory mediators [6, 7]. Meanwhile, IL-17 A-driven immune signals persistently activate macrophages and other innate immune cells, maintaining their pro-inflammatory phenotype and further exacerbating the local inflammatory microenvironment [8]. Notably, this process is often accompanied by significant oxidative stress and mitochondrial dysfunction, which in turn sustain the inflammatory signaling, forming a feedback loop that is difficult to break.

Although treatment strategies for psoriasis, such as topical steroids, immune-modulating small molecules, and targeted biologics, have evolved, they primarily focus on suppressing downstream inflammatory signals or immune output [9], without fundamentally addressing the self-sustaining mechanism of inflammation. Recently, local treatment strategies based on biomaterials have shown promise in reducing systemic side effects [10–12]. However, most studies still focus on the delivery of anti-inflammatory or antioxidant molecules, lacking the ability to regulate the multi-layered pathological state of psoriasis comprehensively. Therefore, breaking this self-sustaining loop requires a multi-dimensional intervention that simultaneously targets oxidative stress, immune dysregulation, and microbial imbalance—a concept we term “inflammatory-state reprogramming.”

Thus, the effective intervention of psoriasis requires a shift from “inflammation suppression” to “inflammatory-state reprogramming.” [13] This approach should focus on multi-dimensional regulation to restore the skin’s local microenvironment to a stable state. An ideal treatment strategy should buffer excessive oxidative stress, restore metabolic homeostasis, and simultaneously regulate immune cell functions, rather than simply targeting a single pathway for short-term intervention. In this context, materials with multi-functional synergistic effects and the ability to precisely control local delivery offer new possibilities for reprogramming the inflammatory state. Tannic acid (TA) is known for its excellent biocompatibility and multi-site coordination capabilities, serving as a stable structural basis for functional materials [14, 15]. Additionally, TA exhibits anti-inflammatory [16], antioxidant [17], and immune-regulatory [18] effects. Cerium-based materials, with their reversible Ce³⁺/Ce⁴⁺ redox cycling, can buffer reactive oxygen species (ROS) levels and restore redox homeostasis in various cell types [19]. Luteolin (Lut), a natural multi-target immune-regulatory molecule, has been shown to modulate multiple inflammation-related signaling pathways [20, 21]. By rationally integrating these components, a composite system that functions across structural, metabolic, and immune levels could be developed to achieve systemic reprogramming of the inflammatory state in psoriasis.

Moreover, as psoriasis lesions are primarily located on the skin, local treatment offers significant clinical advantages [22]. An ideal local treatment platform should possess good skin adhesion, controllable drug release, and mechanical stability to adapt to complex skin environments. Hydrogels, characterized by their high water content, flexibility, and tunable structure, are considered ideal carriers for topical treatments [23, 24]. By incorporating functional nanoplatforms into a hydrogel network that can gel in situ, local retention time can be extended, and physical responses to external stimuli can provide a foundation for localized and sustained treatment [25].

Based on these insights, we developed a light-responsive composite hydrogel system (TA-Ce-Lut-Gel) by integrating TA-Ce-Lut metal-polyphenol nanoparticles into a GelMA/HAMA hydrogel matrix. This design enables localized and sustained drug retention, while allowing NIR-enhanced transdermal delivery. The system was systematically evaluated in vitro and in two IMQ-induced psoriasis mouse models to assess its capacity to reshape the inflammatory microenvironment, ameliorate pathological phenotypes, and restore skin homeostasis. Mechanistic studies further examined its association with the TRAF6/NF-κB/STAT3 signaling pathway and associated changes in skin microbiota composition. Together, this work presents a multi-dimensional therapeutic platform that integrates antioxidant, immunomodulatory, and photothermal functions, offering a potential therapeutic strategy for psoriasis and potentially other chronic inflammatory diseases.

Materials and methods

Materials

TA (T818845), cerium nitrate (Ce(NO₃)₃, C804687), and Lut (C17463887) were purchased from Shanghai Macklin Biochemical Co., Ltd. (Shanghai, China). Lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP, EFL-LAP-250), gelatin methacryloyl (GelMA, EFL-GM-30), and hyaluronic acid methacryloyl (HAMA, EFL-HAMA-150 K) were obtained from EngineeringForLife (Shenzhen, China). Collagenase II (PH1537), Calcein-AM/PI kit (C2015M), EdU kit (C0078S), JC-1 kit (C2006), DCFH-DA kit (S0033S), Calcipotriol ointment (Daivonex, HJ20160070), CCK-8 kit (C0038), and Western blot protease inhibitor (P1046) were obtained from Beyotime Biotechnology (Shanghai, China). ELISA kits for IL-6 (JLCJ221), IL-1β (JLCY8691), TNF-α (JLC17113), IL-10 (JLC17305), and TGF-β (JLCJ236) were purchased from Jiangxi Jianglanchun Biotechnology Co., Ltd. (Jiangxi, China). The Pierce™ ECL Western (32209) was purchased from Thermo Fisher Scientific (Waltham, MA, USA). Imiquimod (IMQ) cream (5%) (H20030129) was sourced from Sichuan Mingxin Pharmaceutical Co., Ltd. (Chengdu, China). The primary antibodies used included rabbit anti-F4/80 (GB113373), rabbit anti-CD68 (GB113109), rabbit anti-IL-17 (GB11110-1), rabbit anti-TNF-α (GB115701), rabbit anti-IL-1β (GB11113), the TMR (Red) TUNEL Apoptosis Assay Kit (G1502), HRP-conjugated secondary antibodies, and IF488-Tyramide (TSA) reagents were purchased from Servicebio (Wuhan, China).

Preparation and characterization of nanoparticles

Preparation of TA-Ce nanoparticles

Metal-polyphenol nanoparticles were constructed by coordination bonds [26]. Briefly, 73.2 mg TA was dissolved in 5 mL pH 7.4 PBS. 62.5 mg Ce(NO₃)₃ was dissolved in 8 mL pH 7.4 PBS. After ultrasonic mixing, 500 µL of TA solution was added to 40 mL of deionized water, followed by magnetic stirring at 900 rpm for 5 min. Next, 800 µL of cerium solution was added, and the mixture was stirred at 900 rpm for 1.5 h. After completion, the solution was filtered through a 0.22 μm membrane and divided into centrifuge tubes. The samples were centrifuged at 14,000 rpm for 15 min, and the supernatant was discarded. The residual 0.1 mL was re-suspended in 0.9 mL of deionized water, followed by further centrifugation. This process was repeated three times. The final samples were combined, lyophilized at −40 °C for 3 days, and stored at low temperature for future use.

Preparation of TA-Ce-Lut nanoparticles

73.2 mg TA was dissolved in 5 mL pH 7.4 PBS. 62.5 mg Ce(NO₃)₃ was dissolved in 8 mL pH 7.4 PBS. 25.75 mg Lut was also dissolved in 8 mL of pH 7.4 PBS. After ultrasonic mixing, 500 µL of TA solution was added to 40 mL of deionized water and stirred at 900 rpm for 5 min. Then, 800 µL of cerium solution was added, followed by stirring for 5 min. Afterward, 800 µL of Lut solution was added and stirred at 900 rpm for 1.5 h. The mixture was filtered through a 0.22 μm membrane, aliquoted into centrifuge tubes, and subjected to the same centrifugation process as described for TA-Ce nanoparticles. The samples were lyophilized at −40 °C for 3 days and stored at low temperature for later use.

Characterization of nanoparticles

The morphology of the nanoparticles was examined using scanning electron microscopy (SEM, Apreo, Thermo Fisher Scientific, USA). The crystalline structure was analyzed by X-ray diffraction (XRD, D8 ADVANCE, Bruker, Germany) in the 2θ range of 5°−60° at a scan rate of 5°/min. Functional groups were identified by Fourier-transform infrared (FT-IR) spectroscopy (Nicolet IS10, Thermo Fisher Scientific, USA) in the range of 4000–400 cm^− 1. Thermal stability was assessed by thermogravimetric analysis (TGA, STA200, Hitachi, Japan) in a nitrogen atmosphere from 25 to 800 °C at a heating rate of 10 °C/min.

Particle size and stability

The hydrodynamic diameter and polydispersity index (PDI) of the nanoparticles were measured using a Malvern Zetasizer Nano ZS (Malvern Instruments, UK). The samples were diluted in deionized water and measured three times at room temperature. The colloidal stability was assessed by tracking the nanoparticle size over time. Dispersions were stored in the dark at room temperature for 7 days, and the size and PDI were measured at regular intervals.

Determination of drug loading and encapsulation efficiency

The Lut content was quantified by high-performance liquid chromatography (HPLC). The chromatographic conditions were as follows: Agilent ZORBAX SB-C18 column (4.6 mm × 150 mm, 5 μm); mobile phase: methanol: 0.2% phosphoric acid aqueous solution (60:40, v/v); flow rate: 1.0 mL/min; column temperature: 30 °C; detection wavelength: 350 nm; injection volume: 10 µL. The retention time of Lut was approximately 3.6 min.

Drug loading (DL) was determined using the direct method: a known mass of lyophilized nanoparticle powder was dissolved in methanol and sonicated to release the drug completely. The solution was then filtered through a 0.22 μm membrane filter, and the Lut concentration was measured by HPLC.

Encapsulation efficiency (EE) was determined using the indirect method: the freshly prepared nanoparticle suspension was centrifuged at 20,000 × g for 45 min to separate free Lut. The supernatant was collected, and the concentration of free Lut was determined by HPLC.

DL and EE were calculated as follows:

1here W_total is the total amount of drug added, W_free is the amount of free drug measured in the supernatant, and W_{nanoparticles} is the total weight of the lyophilized nanoparticle powder.

In vitro hemolysis test

Whole blood was collected from experimental mice, and red blood cells (RBCs) were isolated by centrifugation and diluted with pH 7.4 PBS to prepare a 2% suspension. Various concentrations (25–100 µg/mL) of nanoparticles were incubated with the RBCs suspension at 37 °C for 2 h. Deionized water and pH 7.4 PBS were used as positive and negative controls, respectively. After centrifugation, the supernatant was collected, and the absorbance (OD) was measured at 540 nm.

In vitro cell studies

Cell culture and model establishment

The mouse macrophage cell line RAW264.7 and the human immortalized keratinocyte cell line (HaCaT) were obtained from the Cell Resource Center of the Institute of Basic Medical Sciences, Chinese Academy of Medical Sciences. HaCaT cells were cultured in DMEM medium supplemented with 10% fetal bovine serum (FBS). An in vitro psoriasis model was established by stimulating HaCaT cells with an M5 cytokine cocktail (TNF-α, IL-17 A, IL-22, IL-1α, OSM, 10 ng/mL each) for 24 h. Subsequent drug treatment was performed, and the light-exposed group received 808 nm laser irradiation (1.0 W/cm², 5 min) after drug application to enhance local transdermal delivery.

Cell viability assay

Cell viability was assessed using the CCK-8 assay. HaCaT cells were seeded in 96-well plates at a density of 5 × 10³ cells/well. After cell attachment, cells were treated with M5 cytokine cocktail and varying concentrations of the samples for 24 h. Fresh medium containing 10% CCK-8 reagent was added to each well, and after 2 h of incubation at 37 °C, absorbance was measured at 450 nm.

Live/dead staining

Cell viability was determined using the Calcein-AM/PI kit. After treatment with the respective drugs, HaCaT cells were washed with pH 7.4 PBS and incubated with a working solution of 2 µM Calcein-AM and 8 µM PI at 37 °C for 30 min. Cells were imaged using fluorescence microscopy, and the live/dead cell ratio was calculated.

EdU cell proliferation assay

The proliferation of HaCaT cells was measured using the EdU kit. After treatment as described, cells were incubated with 10 µM EdU for 2 h. Cells were then fixed with 4% paraformaldehyde, permeabilized with 0.3% Triton X-100, and stained using a Click reaction. Nuclei were counterstained with DAPI. The percentage of EdU-positive cells was manually quantified using fluorescence microscopy.

Cell scratch assay

HaCaT cells were seeded in 6-well plates at a density of 2 × 10⁵ cells/well and allowed to reach confluence. The cells were then treated with the M5 cytokine cocktail and the respective drugs (including light exposure). Scratches were made using a 200 µL pipette tip. Images of the scratch area were captured at 0, 12, 24, and 36 h. The wound healing area was analyzed using ImageJ software to assess the inhibition of abnormal migration.

Transwell migration assay

Transwell chambers (8.0 μm pore size) were used to assess cell migration ability. HaCaT cells (5 × 10⁴ cells/well) suspended in serum-free medium containing drugs were added to the upper chamber, while the lower chamber contained complete medium with 10% FBS. After 24 h of incubation at 37 °C, non-migrated cells were removed, and cells that had migrated to the lower surface were fixed, stained with crystal violet, and counted.

Mitochondrial membrane potential and ROS detection

Mitochondrial membrane potential and ROS levels were measured using JC-1 and DCFH-DA probes, respectively. After 24 h of drug treatment, HaCaT or RAW264.7 cells were incubated with either JC-1 or DCFH-DA probes at 37 °C in the dark for 30 min. After washing, fluorescence images were obtained using confocal laser scanning microscope (CLSM, Biotek, Winooski, VT, USA).

Intracellular ATP content assay

HaCaT or RAW264.7 macrophages were seeded into 6-well plates at a density of 1 × 10⁵ cells per well. After the corresponding drug treatments, 200 µL of ATP detection lysis buffer was added to each well to fully lyse the cells, and the supernatants were collected. The enhanced ATP assay kit (Beyotime, S0027) was used for detection: 20 µL of supernatant was mixed with 100 µL of pre-prepared ATP detection working solution, and the chemiluminescence value (RLU, relative light unit) was measured. Meanwhile, the protein concentration of each group was determined by the BCA method, and the ATP content was normalized to nmol per milligram of protein (nmol/mg protein).

Macrophage polarization and anti-inflammatory effect tests

RAW264.7 macrophages were polarized towards the M1 phenotype using LPS (1 µg/mL) [27]. Different drug treatment groups were incubated for 24 h, and the NIR group was exposed to 808 nm laser irradiation for the specified duration. The JC-1 and DCFH-DA probes were added, and mitochondrial membrane potential and ROS levels were assessed using CLSM.

RAW264.7 cells were harvested with 0.25% trypsin-EDTA and washed twice with ice-cold PBS. Cells were first stained with Fixable Viability Stain 510 (BD Biosciences) for 15 min at 4 °C in the dark to exclude dead cells. After washing, cells were blocked with anti-CD16/32 Fc block (BD Biosciences) for 10 min at 4 °C to prevent non-specific antibody binding. Cells were then incubated with fluorochrome-conjugated antibodies: PE anti-mouse CD86 (clone GL-1, BD Biosciences) and APC anti-mouse CD206 (clone C068C2, BioLegend) for 30 min at 4 °C in the dark. Isotype-matched control antibodies (PE Rat IgG2a κ, BD Biosciences; APC Rat IgG2a κ, BioLegend) were used to set compensation and define positive gating boundaries. After staining, cells were washed twice and resuspended in 300 µL PBS. Flow cytometry analysis was performed on a BD FACSAria III flow cytometer. The gating strategy was as follows: (1) FSC-A vs. SSC-A to identify the intact macrophage population; (2) FSC-H vs. FSC-A to gate single cells and exclude doublets; (3) Viability stain vs. FSC-A to exclude dead cells; (4) CD86 vs. CD206 to quantify M1 (CD86⁺CD206⁻) and M2 (CD86⁻CD206⁺) macrophage subsets. At least 10,000 single live cells were collected for each sample. Data were analyzed using FlowJo software (version 10.6.2) by two independent researchers blinded to the experimental groups.

RAW264.7 cells were fixed and permeabilized using appropriate fixation and permeabilization buffers in PBS (pH 7.4). Non-specific binding was blocked with bovine serum albumin (BSA). Primary antibodies against CD86 and CD206 were applied overnight at 4 °C. After incubation, fluorescently labeled secondary antibodies and DAPI were applied. CLSM was used to acquire images, and the relative fluorescence intensity of M1 (CD86) and M2 (CD206) markers was quantified to evaluate macrophage polarization.

Inflammatory factor secretion assay

Culture supernatants were collected and centrifuged at 1000 × g for 10 min. The concentrations of pro-inflammatory cytokines (IL-6, IL-1β, and TNF-α) and anti-inflammatory cytokines (IL-10 and TGF-β) were measured using enzyme-linked immunosorbent assay (ELISA) kits.

Preparation and characterization of composite hydrogel

Preparation of GelMA hydrogel

LAP was dissolved in pH 7.4 PBS to prepare a 0.25% (w/v) photoinitiator stock solution, which was stored in the dark. 100 mg GelMA was dissolved in 2 mL of the LAP stock solution at 50 °C for 15 min. The solution was then crosslinked and solidified under UV light at a wavelength of 365 nm and an intensity of 10 mW/cm² to form the GelMA hydrogel.

Preparation of HAMA hydrogel

LAP was dissolved in pH 7.4 PBS to prepare a 0.25% (w/v) photoinitiator stock solution, stored in the dark. 100 mg HAMA was dissolved in 2 mL of the LAP stock solution at 50 °C for 15 min, forming a 5% (w/v) prepolymer solution. Crosslinking and solidification were then performed under UV light (365 nm, 10 mW/cm²) to form the HAMA hydrogel.

Preparation of GelMA/HAMA (GH) composite hydrogel

LAP was dissolved in pH 7.4 PBS to create a 0.25% (w/v) photoinitiator stock solution, stored in the dark. 100 mg GelMA and 100 mg HAMA were dissolved separately in 2 mL of the LAP stock solution at 50 °C for 15 min. The two solutions were then mixed at a 1:1 volume ratio and further stirred at 50 °C for 40 min to ensure uniformity. The resulting GelMA/HAMA (GH) hydrogel was crosslinked and solidified under UV light at 365 nm and 10 mW/cm².

Preparation of TA-Ce-Lut-Gel composite hydrogel

TA-Ce-Lut-Gel was prepared by incorporating TA-Ce-Lut nanoparticles into the GH composite hydrogel matrix. Two solutions were prepared. 10 mg LAP was dissolved in 2 mL of pH 7.4 PBS and 4 mg TA-Ce-Lut was dissolved in 2 mL of pH 7.4 PBS. Each solution was vortexed in the dark until completely dissolved. The two solutions were combined and vortexed again, resulting in a homogeneous 4 mL stock solution containing LAP and TA-Ce-Lut, referred to as the TA-Ce-Lut-containing photoinitiator stock solution. 100 mg GelMA and 100 mg HAMA were weighed and dissolved in 2 mL of the TA-Ce-Lut-containing LAP stock solution, with gentle stirring at 50 °C for 15 min to obtain uniform polymer solutions. The two polymer solutions were mixed thoroughly at a 1:1 volume ratio and stirred for an additional 40 min at 50 °C. The resulting precursor solution, designated TA-Ce-Lut-Gel, was stored in the dark at 4 °C until use. For hydrogel formation, the precursor solution was exposed to UV light (365 nm, 10 mW/cm²) for 10 s to initiate crosslinking.

Morphology and pore size

Hydrogels were lyophilized, sputter-coated with gold, and imaged using SEM at 10 kV. Pore size and porosity were quantified using ImageJ software.

In vitro degradation

Lyophilized hydrogels were weighed (W₀) and placed in pH 7.4 PBS containing 2 U/mL collagenase II at 37 °C. At set intervals, samples were removed from the degradation medium, washed with deionized water, lyophilized, and re-weighed (W_t). The degradation rate was calculated as:

In vitro swelling

Dry hydrogel samples were weighed (W_d) and immersed in pH 7.4 PBS at room temperature. At set intervals, the samples were removed from the pH 7.4 PBS solution, blotted on filter paper to remove surface moisture, and weighed (W_s). The swelling degree was calculated as:

Surface wettability

The static water contact angle was measured at 25 °C using a contact angle meter. A 3 µL drop of deionized water was placed on the hydrogel surface, and the contact angle was immediately recorded using the instrument software. The measurements were conducted at three different locations on each sample, and the average value was obtained.

Water retention

Hydrogel samples were immersed in pH 7.4 PBS at room temperature for 24 h to achieve equilibrium swelling. The samples were then blotted dry and weighed (W₁). The samples were placed in a water bath at 37 °C, and weight measurements were taken at set intervals (W_t). Water retention was calculated as:

In vitro drug release

The in vitro drug release profile from the drug-loaded hydrogels was evaluated using a thermostatic oscillation system. A 5 mg sample of TA-Ce-Lut hydrogel was submerged in 50 mL of pH 7.4 PBS containing 20% (v/v) PEG400 and shaken at 100 rpm in a thermostatic oscillator system (SHZ-82, Jiangsu Jintan Scientific Analytical Instrument Co., Ltd.). At predetermined time points (1, 2, 3, 4, 6, 8, 10, 12, and 24 h), 2 mL of the solution was aspirated and filtered through a 0.22 μm microporous membrane. The drug concentration was quantified using HPLC. After each sampling, pH 7.4 PBS was added to maintain the “leak tank” and the experiment was repeated. Each time point was measured in triplicate.

In vitro transdermal permeation study

A modified Franz diffusion cell method was employed to evaluate the transdermal permeation behavior of different formulations. Male SPF-grade BALB/c mice (6–8 weeks old) were sacrificed by cervical dislocation. The dorsal hair was immediately shaved, and the full-thickness dorsal skin was carefully excised. The skin was rinsed repeatedly with normal saline to remove blood residues, and the subcutaneous fat and connective tissue were meticulously scraped off. The processed skin was cut into appropriate sizes, placed flat on filter paper, and stored at −20 °C for use within one week.

Vertical Franz diffusion cells with an effective diffusion area of 1.77 cm² and a receiving chamber volume of 7.0 mL were used. The mouse skin was mounted between the donor and receiving chambers with the stratum corneum facing the donor chamber and the dermis in contact with the receiving phase. The receiving chamber was filled with 7.0 mL of pre-warmed pH 7.4 PBS buffer containing 20% (v/v) PEG400 to enhance the solubility of Lut. A constant temperature circulation system was activated to maintain the receiving solution at 32± 0.5 °C, and the magnetic stirring speed was set at 300 rpm. The experiment was divided into four groups with an equivalent Lut dose of 11.94 µg/mL: Free Lut group (0.5 mL of pH 7.4 PBS solution containing 11.94 µg/mL Lut was added to the donor chamber), TA-Ce-Lut group (0.5 mL of pH 7.4 PBS suspension containing 200 µg/mL TA-Ce-Lut nanoparticles (equivalent to 11.94 µg/mL Lut) was added to the donor chamber), TA-Ce-Lut-Gel group (0.5 g of TA-Ce-Lut-Gel hydrogel (equivalent to 11.94 µg/mL Lut) was applied to the donor chamber), TA-Ce-Lut-Gel-NIR group (0.5 g of TA-Ce-Lut-Gel hydrogel was applied to the donor chamber, followed immediately by 808 nm near-infrared laser irradiation for 5 min at a power density of 1.0 W/cm²).

At predetermined time points, 0.5 mL of the receiving solution was withdrawn and immediately replaced with an equal volume of pre-warmed blank receiving solution. All samples were filtered through a 0.22 μm microporous membrane, and the concentration of Lut was determined by HPLC using the same chromatographic conditions as described in Sect. 2.2.5. The cumulative permeation rate was calculated accordingly.

After 6 h, the skin was removed and rinsed thoroughly with normal saline to remove residual drug on the surface. The skin was blotted dry with filter paper, and the stratum corneum was removed by tape stripping. The remaining dermal layer was minced and placed in 5 mL of methanol for ultrasonic extraction for 30 min. The extract was centrifuged at 12,000 rpm for 10 min, and the supernatant was filtered through a 0.22 μm membrane. The Lut content in the skin tissue was determined by HPLC.

Rheological testing

Rheological measurements were performed using an Anton Paar MCR92 rheometer (Anton Paar, Austria) with a 35 mm parallel plate rotor at 25 °C. A strain sweep (1 rad/s, 0.1%−100% strain) was first conducted to determine the linear viscoelastic zone. Subsequently, cyclic oscillatory shear testing was performed at a constant frequency (1 rad/s) to assess gel stability under shear. Frequency sweeps (0.1–100 rad/s) were conducted to determine the viscoelastic properties of the hydrogel.

Adhesion, mechanical properties, and injectability

The macroscopic adhesion, torsion, and injection performance of the hydrogel was evaluated. For adhesion, the hydrogel was applied to various substrates (metal, glass, plastic, and wood) and adhered to glove joints. Dynamic adhesion and flexibility were assessed by attaching the hydrogel to a nitrile glove and photographing the interface stability at different finger joint angles during motion.

Animal studies

Animal model establishment

A psoriasis mouse model was established with IMQ [28, 29]. Male BALB/c mice (6–8 weeks old, weighing 20–22 g) were housed in a Specific pathogen-free (SPF) conditions (temperature: 22 ± 2 °C, humidity: 50 ± 10%, and a 12-hour light/dark cycle), with ad libitum access to food and water. After a one-week acclimatization period, the mice were randomly assigned to seven experimental groups (n = 6 per group): (1) Control group, (2) Model group, (3) TA-Ce group, (4) TA-Ce-Lut group, (5) TA-Ce-Lut-Gel group, (6) TA-Ce-Lut-Gel-NIR group (NIR irradiation at 808 nm, 1.0 W/cm² for 5 min), and (7) Calcipotriol group (positive control). Two independent animal cohorts were used to evaluate therapeutic efficacy in ear and back psoriasis-like inflammation models, respectively.

Ear psoriasis-like inflammation model: Except for the control group, all mice received a daily topical application of 62.5 mg of 5% IMQ cream on the right ear at 8:00 AM for 6 consecutive days. At 5:00 PM each day, the corresponding treatment preparations or calcipotriol ointment were topically applied to the same ear. For hydrogel-treated groups, in situ crosslinking was performed immediately after application using 365 nm UV light (10 mW/cm², 10 s). The TA-Ce-Lut-Gel-NIR group received additional 808 nm NIR irradiation (1.0 W/cm², 5 min) right after UV crosslinking.

Back psoriasis-like inflammation model: The dorsal fur of the mice was shaved 24 h prior to the first IMQ application, and a 2 cm × 2 cm square treatment area was demarcated. Except for the Control group, all mice received a daily topical application of 62.5 mg of 5% IMQ cream on the shaved dorsal skin at 8:00 AM for 6 consecutive days. At 5:00 PM each day, the corresponding treatment preparations or calcipotriol ointment were topically applied to the same dorsal area, with equivalent Lut doses maintained across all drug-treated groups (59.7 µg/mouse/day). The crosslinking and NIR irradiation procedures were identical to those used in the ear model. Residual gel and scales were gently removed with saline-soaked cotton swabs before each daily reapplication.

All animal experiments were approved by the Experimental Animal Ethics Committee of the Institute of Materia Medica, Chinese Academy of Medical Sciences (Approval No.: IMM‑S‑25‑0167).

Scoring and phenotypic observation

Body weight was monitored daily throughout the experiment, and ear thickness was measured using an electronic digital micrometer. The modified Psoriasis Area and Severity Index (PASI) was employed to assess erythema, scaling, and thickening of the dorsal skin. Each parameter was scored on a scale of 0 to 4 (0: none; 1: mild; 2: moderate; 3: severe; 4: very severe), with the total score being the sum of the individual parameter scores.

Tissue sectioning and staining

At the conclusion of the study, psoriatic skin lesions and major organs (heart, liver, spleen, lungs, kidneys) were harvested and fixed in 4% paraformaldehyde. Tissues were subsequently embedded in paraffin, sectioned, and stained with H&E for morphological examination and epidermal thickness measurement. TUNEL staining was performed to assess apoptotic cell death. The expression of pro-inflammatory cytokines (IL-1β, IL-17, and TNF-α) was evaluated through immunohistochemistry. For macrophage phenotype analysis, fluorescently labeled antibodies (FITC-anti-F4/80, PE-anti-CD86, and APC-anti-CD206) were used, and M1/M2 macrophage marker expression was assessed using a confocal microscope.

Western blot analysis

Protein extraction was carried out using RIPA lysis buffer containing protease and phosphatase inhibitors. Protein concentrations were determined using the BCA method. 20 µg Equal amounts of protein were separated by SDS-PAGE and transferred onto PVDF membranes. The membranes were blocked with 5% skim milk for 1 h at room temperature, followed by overnight incubation at 4 °C with primary antibodies (TRAF6, STAT3, p-STAT3, JNK, p-JNK, ERK, p-ERK, p65, p-p65, p38, and p-p38) at a dilution of 1:1000. After washing, the membranes were incubated with HRP-conjugated secondary antibodies (1:3000) for 1 h at room temperature, followed by development using an ECL detection kit (Thermo Fisher, USA).

Network pharmacology analysis

Potential targets of the flavonoid compound Lut were predicted using the SwissTargetPrediction database. Psoriasis-related targets were retrieved from the GeneCards database using “psoriasis” as the search term. The Venny 2.1 online tool was employed to identify the overlapping targets between these two datasets, and a Venn diagram was generated. The overlapping targets were subsequently imported into the STRING database to construct a protein-protein interaction (PPI) network with a confidence score greater than 0.7. The resulting network was visualized and analyzed using Cytoscape software. Topological analysis of the network data was performed using the CytoHubba plugin within Cytoscape to identify key core targets.

Gene Ontology (GO) functional enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of the overlapping targets were carried out using the clusterProfiler package in R. These analyses aimed to investigate the potential molecular mechanisms and key signaling pathways associated with Lut. Core targets identified in the enriched pathways (TRAF6) were selected for further molecular docking simulations using AutoDock Vina software. The binding modes of Lut to these biological targets were visualized and analyzed using PyMOL software.

Skin microbiota analysis

Skin microbiota samples were collected from the surfaces of skin lesions in each experimental group. PCR amplification of the 16 S rRNA gene was performed using primer pairs targeting the V3-V4 region (primer sequences: CCTAYGGGRBGCASCAG, GGACTACNNGGGTATCTAAT). Raw sequencing data were processed using the QIIME2 pipeline for quality control and preprocessing. The processed data were analyzed using alpha-diversity metrics, including the Shannon diversity index, Faith’s phylogenetic diversity (PD), and Simpson diversity index, to assess the diversity of the microbiota communities. Beta-diversity analysis, based on Bray-Curtis distances, was performed through principal coordinates analysis (PCoA) to evaluate the differences in community structure between the groups.

Linear discriminant analysis effect size (LEfSe) analysis was conducted with an LDA threshold of > 3.0 and a significance level of p < 0.05 to identify differentially abundant taxa among the groups. Additionally, predicted metabolic pathways of the skin microbiota were inferred using the Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt2) software, providing insights into the potential functional capabilities of the microbial communities.

Statistical analysis

All data are presented as the mean ± standard deviation (SD). Statistical analysis was performed using GraphPad Prism 9.5 software (GraphPad Software, Inc., La Jolla, CA, USA). The choice of statistical tests was based on the distribution of the data. For comparisons between two groups, Student’s t-test was employed. For comparisons involving three or more groups, one-way analysis of variance (ANOVA) was used, followed by post-hoc analysis with Tukey’s test to evaluate differences between individual groups. Non-parametric tests were applied for data that did not follow a normal distribution. A p value of < 0.05 was considered statistically significant.

Results

Construction and physicochemical characterization of TA-Ce-Lut nanoparticles

To actively regulate the inflammatory microenvironment rather than merely passively deliver anti-inflammatory drugs, a multifunctional nanosystem was developed. The TA-Ce-Lut nanoparticle composite was constructed through a polyphenol-metal coordination strategy (Fig. 1A). TA and Ce³⁺ were coordinated via polyphenolic hydroxyl groups, forming a stable TA-Ce complex. Lut was then added and incorporated into this coordination network. After ultracentrifugation and freeze-drying, the product color shifted from light yellow (TA-Ce) to dark yellow (TA-Ce-Lut), confirming successful Lut loading into the TA-Ce framework. SEM images showed that TA-Ce nanoparticles had an irregular, sheet-like aggregated morphology with a rough surface. After Lut incorporation, the TA-Ce-Lut nanoparticles exhibited a more uniform and well-dispersed structure, appearing as dense, irregular spheres (Fig. 1B). Dynamic light scattering (DLS) analysis revealed that the hydrated particle size of TA-Ce was approximately 146.70 ± 0.61 nm, while the average size of TA-Ce-Lut increased to 177.30 ± 0.55 nm (Fig. 1C). This increase in size correlates with Lut loading, and no abnormal aggregation was observed. Particle size stability tests conducted over 7 days confirmed that both formulations maintained stable sizes and PDI. TA-Ce-Lut remained unchanged during this period, with a PDI of 0.162 ± 0.012, indicating good colloidal stability (Fig. 1D).

Fig. 1.

Preparation, physicochemical characterization, and hemocompatibility of TA-Ce-Lut nanoparticles. (A) Schematic diagram illustrating the self-assembly preparation process of TA-Ce and TA-Ce-Lut nanoparticles. (B) Representative SEM images of TA-Ce and TA-Ce-Lut nanoparticles (scale bars: 2000 nm for low magnification, 500 nm for high magnification). (C) Particle size distributions of TA-Ce and TA-Ce-Lut measured by DLS. (D) Stability assessment of TA-Ce and TA-Ce-Lut nanoparticles over a 7-day period at room temperature (n = 3). (E) XRD patterns of TA, Ce³⁺, Lut, TA-Ce, and TA-Ce-Lut. (F) FT-IR spectra of TA, Ce³⁺, Lut, TA-Ce, and TA-Ce-Lut. (G) TGA and derivative thermogravimetry (DTG) curves of TA-Ce and TA-Ce-Lut. (H) DSC thermograms illustrating the thermal stability and crystallinity. (I) Evaluation of drug loading (DL) and encapsulation efficiency (EE) of Lut within TA-Ce-Lut (n = 5). (J) Quantitative hemolysis rates of TA-Ce and TA-Ce-Lut nanoparticles at the indicated concentrations (25–100 µg/mL) in the hemolysis assay. Deionized water and pH 7.4 PBS served as positive and negative controls, respectively (n = 3)

The possible interactions among TA, Ce³⁺, and Lut within the nanoparticles were further investigated. XRD and FT-IR spectroscopy were used to examine the material structure. For TA-Ce-Lut, characteristic peaks of the individual components were significantly attenuated or absent, including peaks for TA at 2θ = 13.1° and 24.4°, for Ce at 10.4° and 20.6°, and for Lut at 7.6°, 25.8°, and 26.8°. This resulted in an amorphous or low-crystallinity structure (Fig. 1E), suggesting that Lut was integrated into the composite rather than being simply physically mixed. FT-IR analysis revealed red shifts in the hydroxyl (from 3415.92 cm^− 1 to 3396.35 cm^− 1) and carbonyl (from 1715.27 cm^− 1 to 1716.91 cm^− 1) peaks of TA. These shifts, along with the appearance of a Ce-O bond at 552.21 cm^− 1, suggested possible coordination and hydrogen bonding interactions among the components (Fig. 1F). Thermal stability was assessed using TGA and DSC. TGA revealed that TA-Ce-Lut underwent multi-step thermal degradation. A 21% weight loss occurred between 27 and 401 °C, followed by a 39% loss between 401 and 800 °C, indicating an organic-inorganic hybrid structure and providing greater thermal stability compared to the individual components (Fig. 1G). DSC analysis further supported the formation of the TA-Ce-Lut nanocomplex. The characteristic endothermic peak of pure Lut at 339 °C shifted to 250 °C in the TA-Ce-Lut complex, and the distinct thermal peaks of TA and cerium were no longer observed, indicating strong intermolecular interactions and coordination bond formation between the three components (Fig. 1H). The DL capacity of TA-Ce-Lut was evaluated, yielding an EE of 72.89 ± 5.32% and a loading content of 5.97 ± 0.29% (Fig. 1I). These values indicate sufficient Lut loading for subsequent studies, while the encapsulation within the TA-Ce matrix is expected to minimize burst release. To assess the hemocompatibility of TA-Ce and TA-Ce-Lut in preparation for in vivo studies, hemolysis was tested at concentrations up to 100 µg/mL. Both nanoparticles exhibited hemolysis rates below 5% at all tested concentrations (Fig. 1J), which meets the safety standard for biomedical materials and establishes a solid foundation for subsequent in vivo experiments.

In summary, the TA-Ce-Lut nanosystem was successfully developed with a stable structure, well-defined physicochemical properties, and favorable biocompatibility. These characteristics provide a structurally stable and biocompatible platform for further studies of oxidative stress and inflammation-related models.

Regulation of psoriasis-like keratinocyte abnormal behavior by TA-Ce-Lut nanoparticles

Keratinocytes are both targets of inflammatory signals and active participants in amplifying inflammation and tissue remodeling in psoriasis [30, 31]. To assess the effects of TA-Ce-Lut on keratinocyte behavior under pathological conditions, HaCaT keratinocytes were induced into a psoriasis-like inflammatory state using an M5 cytokine cocktail. The cells were then treated with TA-Ce, TA-Ce-Lut, or TA-Ce-Lut combined with NIR irradiation (TA-Ce-Lut-NIR) to evaluate the impact of these treatments on cell status (Fig. 2A).

Fig. 2.

In vitro effects of TA-Ce-Lut-NIR on viability, migration, and survival of M5-induced keratinocytes. (A) Schematic representation of the experimental design, treatment protocols, and group assignments. (B) Cell viability of HaCaT cells assessed by CCK-8 assay following treatment with varying concentrations of TA-Ce, TA-Ce-Lut, and TA-Ce-Lut with NIR irradiation (808 nm, 1.0 W/cm², 5 min) (n = 6). (C) Representative live/dead staining images (Scale bars: 100 μm). (D) Corresponding quantitative analysis of cell viability (n = 3). (E) Representative images of wound closure at 0, 24, and 36 h post-scratching (scale bars: 500 μm). Quantitative analysis of wound closure rates at (F) 24 h and (G) 36 h (n = 3). (H) Representative images of Transwell migration assay (scale bars: 200 μm). (I) Corresponding statistical analysis of migrated cells (n = 5)

First, the effects of the different nanoparticles on M5-induced HaCaT cell viability were assessed using the CCK-8 assay. Based on these results, a concentration of 200 µg/mL was selected for subsequent experiments. The viability of cells in the M5-stimulated group was significantly higher than in the unstimulated control, indicating increased proliferation, a characteristic of keratinocytes in an inflammatory environment. Treatment with TA-Ce nanoparticles alone showed no obvious cytotoxicity at low to moderate concentrations, resulting in a modest reduction in cell viability. In contrast, TA-Ce-Lut treatment produced a more pronounced decrease in viability. When combined with NIR irradiation, TA-Ce-Lut further reduced cell viability in a concentration-dependent manner, suggesting that the observed effect reflected a mitigation of the abnormal proliferative state rather than acute cytotoxicity (Fig. 2B). To determine whether reduced viability reflected cell death or behavioral changes, live/dead staining was performed. The model group displayed dense green fluorescence, indicating high cell activity and abnormal cell viability and survival. Treatment with TA-Ce treatment resulted in a moderate increase in live cells compared to the model group. TA-Ce-Lut treatment led to red fluorescence, corresponding to the clearance of abnormal cells. NIR irradiation with TA-Ce-Lut further increased red fluorescence and substantially reduced the percentage of live cells (Fig. 2C, D). These results demonstrate that TA-Ce-Lut, particularly with NIR assistance, effectively reduced the viability and survival of hyperproliferative keratinocytes under psoriatic conditions. Abnormal keratinocyte migration contributes to skin thickening and scale formation in psoriasis [31]. The effects of various treatments on cell migration were assessed using scratch and Transwell assays. In the model group, HaCaT cells exhibited enhanced migration over 36 h (Fig. 2E-G). Treatment with TA-Ce-Lut modestly inhibited migration, while TA-Ce-Lut-NIR showed the most pronounced inhibitory effect. Transwell assays confirmed these findings, with fewer migrated cells in the TA-Ce-Lut-NIR group compared to the model group (Fig. 2H, I).

These results show that TA-Ce-Lut reduced keratinocyte viability under the tested conditions, both with and without NIR irradiation. This effect may be associated with the regulation of abnormal proliferative activity. TA-Ce-Lut also inhibited keratinocyte migration under inflammatory conditions. Overall, these findings provide in vitro evidence that TA-Ce-Lut may modulate key cellular behaviors relevant to psoriasis-like inflammatory responses. NIR irradiation was applied as an external experimental condition and was not interpreted as a mechanistically defined driver in this study.

Regulation of oxidative stress and mitochondrial function in keratinocytes by TA-Ce-Lut

In psoriatic inflammation, abnormal keratinocyte proliferation and migration are associated with oxidative stress and mitochondrial dysfunction [32, 33]. To investigate the impact of TA-Ce-Lut on intracellular redox balance and mitochondrial function, we used an M5-induced HaCaT cell model. The effects of different treatments on mitochondrial function, proliferation, and oxidative stress were evaluated (Fig. 3A).

Fig. 3.

In vitro assessment of mitochondrial membrane potential, ROS levels, and cell proliferation in M5-induced HaCaT cells. (A) Schematic illustration of the experimental design for assessing mitochondrial membrane potential, ROS, and cell proliferation in M5-stimulated keratinocytes. (B) Representative fluorescence images of JC-1 staining, showing changes in mitochondrial membrane potential (scale bar: 100 μm). (C) Quantitative analysis of the JC-1 aggregate/monomer ratio across treatment groups. (D) Representative fluorescence images of EdU incorporation assay, showing cell proliferation (scale bar: 100 μm). (E) Statistical quantification of EdU-positive cells (n = 5). (F) Representative fluorescence images of ROS generation detected by DCFH-DA probe (scale bar = 100 μm). (G) Quantitative analysis of relative ROS levels (n = 3)

Mitochondrial function was assessed using JC-1 staining. In healthy mitochondria, JC-1 forms aggregates that emit red fluorescence, while in damaged mitochondria, JC-1 remains in monomeric form, emitting green fluorescence. Control cells predominantly exhibited red fluorescence (Fig. 3B). In contrast, M5-induced model cells showed significantly reduced red fluorescence and increased green fluorescence, indicating a decrease in mitochondrial membrane potential. Treatment with TA-Ce-Lut increased red fluorescence compared to the model group, and combining TA-Ce-Lut with NIR irradiation further improved this effect. Quantitative analysis of the JC-1 fluorescence intensity ratio (aggregates/monomers) confirmed a gradual increase across treatment groups, with the TA-Ce-Lut-NIR group showing the highest aggregate/monomer ratio compared to the model group (Fig. 3C). To further validate mitochondrial energy metabolism function, we performed intracellular ATP content detection using a luciferase-based assay. As shown in Figure S1, M5 stimulation resulted in decreased ATP levels in HaCaT cells compared to controls. Notably, the TA-Ce-Lut-NIR group showed the most significant recovery. These results were highly consistent with the JC-1 staining data, suggesting a correlation between mitochondrial membrane potential and intracellular ATP levels. Cell proliferation was assessed using the EdU incorporation assay. The proportion of EdU-positive cells was quantified for each treatment group. Model group cells exhibited a higher percentage of EdU-positive cells, indicating that inflammatory conditions induced hyperproliferation of keratinocytes. Treatment with TA-Ce-Lut reduced proliferative activity, while combining TA-Ce-Lut with NIR irradiation further reduced the proportion of EdU-positive cells approaching levels observed in the control group (Fig. 3D, E). Intracellular ROS levels were measured using the DCFH-DA probe. The model group showed a marked increase in ROS fluorescence (Fig. 3F, G). Treatment with TA-Ce reduced ROS levels, with further reduction observed upon treatment with TA-Ce-Lut. The lowest ROS levels were detected in the TA-Ce-Lut-NIR group. This progressive reduction in ROS was accompanied by improved mitochondrial membrane potential and reduced abnormal proliferation.

These findings suggest that TA-Ce-Lut attenuates abnormal keratinocyte proliferation associated with psoriatic inflammation, which is accompanied by alleviation of oxidative stress and improved mitochondrial membrane potential. The combination with NIR irradiation was associated with further improvement of these effects, consistent with previous reports [34]. These results provide in vitro support for the immunomodulatory and therapeutic effects of TA-Ce-Lut-NIR in vivo.

TA-Ce-Lut modulates macrophage polarization state

Chronic inflammation in psoriasis involves macrophage phenotype switching from a repair-oriented to a pro-inflammatory state, exacerbating inflammation. Elevated oxidative stress in keratinocytes further amplifies this immune-inflammatory response. The effect of TA-Ce-Lut on macrophage functional state was investigated using an in vitro system (Fig. 4A).

Fig. 4.

TA-Ce-Lut-NIR treatment is associated with reduced oxidative stress and changes in M1/M2 macrophage markers. (A) Schematic representation of macrophage polarization regulation under LPS stimulation. (B) Representative fluorescence images of JC-1 staining showing mitochondrial membrane potential, with green indicating monomeric form (impaired mitochondria) and red indicating aggregate form (healthy mitochondria) (Scale bar: 100 μm). (C) Representative fluorescence images of DCFH-DA staining for intracellular ROS levels (green fluorescence) with nuclei counterstained by DAPI (blue) (Scale bar: 100 μm). (D) Representative immunofluorescence images showing macrophage surface markers, with green indicating CD86 (M1 marker), red indicating CD206 (M2 marker), and blue indicating DAPI (nuclei) (Scale bar: 100 μm). (E) Flow cytometry plots showing the expression of CD86 (M1) and CD206 (M2) in macrophages under different treatment conditions. (F) Quantitative analysis of JC-1 fluorescence intensity, indicating mitochondrial membrane potential (n = 3). (G) Quantitative analysis of ROS levels based on integrated density measurements (n = 3). (H) Quantitative analysis of the relative area of CD86-positive (M1) and CD206-positive (M2) cells (n = 3). (I) ELISA quantification of pro-inflammatory cytokines (IL-6, IL-1β, TNF-α) and anti-inflammatory factors (IL-10, TGF-β) in the culture supernatants (n = 6)

Mitochondrial function in macrophages was assessed using JC-1 staining. Control macrophages predominantly exhibited red JC-1 aggregate fluorescence, indicating normal mitochondrial function. In contrast, LPS-stimulated macrophages showed significantly reduced red fluorescence and increased green fluorescence, demonstrating a decrease in mitochondrial membrane potential. Treatment with TA-Ce-Lut led to modest increases in the JC-1 aggregate/monomer ratio, while the TA-Ce-Lut-NIR group showed the highest red fluorescence intensity among the treatment groups (Fig. 4B, F). Complementary to the membrane potential assay, intracellular ATP content was quantified as a direct functional readout of mitochondrial oxidative phosphorylation. LPS stimulation induced a marked depletion of cellular ATP stores, consistent with severe mitochondrial damage (Figure S2). A progressive increase in ATP levels was observed across the treatment groups, with TA-Ce-Lut-NIR treatment achieving the highest ATP levels among the treatment groups. These findings were consistent with the JC-1 results, suggesting an association between improved mitochondrial membrane potential and increased ATP levels in treated macrophages. Intracellular ROS levels were assessed using the DCFH-DA probe. Model macrophages displayed a marked increase in ROS fluorescence intensity, confirming that LPS stimulation elevated oxidative stress. Treatment with TA-Ce-Lut reduced ROS accumulation. The TA-Ce-Lut-NIR group showed the lowest ROS fluorescence intensity, indicating reduced oxidative stress in inflammatory conditions (Fig. 4C, G). Macrophage polarization was analyzed using immunofluorescence and flow cytometry. Model macrophages showed significantly increased CD86 signal (pro-inflammatory M1 marker) with unchanged CD206 signal (anti-inflammatory M2 marker). Treatment with TA-Ce-Lut decreased CD86 intensity and increased CD206 intensity. Among all treatment groups, TA-Ce-Lut-NIR exhibited the lowest CD86 and highest CD206 signal intensities, suggesting a shift in macrophage marker expression toward an anti-inflammatory profile (Fig. 4D, H). Flow cytometry confirmed that TA-Ce-Lut enhanced the proportion of CD206-positive macrophages while reducing CD86-positive cells (Fig. 4E). Inflammatory cytokine levels were measured in the conditioned media. The model group exhibited significantly elevated levels of pro-inflammatory cytokines (IL-6, IL-1β, and TNF-α) without significant changes in anti-inflammatory cytokines (IL-10 and TGF-β) (Fig. 4I). TA-Ce-Lut treatment partially reduced pro-inflammatory cytokines, and the TA-Ce-Lut-NIR group demonstrated the most pronounced effects across all measured indicators, indicating a reduction in the macrophage pro-inflammatory profile.

TA-Ce-Lut treatment was associated with changes in macrophage M1/M2 polarization markers. This treatment promoted a shift from a pro-inflammatory to an anti-inflammatory profile, with NIR irradiation was used to assist the immunomodulatory effect of TA-Ce-Lut. These findings provide in vitro evidence for the therapeutic potential of this nanosystem in the psoriatic inflammatory environment.

Construction and physicochemical properties of TA-Ce-Lut-Gel hydrogel

To achieve stable local delivery of the TA-Ce-Lut nanosystem in vivo, it was incorporated into a composite hydrogel platform. GelMA and HAMA were selected as the base materials due to their excellent biocompatibility and photocrosslinking ability. However, GelMA alone often suffers from relatively limited mechanical strength and rapid degradation, while HAMA forms a dense network that may limit drug diffusion and cellular infiltration [35]. To overcome these individual limitations and synergistically combine their advantages, we fabricated a composite GH hydrogel by blending GelMA and HAMA. This formulation was designed to be injectable and undergo in situ gelation upon UV exposure, thereby ensuring local retention and sustained drug release within the psoriatic lesion.

The microstructures of different hydrogel systems were characterized. GelMA hydrogel exhibited relatively large pores (61.23 ± 14.27 μm), while HAMA hydrogel showed a denser network structure (23.07 ± 7.61 μm). The composite GH hydrogel displayed an interconnected porous structure with an intermediate average pore size (36.87 ± 6.41 μm) (Fig. 5B). This pore size allowed for stable structural support while enabling uniform nanoparticle distribution and diffusion [36]. The wettability and degradation properties of the hydrogels were assessed. Hydrophilicity was evaluated by measuring contact angles. GelMA exhibited the lowest hydrophilicity, followed by GH, while HAMA showed the highest hydrophilicity (contact angles of 40.8° ± 7.34°, 19.33° ± 3.31°, and 13.17° ± 1.41°, respectively) (Fig. 5C). Mass residue and water retention measurements revealed that the GH hydrogel exhibited a moderate degradation rate. On day 1, residual weight was 15.21 ± 1.40%, and stabilized at 12.64 ± 1.23%. GH hydrogel also maintained long-term water retention, providing a moist environment for skin lesions (Fig. 5D, E). The swelling properties of the GH hydrogel remained controlled, with a swelling degree ranging from 22.76 ± 1.37% to 24.04 ± 1.51% over 24 to 72 h, thus preventing excessive swelling that could lead to secondary mechanical stimulation (Fig. 5F). The release profile of Lut from the GH hydrogel was characterized in vitro. The TA-Ce-Lut-Gel system exhibited sustained release, with 32.67% ± 3.51% of the payload released within the first 4 h and a cumulative release of 53.67% ± 2.31% after 24 h. No burst release was observed, indicating stable local drug retention (Fig. 5G). The sustained release profile is consistent with the retention of nanoparticles within the crosslinked GH hydrogel matrix, as suggested by the interconnected porous structure observed by SEM. The coordination bonds within the TA-Ce-Lut nanoparticles may also contribute to the gradual release. These structural features support the potential of TA-Ce-Lut-Gel for localized and sustained drug retention. Through this mechanism, TA-Ce-Lut-Gel enables precise localized drug release, potentially enhancing therapeutic effects while minimizing systemic side effects.

Fig. 5.

Hydrogel morphology, rheology, and mechanical properties. (A) SEM micrographs of GelMA, HAMA, and GH hydrogels (scale bars: 400 μm (main images), 40 μm (insets)). (B) Average pore size of the hydrogels (n = 10). (C) Water contact angle measurement for GelMA, HAMA, and GH hydrogels (n = 3). (D) In vitro degradation of GelMA, HAMA, and GH hydrogels (n = 3). (E) In vitro moisture retention of GelMA, HAMA, and GH hydrogels (n = 3). (F) In vitro swelling ratio of GelMA, HAMA, and GH hydrogels (n = 3). (G) In vitro drug release profile of TA-Ce-Lut-Gel hydrogel (n = 3). Rheological characterization: (H) cyclic shear recovery test, (I) frequency sweeps, and (J) strain sweeps of GelMA, HAMA, and GH hydrogels (G′, G′′ represent storage modulus and loss modulus, respectively). (K) Practical evaluation of GH hydrogel: flexibility under compression and torsion with rapid shape recovery. (L) Adhesion properties of the GH hydrogel. Adhesion of the hydrogel to diverse inorganic and organic substrates, and its stable adhesion under stretching, bending, and twisting

The transdermal delivery performance of different Lut formulations was evaluated using Franz diffusion cells. Free Lut exhibited the lowest 6 h cumulative permeation (4.82 ± 0.76%), which increased 2.6-fold after TA-Ce nanoencapsulation (12.34 ± 1.81%). Incorporation into GH hydrogel slightly reduced permeation (8.76 ± 1.30%), likely due to sustained release, under NIR irradiation conditions, the measured permeation was 21.59 ± 3.33%, 4.5-fold higher than free Lut. Skin retention after 6 h showed a similar hierarchical trend: free Lut (2.12 ± 0.37 µg/cm²) < TA-Ce-Lut (3.78 ± 0.59 µg/cm²) < TA-Ce-Lut-Gel (5.09 ± 0.78 µg/cm²) < TA-Ce-Lut-Gel-NIR (7.89 ± 1.29 µg/cm²). These findings suggest that the combination of nanoparticle encapsulation, hydrogel adhesion, and NIR-enhanced permeation contributed to improved local drug delivery.

Rheological properties were examined to assess the mechanical stability and processability for in vivo application. Time sweep and frequency sweep tests revealed that the storage modulus (G′ ~ 10² Pa) of GH hydrogel consistently exceeded the loss modulus (G′′ ~ 10¹ Pa) across the tested frequency range, demonstrating an elastic-dominant behavior (Fig. 5H, I). GH hydrogel remained within the linear viscoelastic region at low strain (< 0.1%). The G′ was approximately 200 Pa, significantly higher than the G′′, indicating a stable network structure. With increasing strain, both moduli decreased due to strain softening. A crossover of G′ and G’’ occurred at strain amplitudes of 1.5%−2%, suggesting sensitive mechanical properties and favorable deformability under shear conditions (Fig. 5J). To validate its practicality, macroscopic operational tests were performed. The GH hydrogel withstood compression and torsion, demonstrating flexibility and favorable processability (Fig. 5K). The hydrogel also exhibited stable adhesion to diverse inorganic and organic substrates under stretching, bending, and twisting without interfacial detachment (Fig. 5L).

In conclusion, TA-Ce-Lut-Gel meets the basic requirements for an ideal hydrogel for topical skin application, including appropriate microstructure, physicochemical properties, and mechanical behavior. This provides a stable platform for subsequent in vivo therapeutic efficacy studies in psoriasis animal models.

In vivo evaluation of TA-Ce-Lut-Gel-NIR for treatment of psoriatic-like skin lesions

The in vivo therapeutic effects of TA-Ce-Lut-Gel were assessed using an IMQ-induced psoriasis model in mouse ears and backs. Therapeutic efficacy, biosafety, and immunomodulatory effects were evaluated under NIR irradiation conditions. (Figures 6A and 7A). In the ear model, IMQ induced significant erythema and thickening in the model group, with a maximum thickness of 445.6 ± 50.8 μm on Day 7. In contrast, the TA-Ce-Lut-Gel-NIR treatment group showed alleviated inflammatory responses. maintaining ear thickness below 240.5 ± 22.8 μm on Day 7 (Fig. 6B, C). H&E staining showed improved stratum corneum and granular layer morphology in the TA-Ce-Lut-Gel-NIR group, with reduced epidermal hyperplasia and hyperkeratosis compared with the model group (Fig. 6D). TUNEL staining showed increased apoptotic cells in the TA-Ce-Lut-Gel-NIR group compared with the model group. Consistent with the anti-proliferative effect observed in vitro in the psoriasis model, while the model group exhibited minimal apoptosis (Fig. 6F, G).

Fig. 6.

TA-Ce-Lut-Gel-NIR alleviates IMQ-induced psoriasis-like skin inflammation without systemic toxicity. (A) Schematic illustration of the experimental design: daily topical application of IMQ cream (62.5 mg/day) from Day 0 to Day 7, followed by tissue collection on Day 8. (B) Daily record of ear thickness across groups (n = 6). (C) Representative photographs of mouse ears on Day 8 showing erythema, scaling, and thickening. (D) Representative H&E stained skin sections (scale bars: 100 and 40 μm). (E) Quantitative analysis of epidermal thickness (n = 3). (F) Representative immunofluorescence images (blue: DAPI; red: TUNEL positive cells; scale bars: 100 and 40 μm). (G) Statistical quantification of the apoptotic index (n = 3). Systemic biosafety evaluation: (H and I) gross morphology and (J and K) H&E-stained sections of spleen and liver (scale bars: 50 and 20 μm), along with (L) representative images of major organs (heart, liver, spleen, lungs, kidneys). (M) Schematic showing the multi-organ harvesting process for safety profiling

Fig. 7.

Establishment of the IMQ-induced psoriasis-like skin inflammation model and therapeutic/safety assessment. (A) Schematic diagram of the experimental procedure. (B) Representative photographs showing the dorsal skin of mice from the Control, Model, and treatment groups (TA-Ce, TA-Ce-Lut, TA-Ce-Lut-Gel, TA-Ce-Lut-Gel-NIR, and Calcipotriol) on Days 1, 3, 5, and 7. (C) High-magnification images of the dorsal skin on Day 8, illustrating the severity of scaling, erythema, and skin thickening among different groups. (D) Representative H&E stained sections of dorsal skin tissues (scale bars: 100 μm and 40 μm). Dotted lines indicate epidermal thickness. (E) Dynamic changes in body weight (g) of mice recorded daily over the 7-day experimental period (n = 6). (F) Daily assessment of skin inflammation using the Psoriasis Area and Severity Index (PASI), including scale score (0–4), thickness score (0–4), erythema score (0–4), and cumulative PASI score (0–12) (n = 6). (G) Serum biochemical parameters measured on Day 8, including AST, ALT, BUN, and CREA to evaluate hepatic and renal functions (n = 6)

Safety evaluation was performed through histological examination and serum biochemistry analysis of AST, ALT, BUN, and CREA. In the model group, the spleen exhibited megakaryocyte hyperplasia consistent with systemic inflammation associated with the IMQ-induced psoriasis model. The TA-Ce-Lut-Gel-NIR group showed intact major organs without signs of toxicity (Figs. 6H–L). Lower BUN and CREA levels were observed in the TA-Ce-Lut-Gel-NIR group compared with the model group. (Fig. 7G).

In the back psoriasis model, TA-Ce-Lut-Gel-NIR improved disease scores in the animal model. In the model group, skin lesions and erythema appeared on day 3, progressing to extensive tissue destruction by day 4. In contrast, the TA-Ce-Lut-Gel-NIR treatment group avoided this level of progression, showing greater skin smoothness compared to the calcipotriol group. Histopathological examination showed similar trends (Figs. 7C, D). The model group exhibited signs of hyperkeratosis and parakeratosis, while the treatment group displayed stratum granulosum thickness and stratum corneum depth comparable to those of the control group. PASI score evaluation showed that treatment improved lesion symptoms, with stratum granulosum morphology approaching that of the control group (Fig. 7F). Immunohistochemistry showed that pro-inflammatory cytokine levels (IL-1β, IL-17, and TNF-α) in the treatment group were markedly lower than those in the model group (Figs. 8B, D).

Fig. 8.

Evaluation of apoptosis, inflammatory cytokines, and macrophage markers in the psoriasis-like skin inflammation model. (A) Representative images of TUNEL staining for detecting apoptosis (scale bar: 100 μm). (B) Immunohistochemistry staining images of IL-1β, IL-17, and TNF-α (scale bar: 100 μm). (C) Macrophage marker staining (CD86, pro-inflammatory marker; CD206, anti-inflammatory marker; scale bar: 100 μm). (D) Quantitative analysis of the IL-1β, IL-17, and TNF-α (n = 3). (E) Quantitative analysis of the CD206 and CD86 (n = 3)

To further investigate the effects of TA-Ce-Lut-Gel-NIR on the immune microenvironment, immunofluorescence double staining of skin tissue was performed showed that treatment significantly reduced the number of pro-inflammatory (CD86-positive, M1-type) macrophages in lesional areas while increased CD206-positive (M2-type) macrophages typically involved in wound healing (Figs. 8C, E). These results indicate that TA-Ce-Lut-Gel-NIR treatment was associated with reduced CD86-positive macrophages and increased CD206-positive macrophages in lesional skin, accompanied by improved mitochondrial membrane potential and reduced oxidative stress. These changes suggest potential modulation of the psoriatic inflammatory microenvironment.

The TA-Ce-Lut-Gel-NIR composite system combines nanoparticle-mediated immunomodulation with hydrogel-based sustained local delivery, with NIR irradiation was used to enhance transdermal permeation. The treatment was associated with reduced epidermal hyperplasia, increased apoptosis of hyperproliferative keratinocytes, and a shift toward increased CD206-positive macrophages in lesional skin. These findings suggest the potential of this platform for local treatment of psoriasis, though further studies are needed to fully characterize the underlying mechanisms and optimize clinical translation.

Potential molecular pathways modulated by TA-Ce-Lut-Gel-NIR

We explored potential molecular pathways associated with TA-Ce-Lut-Gel-NIR in psoriatic skin inflammation. Lut is a component of metal-polyphenol nanoparticles (TA-Ce-Lut). It has been reported to have anti-inflammatory effects. We used it to identify overlapping targets with psoriasis. We used a network pharmacology approach. Venn diagram analysis identified 170 overlapping targets. These targets were used for pathway analysis. (Fig. 9A). Network topology analysis was performed. KEGG pathway analysis was performed. The results suggested involvement of inflammatory pathways, including MAPK, PI3K-Akt, and IL-17/TNF. GO analysis showed enrichment of biological processes, including cytoplasmic kinase activity and protein phosphorylation. (Figures S4, S5). Western blot analysis was performed. The results showed reduced phosphorylation of proteins in MAPK and IL-17/NF-κB pathways, including ERK, JNK, p38, TRAF6, and p65. STAT3 was also downregulated. STAT3 is associated with Th17 cell differentiation. We performed molecular docking simulations. We tested binding between Lut and TRAF6. Docking results suggested a potential binding mode, with a binding energy of −7.085 kcal/mol. The interaction involved hydrogen bonds and hydrophobic interactions. (Fig. 9C). These results suggest a possible binding mode. Experimental validation is still needed.

Fig. 9.

Proposed working model illustrating how TA-Ce-Lut-Gel-NIR may alleviate psoriasis-like skin inflammation through modulation of the TRAF6/NF-κB/STAT3 axis. (A) Venn diagram showing differentially expressed genes in psoriasis and those modulated by Lut treatment. (B) Topological structure analysis of the protein-protein interaction (PPI) network, highlighting key inflammatory targets. (C) Molecular docking simulation results showing predicted binding between Lut and key inflammatory proteins, with calculated binding energies. (D) Schematic diagram of inflammatory signaling pathways examined in this study, including MAPK and NF-κB pathways. (E) Western blot analysis of core pathway protein expression: TRAF6, p65, p-STAT3, ERK, p-JNK, p-p38, and β-actin. (F) Schematic summary of the proposed working model, in which TA-Ce-Lut-Gel-NIR treatment is associated with reduced phosphorylation of TRAF6, p-p65, and p-STAT3, accompanied by decreased pro-inflammatory cytokine levels

Western blot analysis was performed. Protein expression in signaling pathways was analyzed. In the model group, TRAF6, p-p65, and p-STAT3 levels were increased. Lower phosphorylation levels of these proteins were observed in the TA-Ce-Lut-Gel-NIR group compared to the model group (Fig. 9E). These changes were accompanied by reduced levels of downstream pro-inflammatory cytokines. (Fig. 9F).

Collectively, these results suggest a regulatory effect of TA-Ce-Lut-Gel-NIR on inflammatory pathways. The evidence is mainly correlative.

Skin microbiota composition changes After TA-Ce-Lut-Gel-NIR treatment

Psoriasis is frequently associated with significant disruption of the skin microbiota, characterized by reduced diversity and overgrowth of opportunistic pathogens such as Staphylococcus [39]. To investigate whether the therapeutic effects of TA-Ce-Lut-Gel-NIR extend to the restoration of skin microecological homeostasis, we performed 16 S rRNA sequencing on skin samples from all treatment groups. In the model group, there was a near-total loss of skin microbial diversity, with significantly reduced alpha diversity indices, including Chao1 (species richness), Pielou’s evenness (species distribution uniformity), Shannon (species diversity), and Faith’s PD (phylogenetic diversity), compared to healthy controls (Figs. 10A-D). Treatment with TA-Ce-Lut-Gel-NIR was associated with increased microbial diversity relative to the model group. Alpha diversity indices in the treatment group approached those of healthy controls. PCoA analysis revealed a distinct separation between the microbial community structure of treated mice and disease-associated clusters, with a convergence toward the healthy control group (R = 0.485, p = 0.01). Notably, the TA-Ce-Lut-Gel-NIR treatment cluster shifted toward the control group, whereas the model and other treatment groups remained distinct (Figs. 10E, F). OTU analysis showed that the number of OTUs was higher in treated mouse skin (146) compared to the model group (92). (Fig. 10G). Taxonomic analysis revealed differences in bacterial community composition among groups. In the model group, the genus Staphylococcus comprised nearly 98% of the bacterial population, reflecting a severely skewed, single-community composition (Fig. 10H). Following treatment with TA-Ce-Lut-Gel-NIR, this proportion decreased to 24%, providing ecological space for beneficial bacteria to grow. Treatment also promoted the expansion of other beneficial bacterial taxa in the skin of treated mice. Notably, treatment promoted the growth of Corynebacterium and other Actinobacteria, which are known to contribute to skin barrier function and immune homeostasis. LEfSe analysis and phylogenetic tree reconstruction revealed profound shifts in the composition of the skin microbiota across treatment groups, which were accompanied predicted changes in microbial functional profiles based on PICRUSt2 analysis (Figs. 10I, J). The model group exhibited a pathogenic metabolic state, characterized by upregulation of key energy metabolism pathways, including the TCA cycle, glycolysis/gluconeogenesis, and pyruvate metabolism. These metabolic adaptations supported the rapid proliferation of opportunistic pathogens. In the TA-Ce-Lut-Gel-NIR group, PICRUSt2 analysis predicted enrichment of pathways associated with vancomycin group antibiotic biosynthesis (LDA 3.9), ansamycin biosynthesis, and bacterial chemotaxis. This predicted enrichment was consistent with the observed reduction in Staphylococcus abundance Notably, the predicted enrichment of the vancomycin group antibiotic biosynthesis pathway was accompanied by a marked reduction in Staphylococcus abundance. The calcipotriol group showed enrichment in thiamine and pentose metabolism pathways. The control group maintained a baseline metabolic state, characterized by nicotinate and nicotinamide metabolism. These findings indicate that TA-Ce-Lut-Gel-NIR treatment was associated with changes in skin microbiota composition, including reduced Staphylococcus abundance and increased relative abundance of Corynebacterium and other Actinobacteria. Whether these microbial changes represent a direct effect of the treatment or a secondary consequence of reduced inflammation remains to be determined. It should be noted that the current microbiome analysis is primarily descriptive and correlative. The functional pathway results are predictions based on 16 S rRNA gene sequencing and do not represent direct evidence of actual bacterial metabolic activity. Causal relationships between microbial alterations and therapeutic efficacy have not been established through experiments such as microbiota transplantation. These will be the focus of our future research.

Fig. 10.

Analysis of skin microbial community structure and functional enrichment. Alpha diversity indices of skin microbial communities across groups (n = 3), including (A) Chao1, (B) Pielou’s evenness, (C) Shannon, and (D) Faith’s PD indices, indicating microbial diversity. (E) PCoA plot based on Bray-Curtis distance illustrating beta diversity among groups. (F) ANOSIM analysis showing the significance of inter-group differences (R = 0.485, P = 0.01). (G) Venn diagram displaying the number of shared and unique ASVs across groups. (H) Krona chart visualizing the taxonomic composition and relative abundance of microbial communities in each group. (I) LEfSe-derived cladogram identifying significantly enriched taxonomic clades in different groups. (J) Predicted functional pathways inferred by PICRUSt2 using the KEGG database, with bar plots representing LDA scores (log10) for pathways with significant differential abundance (LDA score > 2.0)

Discussion

Psoriasis is a chronic, immune-mediated inflammatory skin disease affecting approximately 2–3% of the global population [40, 41]. Treatment of this condition is challenging due to three primary factors: abnormal keratinocyte proliferation, immune system dysregulation, and disruption of the skin microbial barrier. Conventional therapies often have limited efficacy due to several factors, including the marked thickening of the stratum corneum at psoriatic lesions, which hinders transdermal drug delivery. Additionally, psoriatic lesions are often located at highly mobile sites, leading to drug detachment [42]. Long-term use of glucocorticoids raises safety concerns due to side effects. As a result, these therapies do not address the complex and multifactorial pathology of psoriasis comprehensively.

In this study, we developed a metal-polyphenol nanoparticle-based system, TA-Ce-Lut, which was constructed by self-assembling TA, Ce³⁺, and Lut. The nanoparticles exhibited an irregular spherical morphology, which increased the specific surface area. This structural design was also consistent with the sustained release profile observed in vitro. TA served as the structural backbone, while Lut conferred anti-inflammatory and antioxidant properties to the system. Lut, though a potent anti-inflammatory agent, has poor water solubility and low bioavailability [43], which limits its pharmaceutical potential. Nanotechnology-based strategies have been employed to improve Lut’s solubility, and the integration into a stable metal-phenolic network further improved its cellular delivery. In the IMQ-induced psoriasis model, TA-Ce-Lut-Gel-NIR treatment was associated with reduced phosphorylation of TRAF6, p-p65, and p-STAT3. These findings suggest that nanoparticle-based delivery strategies may enhance the therapeutic application of natural products in treating inflammatory skin diseases.

In vitro experiments suggested complementary roles of the three components: Lut was associated with anti-proliferative and anti-inflammatory effects, TA provided structural coordination, and Ce³⁺ contributed to ROS scavenging. Consistent with previous studies, the antioxidant and anti-inflammatory properties of TA, Ce³⁺, and Lut nanoparticles have been demonstrated in the treatment of cardiovascular diseases such as myocardial infarction [44]. However, to our knowledge, this study represents the first evaluation of such a nanosystem for dermatological applications, particularly for immune-mediated inflammatory diseases like psoriasis. In this study, we expanded the TA-Ce-Lut platform by targeting the key pathological features of psoriasis. A one-pot nanoparticle preparation method was developed, which streamlined the fabrication process and improved batch-to-batch consistency. The incorporation of these nanoparticles into a hydrogel composite system for topical delivery allows for site-specific drug administration, sustained release, and enhanced therapeutic efficacy, while supporting patient safety and improving treatment adherence. This approach provides a foundation for further preclinical development of this nanosystem for dermatological applications.

To overcome the challenge of sustained drug release from nanoparticles applied directly to the skin surface, TA-Ce-Lut was encapsulated within a GH hydrogel. The GH hydrogel was designed to provide mechanical support at psoriatic lesions, with rheological testing showing stable viscoelastic properties across the tested frequency range. The GH hydrogel also provides structural support for thickened psoriatic plaques, especially in highly mobile regions. Unlike conventional applications of GelMA/HAMA hydrogels in static tissue engineering scenarios such as nerve repair [35], our system is specifically tailored to withstand the dynamic mechanical stresses of skin, where compression, torsion, and shear are constant challenges. The high stability of the GH hydrogel matrix protects it from shear stress, preventing network collapse and delamination. The material exhibited strain-softening behavior upon exceeding the linear viscoelastic region, which may contribute to maintaining conformal contact with the skin surface under large deformations during movement. To enhance treatment precision, TA-Ce-Lut nanoparticles were incorporated into a photocrosslinked GH hydrogel, forming a drug depot with adhesive properties and the capacity for sustained local release. The 365 nm crosslinking (10 mW/cm², 10 s) was optimized for clinical use: the 100 mJ/cm² dose is 2–3× lower than standard NB-UVB and well below the erythema threshold. The 10-second procedure uses a portable device and takes < 1 min total. The TA-Ce-Lut-Gel-NIR group showed enhanced therapeutic efficacy compared to the non-irradiated group. The in vitro transdermal permeation study showed that a single NIR irradiation increased Lut permeation by approximately 2.5-fold [45]. In vitro, NIR irradiation was also associated with enhanced ROS clearance. Previous studies have suggested several mechanisms that could account for these effects: NIR irradiation has been reported to reversibly loosen the lipid bilayer [50–52]; NIR irradiation has been shown to promote the Ce³⁺/Ce⁴⁺ redox cycle [37, 38]; local mild hyperthermia has been reported to inhibit pro-inflammatory cytokine secretion and modulate macrophage polarization [53, 54]. It should be noted that direct photothermal characterization and in vivo temperature monitoring were not performed in this study; therefore, the contribution of these proposed mechanisms remains to be experimentally validated.

The TA-Ce-Lut-Gel-NIR system exhibits a multifaceted regulatory profile on psoriatic inflammation at both the cellular and molecular levels. Its potent ROS scavenging capacity and protective effects on mitochondria are associated with reduced activity of the ROS-NF-κB/STAT3 inflammatory pathways, which may contribute to a balanced immune state. A critical feature of psoriasis is the imbalance in macrophage polarization, particularly the shift between pro-inflammatory M1 and anti-inflammatory M2 marker expression [46]. In our study, TA-Ce-Lut-Gel-NIR treatment correlated with a reduced presence of pro-inflammatory M1 macrophages alongside an increased M2 marker expression, changes that potentially support a pro-repair therapeutic effect. The addition of NIR light appeared to further enhance these trends. This observation is potentially associated with the improved activity of TA-Ce-Lut nanoparticles. It also correlates with a potential shift of macrophages toward an anti-inflammatory profile. This proposed working model involves oxidative stress regulation and mitochondrial protection. These pathways correlate with an altered immune profile and alleviated local inflammation. They may also support the promotion of skin repair. Furthermore, changes in keratinocyte behavior were accompanied by modulation of the p-ERK/p-p38 signaling pathway. The system is associated with decreased MAPK pathway activity and increased p-JNK activation. This dynamic may support apoptotic cell death and potentially suppress the abnormal proliferation of keratinocytes. This proposed “ERK/JNK switch” presents a potential model for attenuating excessive keratinocyte proliferation in psoriatic lesions, offering a promising strategy for therapeutic intervention in inflammatory skin diseases.

An important dimension of our therapeutic strategy is the maintenance of a balanced skin microbiota. Psoriasis is characterized by severe dysbiosis, with near-complete dominance of pathogenic Staphylococcus and loss of beneficial commensals [47]. The skin microbiome plays a crucial role in maintaining the skin’s defense barrier [48]. TA-Ce-Lut-Gel-NIR treatment was associated with improved alpha-diversity and phylogenetic diversity. These findings suggest a trend toward a more balanced microbial community. The treatment also correlated with a reduction in pathogenic S. aureus. Furthermore, it favored the growth of beneficial commensals such as Corynebacterium, which helps maintain an acidic skin environment and supports immune system development. Notably, the TA-Ce-Lut-Gel-NIR group showed a significant enrichment of the vancomycin antibiotic biosynthesis pathway. This finding is associated with changes in Actinobacteria abundance. It suggests a potential link to a balanced skin state. Functional predictions indicate an altered metabolic potential in Actinobacteria. This alteration may relate to glycopeptide antibiotic biosynthesis. This pathway could potentially interfere with S. aureus virulence and its quorum-sensing system. These results offer a potential ecological balancing strategy. This approach differs from conventional broad-spectrum antimicrobial treatments, focusing on maintaining a balanced skin microbiota.

This study demonstrates promising therapeutic outcomes. However, we acknowledge several limitations before clinical application. First, human psoriatic plaques vary significantly in thickness, hyperkeratosis, and optical properties [49]. Future studies must establish a quantitative relationship between irradiation parameters and tissue penetration depth. Second, murine models fail to capture the chronic, relapsing nature of human psoriasis. We plan to extend future evaluations to porcine models to validate long-term safety and transdermal delivery efficiency. Third, our mechanistic insights are primarily based on correlative analyses. Future work requires perturbation-based validations, such as genetic knockdowns or specific inhibitor treatments, to establish definitive causal relationships. Fourth, comprehensive in vivo quantitative controls are still needed to decouple the specific molecular contribution of each isolated component. Finally, continuous in vivo photothermal monitoring and dosimetric scaling remain essential next steps. These future efforts will solidify the translational framework of our therapeutic platform.

Conclusion

In summary, we successfully fabricated a programmable TA-Ce-Lut-Gel hydrogel system. This platform integrates functional metal-polyphenol nanoparticles within a supportive GelMA/HAMA matrix. Our findings demonstrate a potential multi-dimensional regulation of psoriatic inflammation. Specifically, the treatment correlates with the modulation of key inflammatory pathways. The cellular analysis also suggests a distinct trend toward M2 macrophage polarization. Concurrently, the intervention is associated with improved skin microbial homeostasis. Through NIR-enhanced transdermal delivery and robust antioxidant activity, TA-Ce-Lut-Gel offers a promising therapeutic strategy for psoriasis. Ultimately, this platform holds significant potential for managing chronic immune-mediated inflammatory diseases.

Author contributions

Z.G. and S.W. contributed to the conceptualization of the study. Z.C., J.Z., S.H. and S.W. developed the study methodology. Z.C. conducted the validation, while Z.C., J.Z. and S.W. performed the investigation. Z.C., S.H. and M.J. carried out the formal analysis. Z.C., S.H., G.L. and S.W. completed the software-related work. Z.C. and G.L. were responsible for the visualization. Z.C. wrote the original draft of the manuscript, and S.W. handled the manuscript review and editing. J.Z., M.J., J.J., Z.N., Z.G. and S.W. provided the research resources. J.J., Z.N., Z.G. and S.W. supervised the study. M.J., J.J., Z.N., Z.G. and S.W. acquired the funding support. All authors reviewed and approved the final version of the manuscript.

Funding

This work was supported by the Fundamental Research Funds for the Central Universities, Peking Union Medical College (3332025063), National Natural Science Foundation of China (82360441, 82073778, 82104106), CAMS Innovation Fund for Medical Sciences (CIFMS) (2021-I2M-1-026, China), Natural Science Foundation of Jilin Province General Program (YDZJ202501ZYTS250).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

All animal experiments were approved by the Experimental Animal Ethics Committee of the Institute of Materia Medica, Chinese Academy of Medical Sciences (Approval No.: IMM‑S‑25‑0167). All procedures followed ethical standards during the experiment.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.