학술

Alginate hydrogel co-loaded with phellodendron bark extract and HucMSC-exosomes for combined antibacterial, anti-inflammatory, and pro-repair acne effects

PLOS ONE
CC BY
이 매체는 공공·자유 라이선스로 본문을 직접 표시합니다.

Figures
Abstract
Acne vulgaris is a common inflammatory skin disorder involving follicular hyperkeratosis, Cutibacterium acnes proliferation, and aberrant inflammation, often leading to scarring. Current antibiotic treatments face resistance and limited repair capacity. Here, we developed an alginate-based hydrogel co-loaded with Phellodendron bark extract (PBE) and human umbilical cord mesenchymal stem cell-derived exosomes (HucMSC-Exo) (PE gel). Antibacterial activity of PBE was evaluated against C. acnes, S. aureus, and E. coli. Drug release and in vitro durability were characterized. A rat ear acne model assessed swelling resolution, bacterial load, histopathology, and inflammatory/repair-related markers. PBE exhibited potent broad-spectrum antibacterial activity (MIC ~ 5 mg/mL against C. acnes), which was retained in the presence of exosomes. Zeta potential showed a bimodal distribution, indicating the coexistence of distinct particle populations. PE gel sustained PBE release ≥24 h and prolonged antibacterial activity >72 h. In vivo, PE gel achieved 74.9% ear swelling resolution on day 11, reduced bacterial load >99.8%, lowered follicular hyperkeratosis to 11.2%, and increased collagen maturity to 73.4%. PBE suppressed TNF-α/IL-6, while exosomes enhanced VEGF/CD31-mediated angiogenesis and collagen remodeling. PE gel may help disrupt the acne pathological cycle through a combined antibacterial, anti-inflammatory, and pro-repair effects, representing a potential non-antibiotic strategy for acne management.
Citation: He Y, Wang Y, Zhi Y, Li M, Wu M, Mai Y, et al. (2026) Alginate hydrogel co-loaded with phellodendron bark extract and HucMSC-exosomes for combined antibacterial, anti-inflammatory, and pro-repair acne effects. PLoS One 21(11): e0363272. https://doi.org/10.1371/journal.pone.0363272
Editor: Ahmed Atwa, Al-Azhar University Faculty of Science for Boys in Cairo, EGYPT
Received: June 3, 2026; Accepted: November 3, 2026; Published: November 20, 2026
Copyright: © 2026 He et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data are in the manuscript and/or supporting information files.
Funding: This research was funded by Guangdong Yiyang Healthcare Charity Foundation (Grant No. JZ2025106; URL:http://www.yy-health.org/), Jiangmen Basic and Applied Basic Research Key Project (Grant No. 2520002000140; URL: http://www.jiangmen.gov.cn/bmpd/jmskjj/), The Science and Technology Plan Project in the Field of Healthcare in Jiangmen (Grant No. 2026YL01008), Guangdong Medical Research Foundation (Grant No. B2026147) and Wu Jieping Medical Foundation (Grant No. 320.6750.2025-18-97, 320.6750.2025-18-108, 320.6750.2025-18-105). The APC was funded by the corresponding author.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
Acne vulgaris is a common chronic inflammatory disease of the pilosebaceous unit [1], characterized by a complex pathological process involving multiple interacting factors [2]. Its core initiation lies in the dysfunction of the pilosebaceous unit, primarily manifesting as follicular hyperkeratosis [3], which obstructs sebum excretion and leads to the formation of microcomedones [4–6]. The subsequent overproliferation of Cutibacterium acnes not only directly activates the innate Immune system (e.g., via NOD-like receptor pathways) [7,8],but also secretes various enzymes and chemotactic factors, recruiting and activating inflammatory cells such as neutrophils and macrophages, thereby triggering a local inflammatory cascade [9–11]. A persistent inflammatory state further damages the follicular wall, inciting deep dermal inflammation and potentially resulting in aberrant tissue repair and pathological fibrotic scarring [9,12]. Consequently, acne progression is fundamentally a vicious cycle comprising “follicular hyperkeratosis-microbial colonization-inflammatory infiltration-tissue damage” [1,5,13]. Therefore, in order to treat acne, we need to break this cycle.
Current clinical treatments for acne, such as retinoids, benzoyl peroxide (BPO), and antibiotics (e.g., clindamycin, doxycycline), mainly aimed at reducing microbial load in tissues [14,15]. However, the long-term use of antibiotics has led to the emergence of resistant C. acnes strains, markedly reducing their efficacy and increasing the risk of recurrence [16,17]. More importantly, most existing treatments focus on inhibiting bacteria or correcting keratinization [18,19],but possess limited capacity to repair inflammation-induced tissue damage, promote the regeneration of normal structure, or prevent scar formation [19–21]. This “destruction-over-construction” therapeutic model struggles to fundamentally break the aforementioned vicious cycle, highlighting the urgent need for novel therapies that integrate potent antimicrobial and proactive repair functions.
To break through this bottleneck, this study proposes a combined therapeutic strategy, which is the combination of a natural antibacterial agent, Phellodendron bark Extract (PBE), and Human Umbilical Cord Mesenchymal Stem Cell-derived Exosomes (HucMSC-Exo). On the one hand, PBE, a traditional Chinese medicine rich in alkaloids like berberine (CAS: 208-418-8) and phellodendrine (CAS: 6873-13-8), has been proven to have spectral antibacterial and anti-inflammatory effects [22,23], which is a promising antibiotic alternative to effectively eliminate C. acnes and inhibit biofilm formation [24,25]. On the other hand, HucMSC-Exo, carriers of the paracrine functions of their parent cells, can precisely regulate immune responses [26], promote angiogenesis [27], accelerate epithelial regeneration [28], and facilitate collagen remodeling through the delivery of rich bioactive molecules (e.g., miRNAs, cytokines), thereby empowering tissue repair [29–31].
We hypothesize that the mixture of PBE and HucMSC-Exo may act in a combined manner: PBE creates and maintains a low-bacterial, “clean” microenvironment, clearing obstacles for HucMSC-Exo to exert optimal pro-repair efficacy. Then, HucMSC-Exo-driven tissue regeneration helps restore skin barrier function, thus reducing opportunities for bacterial colonization, thus establishing a virtuous cycle. To deal with the short duration of PBE action and the low targeting efficiency of Exo delivery, we further engineered an alginate-based hydrogel as a delivery vehicle for both. Alginate, an ionically crosslinked hydrogel, is particularly suitable for protecting the activity of biological macromolecules like Exo due to its mild gelation conditions. This hydrogel is expected to realize the continuous release of PBE, extend its antibacterial time, and act as the carrier of Exo, so that it can be slowly released locally at the lesion site.
In conclusion, current research on acne treatment predominantly focuses on single-function strategies, such as solely antibacterial or purely reparative approaches, leading to issues including low tissue repair efficiency, poor repair quality, and persistent infection. Existing treatment systems lack a delivery platform capable of simultaneously targeting the triple facets of “bacteria, inflammation, and tissue damage.” To address this gap, we designed an alginate-based hydrogel (PE gel) co-loaded with PBE and HucMSC-Exo. Our study first confirms via zeta potential, DLS, and TEM analyses that PBE and Exo remain as distinct particle populations in the mixture without forming new complexes or compromising their respective functions. Furthermore, unlike previous hydrogels that merely serve as passive carriers, our system achieves sustained PBE release (≥24 h) and combined action: PBE creates a low-bacterial, anti-inflammatory microenvironment, enabling Exo to exert maximal pro-repair effects. In our experimental design, C. acnes was used as the primary acne-relevant pathogen for both in vitro and in vivo studies. Staphylococcus aureus and Escherichia coli were additionally included in the in vitro antibacterial assays as representative Gram-positive and Gram-negative strains, respectively, to evaluate the broad-spectrum antibacterial potential of the hydrogel beyond C. acnes. Through in vitro characterization and in vivo validation in a C. acnes-induced rat ear acne model, we demonstrate that this system may help interrupt the vicious cycle of acne pathology via a combined antibacterial, anti-inflammatory, and pro-repair effects, providing an integrated, non-antibiotic therapeutic solution for acne management.
2. Materials and methods
2.1. Materials
Human umbilical cord mesenchymal stem cells (Huc-MSCs) were provided by Shenzhen wingor Biotechnology Co., Ltd. (Shenzhen, China). Cell expansion medium (MEMα, Cat# PM150421), Fetal Bovine Serum (FBS, Cat# 164210), and Penicillin-Streptomycin Solution (P/S, Cat# PB180120) were purchased from Procell Life Science & Technology Co., Ltd. (Wuhan, China). The BCA Protein Assay Kit (Cat# P0012S) was obtained from Beyotime Biotechnology (Shanghai, China). Phellodendron bark Extract (PBE) was provided by Shaanxi Xintianyu Biotechnology Co., Ltd. (Shaanxi, China), with its main active components identified by HPLC-MS (S1 Fig). Cutibacterium acnes (Strain ATCC 11827) and its specific culture medium (Cat# T2283), Escherichia coli (Strain ATCC 25922), and Staphylococcus aureus (Strain ATCC 25923) were purchased from the Shanghai Microbiological Culture Collection Center (Shanghai, China). The Live/Dead Bacterial Staining Kit (DMAO/PI, Cat# C2030S) was acquired from Beyotime Biotechnology (Shanghai, China). The Exosome Extraction Kit (Cat# BB-39046) was sourced from BB Biotechnology Co., Ltd. (Guangzhou, China). Sodium Alginate (M/G ≈ 1:1, CAS: 9005-38-3) and Calcium Chloride (CaCl2, CAS: 10043-52-4) were purchased from Shanghai Macklin Biochemical Technology Co., Ltd. (Shanghai, China).
Primary antibodies for western blot (WB)were as follows: anti-Calnexin rabbit polyclonal antibody (Cat# GB111369-100, Servicebio), anti-CD63 rabbit polyclonal antibody (Cat# GB115694-100, Servicebio), recombinant anti-TSG101 rabbit monoclonal antibody (Cat# GB150052-100, Servicebio), recombinant anti-CD9 rabbit monoclonal antibody (GB155697-100, Servicebio), Anti-CD81 rabbit polyclonal antibody (Cat# GB111073-100, Servicebio), anti-Hsp70 rabbit polyclonal antibody (Cat# GB11241-100, Servicebio), and anti-GAPDH mouse monoclonal antibody (Cat# 60004-1-lg, Proteintech). HRP-conjugated goat anti-rabbit IgG and goat anti-mouse IgG were used as secondary antibodies (Cat# SA00001-1, SA00001-2, Proteintech). Primary antibodies and detection reagents for immunohistochemistry (IHC) were as follows: anti-TNF-α rabbit polyclonal antibody (Cat# 17590-1-AP, Proteintech), anti-IL-6 rabbit polyclonal antibody (Cat# 21865-1-AP, Proteintech), anti-CD31 rabbit polyclonal antibody (Cat# 28083-1-AP, Proteintech), and anti-VEGFA rabbit polyclonal antibody (Cat# 26157-1-AP, Proteintech). IHC detection was performed using the Streptavidin-Biotin Complex (SABC) method with Biotin-conjugated goat anti-rabbit IgG(H+L) (Cat# SA00004-2, Proteintech) as the secondary antibody and HRP-conjugated Streptavidin (Cat# SA00001-0, Proteintech). All antibodies and detection reagents were used according to the manufacturers’ recommended protocols.
2.2. Bacterial culture and suspension preparation
E. coli and S. aureus were cultured in LB liquid medium at 37 °C with shaking at 200 rpm for 5 hours in incubated shaker (THZ-92C, Boxun, China) to the logarithmic growth phase. C. acnes was cultured in C. acnes specific medium using the Mitsubishi AnaeroPack system (AnaeroPack, Cat# C-11; Anaerobic Indicator, Cat# C-22; AnaeroPack Rectangular Jar, Cat# C-43) at 37 °C under static conditions for 72 hours to create a strict anaerobic environment. All bacterial cultures were harvested by centrifugation at 4000 ×g for 5 minutes. The pellets were resuspended in sterile physiological saline, and the concentration was accurately adjusted to 1 × 108 CFU/mL using the McFarland standard (DEN-1B, BioSan, Latvia) and confirmed by plate counting.
2.3. Stem cell culture and exosome isolation and purification
Huc-MSCs were cultured routinely in MEMα complete medium supplemented with 10% FBS and 1% P/S at 37 °C in a humidified incubator (Midi 40, Thermo Fisher Scientific, USA) with 5% CO2. When P2-P3 cells reached 85–95% confluence, the medium was replaced with serum-free MEMα basal medium, and the conditioned medium was collected after 48 hours. The collected supernatant was first subjected to differential centrifugation by using high-speed centrifuge (5430R, Eppendorf, Germany): 300 ×g for 10 min to remove floating cells, followed by 2000 ×g for 20 min to remove cell debris and apoptotic bodies. The clarified supernatant was then processed strictly according to the instructions of the exosome extraction kit. Finally, the purified exosome pellet was resuspended in sterile PBS, and the protein concentration was determined using the BCA assay (final concentration 250 μg/mL). Aliquots were stored at −80 °C.
To characterize the isolated exosomes, WB analysis was performed. Exosomes and Huc-MSC whole-cell lysates were lysed in RIPA buffer containing protease inhibitors. Equal protein amounts were separated by SDS-PAGE and transferred onto PVDF membranes. After blocking, membranes were incubated with primary antibodies against CD9, CD63, CD81, TSG101, Hsp70, Calnexin, and GAPDH overnight at 4 °C, followed by HRP-conjugated secondary antibodies. Signals were visualized by enhanced chemiluminescence.
Transmission electron microscopy (TEM) was also performed to detect the morphology of exosomes. Freshly isolated exosomes or the Exo+PBE mixture were adsorbed onto carbon-coated copper grids, negatively stained with uranyl acetate, and imaged using a transmission electron microscope.
2.4. PBE gradient antibacterial assay
To evaluate the antibacterial activity of PBE, we included C. acnes as the primary acne-associated pathogen, along with E. coli and S. aureus as representatives of Gram-negative and Gram-positive bacteria, respectively, as these species are also commonly found in acne lesions. The MIC of PBE was determined using the broth microdilution method in 96-well plates, following the principles established by the European Committee on Antimicrobial Susceptibility Testing [32]. According to EUCAST guidelines, the reference method for MIC determination is broth microdilution, which serves as the gold standard to which other susceptibility testing methods are calibrated. Briefly, the prepared bacterial suspensions were diluted with fresh medium to approximately 1 × 106 CFU/mL and inoculated into a 96-well cell culture plate (100 μL per well). Subsequently, an equal volume (100 μL) of sterile-filtered PBE solution at different concentrations was added to the experimental wells, resulting in final concentrations of 1.25, 2.5, 5, 10, and 20 mg/mL. Wells treated with an equal volume of sterile PBS served as the negative control, and wells containing only medium without bacteria were set as background controls. Each group had three replicates. Each concentration was tested in triplicate wells per plate, and the entire experiment was independently repeated three times. The inoculated 96-well plate was incubated at 37 °C under aerobic (E. coli, S. aureus) or anaerobic (C. acnes) conditions as required. The optical density at 600 nm (OD600) was measured using a Multiskan FC microplate reader (Thermo Fisher Scientific) after 2 and 5 hours of incubation for E. coli and S. aureus, and after 24, 48, and 72 hours for C. acnes, incubation times were selected based on their respective logarithmic growth phases determined in preliminary experiments. Given the slow growth rate of C. acnes, the incubation period was extended to 72 hours to allow sufficient bacterial proliferation, consistent with standard protocols for slow-growing anaerobes. To control for background and color interference from PBE, wells containing PBE at the corresponding concentration in medium without bacteria were included as blank controls for each concentration, and their OD600 values were subtracted from the corresponding test wells. After subtracting the background values, the inhibition rate was calculated using the formula: Inhibition Rate (%) = [(ODNegative Control – ODTest) / ODNegative Control] × 100%.
The MIC was defined as the lowest concentration of PBE that achieved ≥90% reduction in OD600 compared to the negative control [29]. Because complete growth inhibition of natural product extracts is often difficult to achieve due to compound solubility or color interference, this operational threshold was used instead of the absence-of-visible-growth criterion. To control for color interference from PBE itself, wells containing PBE at each concentration in medium without bacteria were included as blank controls, and their OD600 values were subtracted from the corresponding test wells [33]. For C. acnes, this threshold was further corroborated by visual inspection of turbidity, as complete clarity was consistently observed at the MIC. The MIC was additionally verified by CFU-based plate counting on agar plates, as described in Section 2.5. Since C. acnes is the primary pathogenic bacterium in acne vulgaris, the MIC of PBE against C. acnes was adopted as the standard concentration for subsequent experiments.
2.5. Drug-loaded plate antibacterial assay
Sterile PBE powder and/or HucMSC-Exo suspension were added to molten LB agar or C. acnes specific agar medium cooled to approximately 50 °C, achieving final concentrations of 5 mg/mL and 250 μg/mL, respectively. The mixture was poured into sterile Petri dishes to solidify. After solidification, diluted bacterial suspension (~1 × 106 CFU/mL) was evenly spread onto the plates using sterile swabs. The inoculated plates were incubated at 37 °C under aerobic (E. coli, S. aureus) or anaerobic (C. acnes) conditions for one growth cycle. The distribution of colonies was photographed using a high-resolution digital camera (90D, Canon, Japan), and the colony-forming units (CFUs) were counted and analyzed using ImageJ software (National Institutes of Health, USA).
2.6. Biofilm formation – crystal violet staining assay
Bacterial suspensions diluted in corresponding liquid media containing PBE (5 mg/mL) and/or HucMSC-Exo (250 μg/mL) were added to 12-well plates. E. coli and S. aureus were cultured aerobically for 24 hours, while C. acnes was cultured anaerobically for 72 hours without shake to allow biofilm formation. After incubation, the culture medium was carefully aspirated, and the wells were gently washed three times with ice-cold PBS to remove non-adherent planktonic bacteria. The plates were air-dried at room temperature. Then, 500 μL of 0.1% (w/v) crystal violet solution was added to each well and stained for 15 minutes at room temperature. After discarding the stain, the wells were repeatedly rinsed with PBS until the wash solution became colorless, followed by air-drying. The distribution of crystal violet-stained biofilms was photographed using a high-resolution digital camera, and the stained area was quantified using ImageJ software. The biofilm formation rate was calculated as: Biofilm Formation Rate = (Stained AreaTest / Stained AreaControl).
2.7. Bacterial live/dead staining
PBE (final concentration 5 mg/mL) and/or HucMSC-Exo (final concentration 250 μg/mL) were added to the prepared bacterial suspensions (final concentration 1 × 108 CFU/mL). The mixtures were incubated statically at 37 °C. E. coli and S. aureus were cultured for 5 hours, and C. acnes for 72 hours. After incubation, 1 mL of bacterial culture was centrifuged at 4000 ×g for 5 minutes. The supernatant was discarded, and the pellet was resuspended in 1 mL of sterile physiological saline. This washing step was repeated three times to thoroughly remove drug and medium residues. Finally, the bacterial concentration was adjusted to 1 × 108 CFU/mL with physiological saline. A 100 μL aliquot of the washed bacterial suspension was mixed with 1 μL of DMAO/PI working solution and incubated in the dark at room temperature for 15 minutes. After incubation, 10 μL of the stained suspension was placed on a glass slide, covered with a coverslip, and immediately observed and photographed using a Zeiss Axio Imager upright fluorescence microscope (Germany). The numbers of green (DMAO, live or dead) and red (PI, dead) fluorescent signals in random fields of view were counted using ImageJ software, and the bacterial survival and mortality rates were calculated.
2.8. Preparation of drug-loaded hydrogels
Sodium alginate powder was accurately weighed and dissolved in pre-warmed PBS to prepare a 1.5% (w/v) stock solution. This solution was sonicated by using ultrasonic cleaner (SB-1200DT, SCIENTZ, China) in a 60 °C water bath for 12 hours to ensure complete dissolution, followed by overnight degassing in a 60 °C air-bath shaker (THZ-92C, Boxun, China) to remove bubbles. PBE solution (final concentration 5 mg/mL), HucMSC-Exo suspension (final concentration 250 μg/mL), or their mixture (PE) was thoroughly mixed with the alginate solution. This mixture was then combined with a 2% (w/v) CaCl2 cross-linking solution at a volume ratio of 10:1 (alginate mixture: CaCl2) under continuous stirring at about 120 rpm for 2 minutes to induce the formation of a stable hydrogel network. Blank hydrogel (SA gel) without any active components was prepared simultaneously as a control. Four types of hydrogels were ultimately obtained: SA gel, Exo gel (final concentration 250 μg/mL), PBE gel (final concentration 5 mg/mL), and PE gel (final concentration 250 μg/mL of Exo and 5 mg/mL of PBE).
2.9. Characterization of PE complex and PBE release kinetics
The hydrodynamic size distribution and Zeta potential of HucMSC-Exo, PBE, and their physical mixture (PE) were analyzed using a Dynamic Light Scattering instrument (Zetasizer Nano ZS, Malvern Panalytical, UK). Samples were diluted 1:100 in sterile PBS (pH 7.4) to achieve a scattering intensity between 50–200 kcps before measurement and measured at 25 °C with a 173° backscatter angle. Each sample was measured in triplicate.
To evaluate the release behavior of PBE, the PBE gel was immersed in PBS release medium and incubated statically at 37 °C. At predetermined time points (1, 2, 4, 6, 8, 16, 24 hours), 1 mL of the release supernatant was collected and replaced with an equal volume of fresh, pre-warmed PBS. The collected supernatant was filtered through a 0.22 μm filter, and its absorbance was measured at 620 nm using a Multiskan FC microplate reader (Thermo Fisher Scientific).
2.10. In vitro antibacterial test of drug-loaded hydrogels
The previously prepared bacterial suspensions were diluted with fresh medium and inoculated into sterile centrifuge tubes pre-loaded with different hydrogels. An equivalent amount of bacterial suspension without hydrogel served as the blank control. The tubes were incubated statically at 37 °C. Samples were taken after 2 and 5 hours of incubation for E. coli and S. aureus, and after 24, 48, and 72 hours for C. acnes. At each sampling point, the suspension was vortexed thoroughly, and an aliquot was taken to measure the OD600 value. The relative inhibition rate of each hydrogel group at different time points was calculated using the formula: Inhibition Rate (%) = [(ODControl – ODHydrogel) / ODControl] × 100%.
2.11. Rat ear acne model construction and treatment
All animal procedures were approved by the Experimental Animal Welfare and Ethics Committee of Jiangmen Greater Bay Area International Institute for Innovative Healthcare (Approval No.: N2025008, 20 May 2025).
Thirty-six healthy male Sprague-Dawley (SD) rats (180–220 g) were purchased from Zhuhai Bestest Biotechnology Co., Ltd. Upon arrival, animals were group-housed (3–4 per cage) in standard individually ventilated cages (IVC) under specific pathogen-free (SPF) conditions. They were acclimatized for one week in an SPF-level animal laboratory. The laboratory environment was maintained at a constant temperature (22 ± 2) °C, humidity 30% ~ 50%, with a 12/12-hour light/dark cycle. Animals had ad libitum access to a standard laboratory rodent diet and autoclaved water. Environmental enrichment was provided in each cage, including sterile wooden blocks and nesting material, which were replaced weekly. Animal health and behavior were monitored daily by trained personnel.
All animals that completed the acclimatization period and met health criteria were included in the study. Animals were randomly allocated into six experimental groups (n = 6 per group, the sample size is the same as that for subsequent data analysis) using a computer-generated random number sequence (Microsoft Excel RAND function): Blank Control group, Model Control group, SA gel group, PBE gel group, Exo gel group, and PE gel group. The sample size (n = 6 per group) was based on previously published rat ear acne and skin wound studies using comparable models. A post-hoc power analysis using the primary endpoint (ear swelling resolution on day 11) confirmed that n = 6 per group provided >99% power to detect the observed effect sizes (Cohen’s d > 3) at α = 0.05 (two-sided). All animal procedures adhered to the 3R principles.
This model combines local intradermal injection to induce acne-like lesions in the ear with repeated intraperitoneal administration to mimic the systemic immune priming associated with chronic C. acnes colonization, as described in the original protocol [34]. Briefly, under isoflurane inhalation anesthesia (induced at 5% and maintained at 2–3% in oxygen) by using animal anesthesia and respiratory system (RWD Life Science, China), 50 μL of C. acnes suspension (6 × 107 CFU/mL) was slowly injected intradermally into the central area of the right ear pinna, once daily for 3 consecutive days. Subsequently, starting from day 3, 1 mL of the same concentration of bacterial suspension was injected intraperitoneally daily for 7 consecutive days. The left ear received no treatment.
The modeling procedure (intradermal ear injection) was classified as causing mild to moderate, transient discomfort. Based on our pilot study and established protocols for similar dermal inflammation models, the acute pain associated with a precise intradermal injection under deep anesthesia is very brief and resolves rapidly post-procedure without signs of persistent distress (excessive grooming, guarding, or altered locomotion). Therefore, postoperative analgesic administration was not routinely implemented. This decision was explicitly reviewed and approved by the overseeing ethics committee. However, all animals were closely monitored for signs of pain or distress (vocalization, lethargy, reduced food/water intake) for at least 2 hours post-procedure and twice daily thereafter. Had any signs of significant pain been observed, rescue analgesia (buprenorphine, 0.05 mg/kg, subcutaneously) would have been administered immediately. No such intervention was required in this study.
The day after modeling completion was designated as treatment day 0 (D0). Treatment involved topically applying approximately 1 g (covering the surface of acne) of the respective hydrogel to the right ear lesions twice daily for 11 consecutive days. Ear thickness at the same site was precisely measured using a digital micrometer on day −9 (pre-modeling), D0, D7, and D11. The swelling resolution rate was calculated as: Swelling Resolution Rate (%) = [(ThicknessD0 – ThicknessDn) / (ThicknessD0 – ThicknessPre-modeling)] × 100%. Redness, inflammation, and recovery of the ear were observed and recorded daily. To minimize confounding factors, the order of daily treatment administration and ear thickness measurement was randomized each day. Furthermore, cages were repositioned on the rack in a randomized manner twice weekly to account for any environmental gradients within the animal room. The investigators responsible for the daily clinical observation of ear redness and inflammation, as well as all personnel conducting the subsequent histopathological analysis and bacterial colony counting, were blinded to the group allocations throughout the experiment and data analysis phase. Coded identifiers were used for samples and data sheets until all analyses were complete.
2.12. Tissue bacterial load and histopathological analysis
Twenty-four hours after the last treatment, all rats were humanely euthanized for tissue collection. Animals were first placed in an induction chamber and deeply anesthetized with 5% isoflurane in oxygen. Loss of consciousness and abolition of pedal withdrawal reflex were confirmed. Subsequently, while under deep surgical plane anesthesia, euthanasia was performed by cervical dislocation, conducted by a highly trained technician.
Prior to the planned endpoint, predefined humane endpoint criteria were established and required immediate euthanasia if observed: 1) Severe ulceration or necrosis at the injection site exceeding 20% of the ear area; 2) Weight loss >20% of baseline body weight within 48 hours; 3) Signs of severe systemic infection (e.g., hunched posture, piloerection, lethargy unresponsive to environmental stimuli, inability to access food or water); 4) Any self-mutilation behavior directed at the ear lesion. No animals reached these humane endpoints during the study, and all survived to the planned experimental termination.
Immediately after euthanasia, the full-thickness skin tissue from the right ear modeling site was aseptically collected by using surgical scissors. The tissue was weighed and immediately partially placed in pre-cooled sterile PBS, cut into approximately 1 mm3 pieces with ophthalmic scissors, mixed with 1 mL PBS, and homogenized using an IKA T10 homogenizer (Germany) at 15,000 rpm for 30 seconds intermittently. The tissue homogenate was serially diluted ten-fold. A 100 μL aliquot of an appropriate dilution was evenly spread onto C. acnes specific agar plates. The plates were incubated anaerobically at 37 °C for 72 hours, after which colonies were counted. Colonies were quantified using ImageJ, and the tissue C. acnes load rate was calculated relative to the Model Control group (set at 100%).
Simultaneously, the other part of the tissue blocks were fixed in 10% neutral buffered formalin for 24–48 hours, dehydrated through a graded ethanol series, embedded in paraffin, and continuously sectioned (4 μm thickness) using an HM340E microtome (Thermo Fisher Scientific). Sections were stained with Hematoxylin and Eosin (H&E) and Masson’s Trichrome. H&E-stained sections were used to measure the cross-sectional area of acne-like pustules and the area of follicular hyperkeratosis. Masson’s trichrome-stained sections were used to analyze collagen fiber maturity and alignment orderliness. All sections were observed and images were captured using a PreciPoint O8 microscopic imaging system (Germany). Quantitative analysis was performed using ImageJ software.
2.13. Immunohistochemical analysis
Paraffin sections were deparaffinized and rehydrated. Antigen retrieval was performed by heat induction in citrate buffer (pH 6.0). Subsequently, sections were incubated overnight at 4 °C with diluted primary antibodies: Anti-TNF-α (1:200), Anti-IL-6 (1:200), Anti-CD31 (1:100), and Anti-VEGFA (1:150). After washing with PBS, sections were sequentially incubated with Biotin-conjugated Goat Anti-Rabbit IgG secondary antibody (1:500) and HRP-conjugated Streptavidin (1:1000), each for 30 minutes at room temperature. Finally, color development was performed using a DAB substrate kit, and nuclei were counterstained with Hematoxylin. Negative controls were performed by replacing the primary antibody with PBS. All sections were processed, developed, and photographed under the same conditions. Microvessel density (MVD) was assessed by immunohistochemical staining for CD31, other positive staining in specific regions was analyzed for positive area by using the IHC Profiler plugin in ImageJ software.
2.14. Safety and biocompatibility evaluation of hydrogel formulation
To evaluate the in vivo safety and biocompatibility of the PE gel, a separate cohort of SD rats was used. The C. acnes-induced rat ear acne model was established as described in Section 2.11. The rats were randomly divided into two groups (n = 6 per group): (1) Control group, which received no treatment after model induction; and (2) PE gel group, which received topical application of the PE gel to the acne lesions twice daily for 11 consecutive days. At the end of the treatment period, rats were euthanized, and blood samples were collected. Serum was separated and subjected to biochemical analysis, including alkaline phosphatase (ALP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), creatinine (CREA-S), total protein (TP II), and urea (UREA). Major organs including the heart, liver, spleen, lung, and kidney were harvested, fixed in 4% paraformaldehyde, embedded in paraffin, sectioned, and stained with hematoxylin and eosin (H&E) for histopathological examination.
2.15. Statistical analysis
All in vitro experiments were performed with three technical replicates per group (n = 3) in a single experimental run. Data are presented as mean ± standard deviation (SD). In vivo experiments were performed with six rats per group (n = 6), and data are presented as mean ± standard error of the mean (SEM) to reflect inter-animal variability.
Statistical analysis was performed using SPSS 27.0 software. Normality was assessed using the Shapiro-Wilk test, and homogeneity of variance was assessed using Levene’s test. Comparisons among multiple groups were conducted using one-way analysis of variance (ANOVA). When homogeneity of variance was satisfied, Tukey’s post hoc test was used for pairwise multiple comparisons to control for the increased risk of type I error associated with multiple group comparisons; LSD was not used. When homogeneity of variance was satisfied, Tukey’s post hoc test was used for pairwise multiple comparisons. When variances were unequal, Dunnett’s T3 test was applied. A p-value less than 0.05 was considered statistically significant. Exact p-values are provided in the figure where applicable. The graphical elements in Figs 2, 4, 6, and 7 were obtained from BioGDP [35] under a CC BY license.
3. Results
3.1. PBE exhibits potent inhibitory activity against C. acnes and broad-spectrum assessment against S. aureus and E. coli and Demonstrates Good Biocompatibility with HucMSC-Exo
HPLC-MS analysis identified the main active components of the PBE used in this study as berberine (CAS: 208-418-8), accounting for 74.81% of the total content, and phellodendrine (CAS: 6873-13-8), accounting for 25.00% (S1 Fig). To validate the preservative and antibacterial capability of PBE, different concentrations of PBE were added to cultures of E. coli, S. aureus, and C. acnes during their logarithmic growth phase and co-incubated for one bacterial growth cycle (5h for E. coli/S. aureus; 72h for C. acnes) (Figs 1A–1C). C. acnes was used as the primary acne-relevant pathogen, while S. aureus and E. coli were included to assess whether the antibacterial effect extended to other clinically relevant Gram-positive and Gram-negative bacteria. For E. coli, the inhibition rate of all concentration groups exceeded 50% at 2h, indicating that the minimum concentration reached the half-maximal inhibitory concentration (IC50) at this time point. The inhibition rate increased with PBE concentration. At 5h, the inhibition rates in all experimental groups decreased significantly compared to their respective 2h rates, potentially due to PBE degradation, microenvironmental changes, or PBE-bacteria interactions. PBE showed similar inhibitory effects against S. aureus and C. acnes. For C. acnes, the inhibition rate exhibited a concentration-dependent increase. At 5 mg/mL, the inhibition rate reached 92.3% ± 3.1% after 72 h of incubation, and no visible turbidity was observed in the culture wells. Concentrations below 5 mg/mL (1.25 and 2.5 mg/mL) resulted in inhibition rates below 50%, with visible turbidity comparable to the growth control. Based on the predefined criterion of ≥90% OD600 reduction, the MIC of PBE against C. acnes was determined to be 5 mg/mL. This was further supported by CFU-based plate counting, which showed an approximately five-order-of-magnitude reduction in colony numbers at 5 mg/mL compared with the control (Fig 2D). Consequently, 5 mg/mL was selected as the working concentration for PBE in subsequent experiments.
(A)E. coli, (B)S. aureus, and (C)C. acnes were treated with increasing concentrations of PBE (0, 1.25, 2.5, 5 mg/mL) during logarithmic growth. Inhibition rates were measured at indicated time points (up to 5 h for E. coli and S. aureus, and up to 72 h for C. acnes). Data represent mean ± SD (n = 3). The IC50 for E. coli was reached at 2 h even at the lowest concentration, and the MIC for C. acnes was determined as 5 mg/mL (≥90% OD600 reduction at 72 h).
(A) Representative agar plates showing bacterial colonies after treatment with control (CTL), Exo, PBE, or PE (PBE + Exo) mixture. (B–D) Quantification of colony‑forming units (CFU) for E. coli, S. aureus, and C. acnes, respectively. PBE and PE groups reduced colony counts by approximately two orders of magnitude for E. coli and S. aureus, and about five orders for C. acnes, compared to CTL. Exo alone did not reduce colony counts. PE showed no significant difference from PBE alone, indicating that Exo does not impair PBE’s antibacterial activity. Data are presented as mean ± SD of three technical replicates (n = 3). Statistical analysis was performed by one-way ANOVA followed by Tukey’s post hoc test. Created with BioGDP.com. Reprinted from [34] under a CC BY license, with permission from the authors, original copyright 2025.
To verify the combined effect of PBE and Exo on bacterial colony counts, agar plates loaded with 5 mg/mL PBE were prepared and bacteria were cultured for one growth cycle (Fig 2A). For E. coli and S. aureus, the colony numbers in the PBE and PE groups were reduced by approximately two orders of magnitude compared to the CTL and Exo groups, showing significant differences (Figs 2B–2C). For C. acnes, the difference reached approximately five orders of magnitude (Fig 2D). These results demonstrate the powerful antibacterial capability of PBE. However, the colony counts in the Exo group itself were not significantly different from the CTL group, indicating Exo lacks inherent antibacterial activity. Crucially, the colony counts in the PE group showed no significant difference from the PBE group, indicating that PBE’s potent antibacterial action is retained when mixed with Exo.
To investigate the effect of Exo and PBE on bacterial biofilm formation, biofilms were cultured in 12-well plates and assessed using the crystal violet staining method (Fig 3A). The biofilm formation rates in the PBE and PE groups were significantly lower than those in the CTL and Exo groups (Figs 3B–3D). Interestingly, compared to the CTL group, the Exo group showed a trend towards more intact biofilms, potentially suggesting that miRNAs or proteins carried by exosomes might support bacterial growth or directly facilitate biofilm formation. In contrast, biofilms in the PBE and PE groups were fragmented, indicating a strong inhibitory effect on biofilm formation, with no significant difference between these two groups.
(A) Crystal violet staining of biofilms formed by the three bacteria under different treatments. (B–D) Quantification of biofilm formation rates (percentage of control) for E. coli, S. aureus, and C. acnes. PBE and PE significantly reduced biofilm formation compared to CTL and Exo groups. Exo showed a slight, non‑significant trend toward increased biofilm integrity. Data are presented as mean ± SD of three technical replicates (n = 3). Statistical analysis was performed by one-way ANOVA followed by Tukey’s post hoc test.
To further assess the impact of Exo and PBE on bacterial viability, bacteria were double-stained with DMAO/PI after treatment. Viable bacteria stain green with DMAO, while dead bacteria stain red with PI. The results aligned with previous findings. Compared to the CTL group, the proportion of live bacteria in the Exo group was possibly higher but not significantly different, whereas the PBE and PE groups showed significantly higher proportions of dead bacteria. Specifically, the bacterial mortality rates for E. coli were 55% (PBE) and 49% (PE); for S. aureus, 50% (PBE) and 63% (PE); and for C. acnes, 55% (PBE) and 70% (PE) (S2 Fig). These findings indicate that the potent antibacterial effect of PBE dominates the microbial state of the microenvironment. Even in the presence of Exo, PBE effectively inhibits bacterial growth and biofilm formation. The strong antibacterial activity of PBE is sufficient to counteract any potential mild promoting effect of Exo on bacterial biofilms, ensuring an overall potent antibacterial outcome in the composite system (PE gel).
In conclusion, PBE is a natural preservative and antibacterial agent with potent, broad-spectrum antibacterial activity and a remarkable ability to inhibit biofilm formation, with a core effective concentration of 5 mg/ml. Furthermore, when co-existing with Exo, PBE’s core antibacterial efficacy remains unaffected and can effectively counteract the potential pro-bacterial or pro-biofilm effects of Exo by inducing bacterial death. Therefore, in acne management, using PBE to protect Exo may create an antibacterial environment, enhancing Exo’s therapeutic efficacy. Simultaneously, PBE itself is a potential core component for acne treatment, potentially producing additive effects with Exo’s therapeutic actions, thereby strengthening acne management.
3.2. Fabrication and performance validation of PE Gel
To counteract the decline in PBE’s antibacterial capacity over prolonged interaction times with bacteria, an alginate hydrogel was employed to load both Exo and PBE, aiming to extend the duration of action. Fig 4 illustrates the preparation flowchart for the PE-loaded hydrogel. Conditioned medium was collected from P2-P3 Huc-MSCs at 85–95% confluence under serum-free conditions. Cell debris, apoptotic bodies, and large particles were removed via differential centrifugation. Exosomes were isolated and purified using an exosome extraction kit, resuspended in PBS, mixed with PBE to form the PE mixture, and finally loaded onto a calcium alginate hydrogel formed by cross-linking sodium alginate with calcium ions. Western blot analysis confirmed that the isolated vesicles were enriched for CD9, CD63, CD81, and TSG101 compared with whole-cell lysates, whereas calnexin, Hsp70, and GAPDH were barely detected in the exosome fraction (S3 Fig A and S1 File). TEM showed that the isolated exosomes exhibited typical cup-shaped vesicular morphology (S3 Fig B). These results confirm the successful isolation and purification of HucMSC-derived exosomes.
(A) Schematic flowchart of PE gel fabrication, including isolation of HucMSC‑Exo from conditioned medium, mixing with PBE, and crosslinking with sodium alginate to form calcium alginate hydrogel. (B) Hydrodynamic size distributions (by DLS) of Exo, PBE, and PE mixture. Exo showed a monodisperse size of ~141 nm, PBE ~ 199 nm, and PE mixture ~187 nm with a higher PDI. (C) Zeta potential distributions; the PE mixture displayed a bimodal profile, indicating coexistence of independent PBE and Exo populations without complex formation. (D) Cumulative release profile of PBE from the hydrogel over 24 h, demonstrating sustained release and maintenance of effective antibacterial concentration.Created with BioGDP.com. Reprinted from [34] under a CC BY license, with permission from the authors, original copyright 2025.
To investigate the interaction between Exo and PBE, their hydrodynamic size and Zeta potential, along with that of the PE mixture, were analyzed (Figs 4B–Fig 4C). Dynamic light scattering (DLS) analysis revealed that Exo presented a monodisperse size distribution with a diameter of 141.4 ± 2.0 nm (PDI = 0.132). In contrast, PBE particles were larger (199.2 ± 1.6 nm) and more heterogeneous (PDI = 0.285), PE mixture exhibited an intermediate size of 187.3 ± 2.6 nm (PDI = 0.291) (Fig 4B). The analysis of zeta potential provides new insights into this result, Exo displayed a strongly negative surface charge of −31.18 ± 0.32 mV, consistent with their stable, lipid-based membrane. PBE, however, showed a near-neutral potential of −3.82 ± 0.79 mV, indicative of poor stability. Critically, the zeta potential distribution for the PE mixture was unequivocally bimodal (Fig 4C), featuring two distinct populations with peak values at approximately −7.4 mV and −26.9 mV. This profile demonstrates the coexistence of two independent particle populations, with their respective surface potentials closely aligning with those of single PBE and Exo. TEM of the Exo + PBE mixture showed intact exosome vesicles that remained separate from PBE particles (S3 Fig C), further confirming that Exo and PBE can still exist alone in a mixture without forming new complexes or significantly altering each other’s surface properties.
To validate the ability of the fabricated hydrogel to maintain an antibacterial environment, the release profile of PBE was characterized (Fig 4D). The instantaneous concentration was calculated based on a pre-established PBE concentration-absorbance standard curve (S4 Fig), and the release profile was plotted. PBE was able to maintain an effective antibacterial concentration over 24 hours, demonstrating that the hydrogel prolongs PBE release, achieving a sustained-release effect.
Furthermore, to simulate practical acne management scenarios, the drug-loaded hydrogel was co-cultured with bacterial suspensions, and its antibacterial ability was assessed at different time points (Figs 5A–5B). Over 72 hours, the PE hydrogel exhibited significantly stronger antibacterial capability compared to the SA hydrogel (Figs 5C–5E). It is worth noting that the SA hydrogel group also showed a certain degree of antibacterial effect, possibly due to bacterial adsorption onto the hydrogel surface or interior, leading to a decrease in the bacterial load in the suspension.
(A–B) Experimental setup for co‑culture of hydrogels with bacterial suspensions and sampling at indicated time points. (C–E) Antibacterial effects against E. coli, S. aureus, and C. acnes, respectively, measured as bacterial viability over 72 h. PE hydrogel exhibited significantly stronger and sustained inhibition compared to SA hydrogel at all time points. SA hydrogel alone showed moderate antibacterial effect, possibly due to bacterial adsorption. Data are presented as mean ± SD of three technical replicates (n = 3). Statistical analysis was performed by one-way ANOVA followed by Tukey’s post hoc test.
In summary, Exo and PBE could be co-loaded into the hydrogel and remained as distinct particle populations. The prepared PE gel enables sustained release of PBE, resulting in superior antibacterial performance.
3.3. PE gel accelerates acne resolution and tissue repair
An in vivo experiment was conducted using a C. acnes-induced rat ear acne model to evaluate the antibacterial and acne-eliminating capabilities of the PE gel (Fig 6A). The drug-loaded hydrogel groups showed high acne resolution rates on both day 7 and day 11 (Fig 6B). On day 7, the swelling resolution rate was not significantly different between the Exo gel group (45.74% ± 9.73%) and PE gel group (59.64% ± 10.57%), whereas a significant difference existed between the PBE gel group (32.93% ± 8.89%) and PE gel group (59.64% ± 10.57%). By day 11, significant differences were observed among all three drug-loaded hydrogel groups. Notably, a significant difference consistently existed between the PBE gel group (39.10% ± 10.57%) and PE gel group (74.91% ± 8.32%) throughout the treatment period. This suggests that Exo likely plays a dominant role in the acne repair process during the treatment phase, while PBE promotes this repair function. Importantly, the model group’s acne resolution rates were only about 2.79% ± 5.36% (day 7) and 6.28% ± 9.17% (day 11), and the SA gel group’s rates were about 10.49% ± 8.23% (day 7) and 11.57% ± 7.29% (day 11), with no significant difference between these two groups. This indicates weak self-healing in this model and limited therapeutic efficacy of SA gel alone.
(A) Schematic of the C. acnes‑induced acne model and treatment regimen with different hydrogels. (B) Swelling resolution rates at day 7 and day 11 for model, SA gel, Exo gel, PBE gel, and PE gel groups. PE gel showed the highest resolution rate (74.91 ± 8.32% at day 11), significantly higher than all other groups. Exo gel also showed moderate resolution, while model and SA gel exhibited minimal self‑healing. Data are presented as mean ± SEM (n = 6 per group). Created with BioGDP.com. Reprinted from [34] under a CC BY license, with permission from the authors, original copyright 2025.
Fig 7A shows photographs of the rat ear acne model on treatment days 7 and 11. Compared to the control group, the modeling site in the model group exhibited obvious redness (or purplish discoloration), inflammation, and swelling, confirming successful model establishment. The difference between the SA gel group and the model group was relatively small. In contrast, the other hydrogel treatment groups showed apparent differences, with reduced redness and swelling, becoming more pronounced by day 11 compared to day 7.
(A) Representative photographs of rat ears on days 7 and 11, showing redness, swelling, and lesion resolution. (B) Bacterial culture results from tissue homogenates (agar plates). (C) Quantification of C. acnes load (percentage relative to SA gel group). The PE gel group had the lowest bacterial load (0.18 ± 0.02%), while Exo gel also significantly reduced the load compared to SA gel, possibly via immune modulation. Data are presented as mean ± SEM (n = 6 per group). Created with BioGDP.com. Reprinted from [34] under a CC BY license, with permission from the authors, original copyright 2025.
To investigate C. acnes colonization in the rat ear tissue, bacterial culture was performed on tissue homogenates from the modeling site upon sacrifice: SA gel group (88.68% ± 0.91%), PBE gel group (47.76% ± 2.36%), Exo gel group (57.03% ± 2.48%), PE gel group (0.18% ± 0.02%) (Fig 7B). The C. acnes load was significantly reduced in all experimental groups, with notable differences between groups (Fig 7C). Interestingly, the C. acnes load in the Exo gel group also decreased significantly, potentially because Exo-mediated repair of acne tissue improves the local microenvironment and activates immune functions, thereby partially eliminating the bacteria. Notably, the C. acnes load in the PE gel group was only about 0.18%, demonstrating the superior management effect of PE gel on acne tissue.
3.4. Histopathological characteristics
Transverse and longitudinal sections of rat tissue were subjected to H&E and Masson’s trichrome staining (Fig 8). The results indicated that all groups except the CTL group developed classic acne-like lesions. The CTL group showed minor non-acne-like lesions, likely due to the injection trauma of saline (Fig 8, S5 Fig).
Representative images of transverse and longitudinal sections of rat ear tissues from each group. H&E staining reveals inflammatory infiltration and cyst formation; Masson’s trichrome stains collagen (blue). The CTL group showed minor non‑acne‑like lesions due to injection trauma. Scale bars indicate 500 µm (100x) and 100 µm (400x) (or as appropriate). Drug‑loaded gel groups, especially PE, showed reduced cyst size and improved collagen organization.
Compared to the MOD group (0.98 mm2 ± 0.09 mm2), except the SA gel group (0.87 mm2 ± 0.08 mm2), the cross-sectional area of acne-like pustules was significantly reduced in the Exo gel (0.51 mm2 ± 0.07 mm2), PBE gel (0.63 mm2 ± 0.09 mm2), and PE gel (0.28 mm2 ± 0.08 mm2) treatment groups, with the PE gel group showing the greatest reduction (Fig 9A). Similarly, except the SA gel group (60.95% ± 11.01%) the area of follicular hyperkeratosis was significantly decreased in these treatment groups (Exo gel: 14.06% ± 4.03%, PBE gel: 19.75% ± 4.73%, PE gel: 11.22% ± 3.34%) compared to the MOD group (71.74% ± 6.35%) (Fig 9B). This indicates that the drug-loaded gels markedly ameliorated key pathological features of acne. Furthermore, compared to the MOD group, in gel groups containing Exo, collagen maturity was significantly increased (MOD: 28.29% ± 4.83%, Exo gel: 72.29% ± 3.22%, PE gel: 73.40% ± 6.75%), and the alignment entropy was significantly decreased (MOD: 0.68 ± 0.07, Exo gel: 0.19 ± 0.04, PE gel: 0.09 ± 0.03), suggesting that Exo within the gels likely plays a leading role in driving the related repair processes, promoting more organized collagen deposition (Figs 9C–Fig 9D).
(A) Cross‑sectional area of acne‑like pustules (mm2). (B) Area of follicular hyperkeratosis (%). (C) Collagen maturity (%). (D) Collagen alignment entropy. PE gel significantly reduced pustule area and hyperkeratosis, and increased collagen maturity while decreasing entropy, indicating effective tissue remodeling. Exo‑containing gels (Exo gel and PE gel) showed superior collagen quality. Data are presented as mean ± SEM (n = 6 per group).
3.5. Combined mechanism: Triple action of antibacterial, anti-inflammatory, and pro-repair effects accelerates acne repair
To further elucidate the pathological manifestations, immunohistochemistry (IHC) was performed on the transverse and longitudinal sections of CD31, VEGF, IL-6 and TNF-α, (Figs 10–12, S6 Fig). Compared to the MOD group, the results showed that treatment groups containing Exo showed a significant increase of VEGF levels (MOD: 6.28% ± 3.13%, Exo gel: 25.09% ± 8.77%, PE gel: 29.48% ± 8.17%), as well as an increase in the number of microvessels (MOD: 12.08 ± 4.75, Exo gel: 55.72 ± 4.94, PE gel: 57.51 ± 6.53) (Figs 11A–11B), suggesting that Exo primarily drives angiogenesis during PE gel treatment.
Representative images of CD31 expression (brown) in transverse and longitudinal sections from each group. CD31 marks microvessels. Groups containing Exo (Exo gel and PE gel) showed increased CD31‑positive vascular structures compared to MOD and PBE gel groups. Scale bars indicate 500 µm (100x) and 100 µm (400x) (or as appropriate).
(A) VEGF expression levels (% positive area). (B) Number of microvessels (CD31 + vessels per field). Exo‑containing groups (Exo gel and PE gel) exhibited significantly higher VEGF expression and microvessel density than MOD and PBE gel groups, indicating that Exo drives angiogenesis during treatment. Data are presented as mean ± SEM (n = 6 per group).
Representative images showing positive staining (brown) for TNF‑α and IL‑6 in lesions from different treatment groups. PBE‑containing groups (PBE gel and PE gel) displayed markedly reduced immunoreactivity compared to MOD and Exo gel groups, indicating strong anti‑inflammatory effects. Scale bars indicate 500 µm (100x) and 100 µm (400x) (or as appropriate).
Conversely, compared to the MOD group, treatment groups containing PBE exhibited significantly lower levels of TNF-α (MOD: 51.49% ± 5.61%, PBE gel: 11.88% ± 3.82%, PE gel: 10.82% ± 4.41%) and IL-6 (MOD: 19.21% ± 3.75%, PBE gel: 0.86% ± 0.36%, PE gel: 0.46% ± 0.32%) (Figs 13A–13B), indicating that PBE likely plays a dominant role in suppressing inflammatory cytokines during PE gel treatment. In summary, the PE gel accelerates acne repair through a triple mechanism involving antibacterial, anti-inflammatory, and pro-repair actions.
(A) TNF‑α positive area (%). (B) IL‑6 positive area (%). PBE and PE gels significantly suppressed both TNF‑α and IL‑6 expression compared to MOD and Exo gel groups. Exo gel alone did not reduce these cytokines, highlighting that PBE is primarily responsible for anti‑inflammatory action. Data are mean ± SEM (n = 6).
3.6. Safety and biocompatibility evaluation of PE gel
To assess the in vivo safety of the PE gel, serum biochemical parameters and major organ histology were evaluated after 14 days of topical treatment. Serum levels of ALP, ALT, AST, CREA-S, TP II, and UREA showed no significant differences between the control and PE gel groups (S7 Fig), indicating no obvious hepatic or renal toxicity under the tested conditions. H&E staining of the heart, liver, spleen, lung, and kidney revealed normal histological architecture in both groups, with no signs of inflammation, necrosis, or other pathological changes (S8 Fig). These results suggest that the PE gel exhibits good biocompatibility and no apparent systemic toxicity following topical application.
4. Discussion
Our results support the potential of PBE as a natural antibacterial agent. Combining stem cell-derived components with natural product extracts has been explored as a combinatorial therapeutic strategy [36], which is conceptually consistent with the PBE–Exo design used in the present study. Its broad-spectrum and potent antibacterial capacity, particularly the strong killing effect and biofilm inhibition against the key acne pathogen, C. acnes, may contribute to disrupting the microbial colonization–inflammation cycle [37,38]. In our experimental design, C. acnes served as the primary and clinically relevant pathogen for both in vitro and in vivo studies. S. aureus and E. coli were additionally included in the in vitro antibacterial and biofilm assays as representative Gram-positive and Gram-negative strains, respectively, to evaluate the broad-spectrum antibacterial potential of the hydrogel beyond C. acnes. Accordingly, the conclusions regarding acne treatment are based primarily on the C. acnes data and the C. acnes-induced in vivo model. An unexpected observation in our study was that HucMSC-Exo alone exhibited a slight but consistent tendency to enhance bacterial biofilm formation, as evidenced by crystal violet staining (Fig 3). While this effect did not reach statistical significance, the trend was reproducible across three independent experiments for all tested bacterial strains. This finding aligns with several previous reports suggesting that exosomes derived from various cell sources may inadvertently support bacterial growth or biofilm integrity.For instance, exosomes may provide lipids, proteins, and nucleic acids that can be utilized by bacteria as additional nutrient sources [39]. Additionally, the phospholipid bilayer of exosomes may serve as a scaffold that facilitates bacterial adhesion and biofilm matrix assembly [40]. Alternatively, exosome-carried microRNAs or growth factors may indirectly modulate bacterial quiescence or metabolic activity, though the precise mechanisms remain to be elucidated [41]. Importantly, this subtle pro-biofilm effect was completely abolished when PBE was co-administered with Exo, and the combination maintained the full antibacterial potency of PBE alone [42]. It should be emphasized that these explanations remain speculative and were not directly tested in the present study; dedicated experiments are needed to determine which, if any, of these mechanisms contribute to the mild pro-biofilm tendency observed for HucMSC-Exo.
This observation carries two significant implications. First, it underscores the importance of including an effective antimicrobial component when using Exo-based therapies in infectious or colonized tissue environments, as Exo alone may not be sufficient to control bacterial burden and could theoretically create conditions conducive to biofilm persistence; Second, it highlights the functional complementarity within the PE gel system: PBE not only directly eliminates bacteria but also neutralizes any potential unintended pro-biofilm activity of Exo, thereby ensuring that Exo’s regenerative functions are exerted in a “clean” microenvironment free from bacterial interference [43–45]. This combined action can be viewed as a complementary antibacterial–repair process, in which PBE-mediated bacterial clearance creates a more favorable microenvironment for Exo-mediated tissue repair.
Applying this pattern requires a delivery system. Alginate hydrogels, known for their biocompatibility and mild gelation conditions [46,47], are ideal for protecting the integrity and bioactivity of delicate biologicals like Exo [48,49]. Furthermore, their tunable porous structure enables the sustained release of PBE, thereby prolonging its antibacterial presence at the target site [50]. To overcome the limited duration of PBE action and the risk of Exo contamination, we introduced an alginate-based hydrogel as the delivery vehicle (Fig 4A). Characterization results showed that the PE gel enabled sustained release of PBE, maintaining an effective antibacterial concentration for up to 24 hours, directly overcoming the rapid decline in efficacy seen with free PBE due to degradation or bacterial interaction (Fig 4D). Simultaneously, particle size and Zeta potential analyses suggested that PBE and Exo coexisted as distinct particle populations without evidence of stable complex formation, implying that the hydrogel mainly serves as a co-delivery matrix for both components rather than stabilizing a specific PBE–Exo complex (Figs 4B–4C). This delivery system supports the combined contribution of antibacterial and regenerative activities at the lesion site, with PBE providing bacterial clearance and Exo promoting repair.
In vivo experiments showed that the PE gel improved therapeutic outcomes compared with single-component gels. Notably, the significant reduction in bacterial load observed in the Exo gel group is not attributable to a direct antibacterial effect of Exo but rather is consistent with the hypothesis of “repair-driven microenvironment remodeling”: Exo, by accelerating tissue repair, restoring skin barrier function, and potentially modulating local immunity, may indirectly create less favorable for C. acnes colonization [51]. Although alginate hydrogels may provide moisture retention and a supportive microenvironment, the SA gel group showed only minimal therapeutic efficacy in our model, indicating that the alginate matrix alone was insufficient. The superior effects of the PE gel are therefore primarily attributable to PBE and HucMSC-Exo, with the hydrogel mainly serving as a delivery vehicle that maintains hydration and enables sustained local release.
The assessment of acne progression and recovery hinges on key histopathological features, primarily the integrity of the pilosebaceous unit and the state of the dermal extracellular matrix [52,53]. The formation of inflammatory cysts and microcomedones, characterized by follicular hyperkeratosis and dilated, rupture-prone follicles, represents a core pathological event in acne, directly contributing to clinical manifestations such as papules, pustules, and subsequent tissue destruction [54,55]. Concurrently, the disordered and immature deposition of collagen fibers, a hallmark of aberrant healing often observed in chronic inflammatory skin conditions, undermines skin structural strength and function, paving the way for atrophic scarring [56,57]. Therefore, quantifying the reduction of cystic structures and the restoration of mature, well-organized collagen are critical metrics for evaluating therapeutic efficacy against acne and its sequelae. In this study, the therapeutic efficacy of the PE gel was first corroborated by histopathological analysis, which revealed significant improvements in these key acne-related pathologies. Specifically, H&E staining demonstrated a marked reduction in the size of acne-like cysts and the area of follicular hyperkeratosis, indicating effective resolution of the core lesions (Figs 9A–9B). Furthermore, Masson’s trichrome staining confirmed enhanced collagen maturity and a more organized fiber arrangement within the dermis, signifying active tissue remodeling and functional repair (Figs 9C–9D).
To delve into the underlying mechanisms driving these observed histological improvements, we turned to molecular-level analysis of key biomarkers governing inflammation and regeneration. The inflammatory cascade in acne is largely driven by pro-inflammatory cytokines such as TNF-α and IL-6, which perpetuate tissue damage and hinder repair [58,59]. Conversely, the process of tissue regeneration relies critically on angiogenesis [58], facilitated by markers like CD31 and VEGF [60,61], which ensure the delivery of oxygen and nutrients necessary for rebuilding functional tissue [62]. Against this mechanistic backdrop, our findings indicate that PBE contributed mainly to antibacterial and anti-inflammatory effects. This was manifested by the extremely low bacterial load in the PE gel group and the significant suppression of these key pro-inflammatory cytokines TNF-α and IL-6 (Fig 13). In contrast, HucMSC-Exo mainly promoted angiogenesis and collagen remodeling, evidenced by the significantly upregulated expression of pro-angiogenic markers CD31 and VEGF in the Exo gel and PE gel groups, which facilitates the delivery of oxygen and nutrients for tissue repair, alongside fundamental improvements in collagen maturity and alignment orderliness (Figs 8, 9, 11). VEGF-mediated angiogenesis is a well-recognized component of regenerative skin repair platforms [63], consistent with the pro-angiogenic effect of Exo observed in our study.
These results indicate that PBE and HucMSC-Exo exert complementary therapeutic functions: PBE mainly contributes to bacterial clearance and suppression of pro-inflammatory cytokines, whereas HucMSC-Exo mainly promotes angiogenesis and collagen remodeling. This combined antibacterial–anti-inflammatory–repair strategy helps disrupt the vicious pathological cycle of acne, although the current data support additive or complementary effects rather than formal pharmacological synergy. Accordingly, we use “complementary” or “combined” rather than “synergistic” throughout the manuscript unless referring specifically to this conceptual distinction.
The PE gel developed in this study shows potential as a candidate for further development. The hydrogel formulation is easy to apply, adheres well to skin folds, and should improve patient compliance [64,65]. More importantly, integrating both active components, PBE and Exo, into a stable hydrogel dressing holds potential for development as a “ready-to-use” acne management product [66], and may represent a promising non-antibiotic strategy requiring further validation, particularly for cases involving antibiotic resistance [67]. However, given the preclinical nature of this investigation, these findings should be interpreted with caution. Additional safety, pharmacokinetic, and efficacy studies in larger animal models, followed by controlled clinical trials, are required before clinical translation can be considered.
Despite the promising findings, this study has several limitations that warrant further investigation. First, although our DLS, zeta potential, and TEM analyses indicate that PBE and Exo coexist as distinct particle populations without evidence of stable complex formation, the precise nature of any weak or transient interactions between them—whether driven by electrostatic forces, hydrophobic interactions, or other mechanisms—remains to be further clarified. Surface plasmon resonance (SPR) or isothermal titration calorimetry (ITC) could be employed to characterize these binding affinities and kinetics [68]. Second, although HucMSC-Exo demonstrated potent pro-repair effects, the exact cargo (e.g., specific miRNAs like miR-21-5p or proteins like TGF-β1) responsible for promoting angiogenesis and collagen remodeling was not identified. Future work involving miRNA sequencing and proteomic analysis of the exosomes, followed by gain- and loss-of-function experiments in target cells, is essential to pinpoint the key active molecules [69,70]. In addition, particle concentration was not measured by nanoparticle tracking analysis (NTA) or comparable methods; BCA-based protein quantification reflects total protein content and does not provide particle number, which should be considered a limitation of the present exosome characterization. Third, our animal model, while well-established for acne research, does not fully recapitulate the chronic, multi-factorial and spontaneously occurring nature of human acne [71]. Rodents differ from humans in pilosebaceous anatomy, sebum composition, skin microbiome, inflammatory responses, and wound healing processes, which may influence treatment responses. In addition, the repeated intraperitoneal administration of C. acnes may have induced systemic immune responses that could influence local ear outcomes; although local parameters were the primary endpoint, this potential confounding effect cannot be fully excluded. Therefore, the present study should be regarded as a preclinical proof-of-concept investigation, and the therapeutic efficacy and safety observed here require further validation in more clinically relevant models and, ultimately, in controlled clinical trials. Advanced in vitro models, such as 3D tissue constructs, may provide more physiologically relevant platforms for future validation of exosome-loaded hydrogels before clinical translation [72]. Fourth, the in vitro experiments were performed in a single experimental run with three technical replicates per group. Although this design was sufficient for the exploratory and proof-of-concept analyses reported here, the lack of independent biological replicates may limit the generalizability of the in vitro results. Future studies should include at least three independent biological replicates to further validate the antibacterial, anti-biofilm, and release-related findings. Finally, although we performed preliminary in vivo safety evaluations, longer-term toxicity, biodegradation, and potential immunogenicity of the hydrogel remain to be investigated. In addition, the release profile of exosomes from the PE hydrogel and the structural integrity and biological activity of exosomes after release were not quantitatively assessed. Although the sustained release of PBE (Fig 4D) and the intact morphology of exosomes in the Exo + PBE mixture (S3 Fig C) provide indirect support, these data do not fully substitute for direct post-release characterization, and future studies should include quantitative exosome-release profiling and functional assessment.
5. Conclusions
In conclusion, we have successfully constructed an alginate based hydrogel, co loaded with PBE and HucMSC-Exo (PE gel), as a comprehensive treatment delivery system for acne vulgaris. PE gel operates through combined antibacterial, anti-inflammatory, and pro-repair effects, PBE ensures a clean microenvironment through bacterial elimination and anti-inflammatory effects, while HucMSC-Exo drives tissue regeneration by promoting angiogenesis and collagen remodeling. This collaborative strategy may help interrupt the pathological cycle of acne and provide a basis for developing non-antibiotic therapeutic options, pending further validation for acne management.
Supporting information
S1 Fig. High-performance liquid chromatography-mass spectrometry (HPLC-MS) analysis of PBE.
https://doi.org/10.1371/journal.pone.0363272.s001
(TIF)
S2 Fig. Bacterial viability assessed by DMAO/PI staining after treatment with Exo, PBE, or PE mixture.
https://doi.org/10.1371/journal.pone.0363272.s002
(TIF)
S3 Fig. Characterization of HucMSC-derived exosomes.
(A) Western blot analysis of exosomal markers (CD9, CD63, CD81, TSG101) and negative markers (Calnexin, Hsp70, GAPDH) in exosomes and HucMSC whole-cell lysates. (B) TEM image of isolated exosomes showing typical cup-shaped vesicular morphology. (C) TEM image of the Exo + PBE mixture showing intact exosome vesicles and separate PBE particles. The original, uncropped western blot images corresponding to (A) are provided in S1 File.
https://doi.org/10.1371/journal.pone.0363272.s003
(TIF)
S4 Fig. Standard curve for PBE concentration determination.
https://doi.org/10.1371/journal.pone.0363272.s004
(TIF)
S5 Fig. Histopathological evaluation of longitudinal sections of rat ear acne tissue by H&E and Masson’s trichrome staining.
https://doi.org/10.1371/journal.pone.0363272.s005
(TIF)
S6 Fig. Cross-sectional immunohistochemical staining for angiogenesis and inflammatory markers.
https://doi.org/10.1371/journal.pone.0363272.s006
(TIF)
S7 Fig. Serum levels of ALP, ALT, AST, CREA-S, TP II, and UREA in control and PE gel-treated rats.
https://doi.org/10.1371/journal.pone.0363272.s007
(TIF)
S8 Fig. H&E staining of major organs (heart, liver, spleen, lung, and kidney) from control and PE gel-treated rats.
https://doi.org/10.1371/journal.pone.0363272.s008
(TIF)
References
- 1. Moradi Tuchayi S, Makrantonaki E, Ganceviciene R, Dessinioti C, Feldman SR, Zouboulis CC. Acne vulgaris. Nat Rev Dis Primers. 2015;1:15029. pmid:27189872
- 2. Podwojniak A, Tan IJ, Sauer J, Neubauer Z, Rothenberg H, Ghani H, et al. Acne and the cutaneous microbiome: A systematic review of mechanisms and implications for treatments. J Eur Acad Dermatol Venereol. 2025;39(4):793–805. pmid:39269130
- 3. Zaenglein AL. Acne vulgaris. N Engl J Med. 2018;379(14):1343–52.
- 4. Drake L, Reyes-Hadsall S, Barbieri JS, Mostaghimi A. New Developments in Topical Acne Therapy. Am J Clin Dermatol. 2022;23(2):125–36. pmid:35041198
- 5. Cruz S, Vecerek N, Elbuluk N. Targeting Inflammation in Acne: Current Treatments and Future Prospects. Am J Clin Dermatol. 2023;24(5):681–94. pmid:37328614
- 6. Jin Z, Song Y, He L. A review of skin immune processes in acne. Front Immunol. 2023;14:1324930. pmid:38193084
- 7. Li X, Luo S, Chen X, Li S, Hao L, Yang D. Adipose-derived stem cells attenuate acne-related inflammation via suppression of NLRP3 inflammasome. Stem Cell Res Ther. 2022;13(1):334. pmid:35871079
- 8. Dreno B, Dekio I, Baldwin H, Demessant AL, Dagnelie M-A, Khammari A, et al. Acne microbiome: From phyla to phylotypes. J Eur Acad Dermatol Venereol. 2024;38(4):657–64. pmid:37777343
- 9. Do TH, Ma F, Andrade PR, Teles R, de Andrade Silva BJ, Hu C, et al. TREM2 macrophages induced by human lipids drive inflammation in acne lesions. Sci Immunol. 2022;7(73):eabo2787. pmid:35867799
- 10. Liu P, Liu X, Zhang L, Yan G, Zhang H, Xu D, et al. ALA-PDT augments intense inflammation in the treatment of acne vulgaris by COX2/TREM1 mediated M1 macrophage polarization. Biochem Pharmacol. 2023;208:115403. pmid:36592708
- 11. Feng Y, Li J, Mo X, Ju Q. Macrophages in acne vulgaris: mediating phagocytosis, inflammation, scar formation, and therapeutic implications. Front Immunol. 2024;15:1355455. pmid:38550588
- 12. Hong Y-K, Hwang D-Y, Yang C-C, Cheng SM, Chen P-C, Aala WJ, et al. Profibrotic Subsets of SPP1+ Macrophages and POSTN+ Fibroblasts Contribute to Fibrotic Scarring in Acne Keloidalis. J Invest Dermatol. 2024;144(7):1491-1504.e10. pmid:38218364
- 13. Clark AK, Saric S, Sivamani RK. Acne scars: how do we grade them?. Am J Clin Dermatol. 2018;19(2):139–44.
- 14. Reynolds RV, Yeung H, Cheng CE, Cook-Bolden F, Desai SR, Druby KM, et al. Guidelines of care for the management of acne vulgaris. J Am Acad Dermatol. 2024;90(5):1006.e1-1006.e30. pmid:38300170
- 15. Eichenfield DZ, Sprague J, Eichenfield LF. Management of Acne Vulgaris: A Review. JAMA. 2021;326(20):2055–67.
- 16. Walsh TR, Efthimiou J, Dréno B. Systematic review of antibiotic resistance in acne: an increasing topical and oral threat. Lancet Infect Dis. 2016;16(3):e23-33. pmid:26852728
- 17. Dessinioti C, Katsambas A. Antibiotics and Antimicrobial Resistance in Acne: Epidemiological Trends and Clinical Practice Considerations. Yale J Biol Med. 2022;95(4):429–43. pmid:36568833
- 18. Xu H, Li H. Acne, the Skin Microbiome, and Antibiotic Treatment. Am J Clin Dermatol. 2019;20(3):335–44. pmid:30632097
- 19. Layton AM, Ravenscroft J. Adolescent acne vulgaris: current and emerging treatments. Lancet Child Adolesc Health. 2023;7(2):136–44. pmid:36525984
- 20. Alexiades M, Kothare A, Goldberg D, Dover JS. Novel 1726 nm laser demonstrates durable therapeutic outcomes and tolerability for moderate-to-severe acne across skin types. J Am Acad Dermatol. 2023;89(4):703–10. pmid:37328000
- 21. Del Rosso J, Sugarman J, Green L, Lain T, Levy-Hacham O, Mizrahi R, et al. Efficacy and safety of microencapsulated benzoyl peroxide and microencapsulated tretinoin for the treatment of acne vulgaris: Results from two phase 3 double-blind, randomized, vehicle-controlled studies. J Am Acad Dermatol. 2023;89(4):719–27. pmid:37356627
- 22. Chen C-Y, Zhang Y. Berberine: An isoquinoline alkaloid targeting the oxidative stress and gut-brain axis in the models of depression. Eur J Med Chem. 2025;290:117475. pmid:40107207
- 23. Patel K, Patel DK. Biological Importance of Phellodendrine in Traditional and Modern Medicines: An Update on Therapeutic Potential in Medicine. Curr Drug Res Rev. 2024;16(2):209–18.
- 24. Liu H, Kong L, Cao D, Zhan X, Gao X, Sun H, et al. Efficacy and mechanism of the Ermiao San series of formulas for rheumatoid arthritis based on Chinmedomics strategy. Phytomedicine. 2024;132:155903. pmid:39047412
- 25. Sun L, Yu Q, Peng F, Sun C, Wang D, Pu L, et al. The antibacterial activity of berberine against Cutibacterium acnes: its therapeutic potential in inflammatory acne. Front Microbiol. 2024;14:1276383. pmid:38249466
- 26. Li W, Si Y, Wang Y, Chen J, Huo X, Xu P, et al. hUCMSC-derived exosomes protect against GVHD-induced endoplasmic reticulum stress in CD4+ T cells by targeting the miR-16-5p/ATF6/CHOP axis. Int Immunopharmacol. 2024;135:112315. pmid:38805908
- 27. Yang J, Xie Y, Xia Z, Ji S, Yang X, Yue D, et al. HucMSC-Exo Induced N2 Polarization of Neutrophils: Implications for Angiogenesis and Tissue Restoration in Wound Healing. Int J Nanomedicine. 2024;19:3555–75. pmid:38638364
- 28. Liu B, Chen L, Huang C, Zhang H, Zhou H, Chen Y, et al. A sprayable exosome-loaded hydrogel with controlled release and multifunctional synergistic effects for diabetic wound healing. Mater Today Bio. 2025;34:102159. pmid:40822934
- 29. Yu Y, Chen M, Guo Q, Shen L, Liu X, Pan J, et al. Human umbilical cord mesenchymal stem cell exosome-derived miR-874-3p targeting RIPK1/PGAM5 attenuates kidney tubular epithelial cell damage. Cell Mol Biol Lett. 2023;28(1):12. pmid:36750776
- 30. Xu L-M, Yu X-X, Zhang N, Chen Y-S. Exosomes from umbilical cord mesenchymal stromal cells promote the collagen production of fibroblasts from pelvic organ prolapse. World J Stem Cells. 2024;16(6):708–27. pmid:38948096
- 31. Chen J, Huo X, Qian M, Xue Q, He Y, Xu P, et al. Exosomes Derived From Human Mesenchymal Stem Cells Mitigate Follicular Interstitial Cell Ferroptosis via the miR-26a-5p/PTEN/GPX4 Axis in Rats with Chemotherapy-Induced Premature Ovarian Insufficiency. Int J Nanomedicine. 2025;20:10195–212. pmid:40873678
- 32. Determination of minimum inhibitory concentrations (MICs) of antibacterial agents by agar dilution. Clin Microbiol Infect. 2000;6(9):509–15.
- 33. Pitic Coţ D-E, Kiş A, Stroia C, Talpoş-Niculescu I-C, Popovici R-A, Ille C-E, et al. New Insights into the Antimicrobial and Wound-Healing Properties of Turmeric-Powder-Derived Curcuma longa Extracts for Oral-Health-Oriented Applications. Biomedicines. 2026;14(5):1078. pmid:42193402
- 34. Huang Y, Liu C, Fu Z, Li C, Wu Y, Jia Q, et al. The combination of RL-QN15 and OH-CATH30 promotes the repair of acne via the TLR2/NF-κB pathway. Eur J Pharmacol. 2025;989:177233. pmid:39740735
- 35. Jiang S, Li H, Zhang L, Mu W, Zhang Y, Chen T, et al. Generic Diagramming Platform (GDP): a comprehensive database of high-quality biomedical graphics. Nucleic Acids Res. 2025;53(D1):D1670–6. pmid:39470721
- 36. AbdRabou MA, Atwa A, Massoud D, Asran AM, Alrayes ZR, Otaibi AMA-, et al. Therapeutic Effect of Mesenchymal Stem Cells and Olive Leaf Extract Against Cytotoxicity Induced by Streptozotocin in Gestational Diabetes Rats. Cell Physiol Biochem. 2024;58(6):702–21. pmid:39720931
- 37. Kuo C-W, Chiu Y-F, Wu M-H, Li M-H, Wu C-N, Chen W-S, et al. Gelatin/Chitosan Bilayer Patches Loaded with Cortex Phellodendron amurense/Centella asiatica Extracts for Anti-Acne Application. Polymers (Basel). 2021;13(4):579. pmid:33671908
- 38. Oge’ LK, Broussard A, Marshall MD. Acne Vulgaris: Diagnosis and Treatment. Am Fam Physician. 2019;100(8):475–84. pmid:31613567
- 39. Kim K-U, Kim W-H, Jeong CH, Yi DY, Min H. More than Nutrition: Therapeutic Potential of Breast Milk-Derived Exosomes in Cancer. Int J Mol Sci. 2020;21(19):7327. pmid:33023062
- 40. Kim M-J, Zarnowski R, Jones R, Nett JE, Andes D. Vesicle inhibition reduces Candida biofilm resistance. Antimicrob Agents Chemother. 2025;69(5):e0004525. pmid:40135881
- 41. Wang X, Liu Y, Dong X, Duan T, Wang C, Wang L, et al. peu-MIR2916-p3-enriched garlic exosomes ameliorate murine colitis by reshaping gut microbiota, especially by boosting the anti-colitic Bacteroides thetaiotaomicron. Pharmacol Res. 2024;200:107071. pmid:38218354
- 42. Qian Z, Bai Y, Zhou J, Li L, Na J, Fan Y, et al. A moisturizing chitosan-silk fibroin dressing with silver nanoparticles-adsorbed exosomes for repairing infected wounds. J Mater Chem B. 2020;8(32):7197–212. pmid:32633312
- 43. Shang S, Zheng F, Tan W, Xing Z, Chen S, Peng F, et al. Piezoelectric Biomaterial with Advanced Design for Tissue Infection Repair. Adv Sci (Weinh). 2025;12(10):e2413105. pmid:39887897
- 44. Tan S, Liu Z, Cong M, Zhong X, Mao Y, Fan M, et al. Dandelion-derived vesicles-laden hydrogel dressings capable of neutralizing Staphylococcus aureus exotoxins for the care of invasive wounds. J Control Release. 2024;368:355–71. pmid:38432468
- 45. Liu H, Wei X, Peng H, Yang Y, Hu Z, Rao Y, et al. LysSYL-Loaded pH-Switchable Self-Assembling Peptide Hydrogels Promote Methicillin-Resistant Staphylococcus Aureus Elimination and Wound Healing. Adv Mater. 2024;36(52):e2412154. pmid:39548922
- 46. Hernández-González AC, Téllez-Jurado L, Rodríguez-Lorenzo LM. Alginate hydrogels for bone tissue engineering, from injectables to bioprinting: A review. Carbohydr Polym. 2020;229:115514. pmid:31826429
- 47. Ren Y, Wang Q, Xu W, Yang M, Guo W, He S, et al. Alginate-based hydrogels mediated biomedical applications: A review. Int J Biol Macromol. 2024;279(Pt 1):135019. pmid:39182869
- 48. Wang T, Yi W, Zhang Y, Wu H, Fan H, Zhao J, et al. Sodium alginate hydrogel containing platelet-rich plasma for wound healing. Colloids Surf B Biointerfaces. 2023;222:113096.
- 49. Lin Y, Guo Q, Liu Q, Wang W, Lv A, Zhang L, et al. An injectable responsive exosome-releasing hydrogel based on sodium alginate restores motor and bladder function by alleviating the injury microenvironment and facilitating distal nerve repair. Int J Biol Macromol. 2025;304(Pt 1):140819. pmid:39929458
- 50. Zou Z, Zhang B, Nie X, Cheng Y, Hu Z, Liao M, et al. A sodium alginate-based sustained-release IPN hydrogel and its applications. RSC Adv. 2020;10(65):39722–30. pmid:35515393
- 51. Song Y, You Y, Xu X, Lu J, Huang X, Zhang J, et al. Adipose-Derived Mesenchymal Stem Cell-Derived Exosomes Biopotentiated Extracellular Matrix Hydrogels Accelerate Diabetic Wound Healing and Skin Regeneration. Adv Sci (Weinh). 2023;10(30):e2304023. pmid:37712174
- 52. Clayton RW, Göbel K, Niessen CM, Paus R, van Steensel MAM, Lim X. Homeostasis of the sebaceous gland and mechanisms of acne pathogenesis. Br J Dermatol. 2019;181(4):677–90. pmid:31056753
- 53. O’Neill AM, Liggins MC, Seidman JS, Do TH, Li F, Cavagnero KJ, et al. Antimicrobial production by perifollicular dermal preadipocytes is essential to the pathophysiology of acne. Sci Transl Med. 2022;14(632):eabh1478. pmid:35171653
- 54. Williams HC, Dellavalle RP, Garner S. Acne vulgaris. Lancet. 2012;379(9813):361–72. pmid:21880356
- 55. Kutlu Ö, Karadağ AS, Wollina U. Adult acne versus adolescent acne: a narrative review with a focus on epidemiology to treatment. An Bras Dermatol. 2023;98(1):75–83. pmid:36253244
- 56. Moon J, Yoon JY, Yang JH, Kwon HH, Min S, Suh DH. Atrophic acne scar: a process from altered metabolism of elastic fibres and collagen fibres based on transforming growth factor-β1 signalling. Br J Dermatol. 2019;181(6):1226–37. pmid:30822364
- 57. Kyriakides TR, Wulsin D, Skokos EA, Fleckman P, Pirrone A, Shipley JM, et al. Mice that lack matrix metalloproteinase-9 display delayed wound healing associated with delayed reepithelization and disordered collagen fibrillogenesis. Matrix Biol. 2009;28(2):65–73. pmid:19379668
- 58. Chen X, Min S, Chen C, Lin X, Wang D, Jiang G. Influence of RETN, IL-1, and IL-6 gene polymorphisms on the risk of acne vulgaris in the Chinese population. J Cosmet Dermatol. 2022;21(10):4965–73. pmid:35279931
- 59. Dréno B. What is new in the pathophysiology of acne, an overview. J Eur Acad Dermatol Venereol. 2017;31 Suppl 5:8–12. pmid:28805938
- 60. Asadi K, Azarpira N, Heidari R, Hamidi M, Yousefzadeh-Chabok S, Nemati MM, et al. Trinitroglycerin-loaded chitosan nanogels accelerate angiogenesis in wound healing process. Int J Biol Macromol. 2024;278(Pt 3):134937. pmid:39179074
- 61. Du E, Li X, He S, Li X, He S. The critical role of the interplays of EphrinB2/EphB4 and VEGF in the induction of angiogenesis. Mol Biol Rep. 2020;47(6):4681–90. pmid:32488576
- 62. Peña OA, Martin P. Cellular and molecular mechanisms of skin wound healing. Nat Rev Mol Cell Biol. 2024;25(8):599–616. pmid:38528155
- 63. El-Fakharany EM, El-Sayed MH, Ali HM, Omar SM, Sayed AM, Atwa A, et al. Ameliorative efficacy of Syzygium aromaticum ethanolic extract and Silymarin® upon carbon tetrachloride-induced liver fibrosis: antioxidant, antibacterial and anti-inflammatory activities. Med Oncol. 2025;43(1):37. pmid:41364285
- 64. Geng X, Qi Y, Liu X, Shi Y, Li H, Zhao L. A multifunctional antibacterial and self-healing hydrogel laden with bone marrow mesenchymal stem cell-derived exosomes for accelerating diabetic wound healing. Biomater Adv. 2022;133:112613. pmid:35527135
- 65. Rao H, Tan JBL. Polysaccharide-based hydrogels for atopic dermatitis management: A review. Carbohydr Polym. 2025;349(Pt B):122966. pmid:39638516
- 66. Xie X, Ao X, Xu R, Lv H, Tan S, Wu J, et al. Injectable, stable, and biodegradable hydrogel with platelet-rich plasma induced by l-serine and sodium alginate for effective treatment of intrauterine adhesions. Int J Biol Macromol. 2024;270(Pt 1):132363. pmid:38754675
- 67. Sardana K, Gupta T, Garg VK, Ghunawat S. Antibiotic resistance to Propionobacterium acnes: worldwide scenario, diagnosis and management. Expert Rev Anti Infect Ther. 2015;13(7):883–96. pmid:26025191
- 68. Wang P, Jia X, Lu B, Huang H, Liu J, Liu X, et al. Erianin suppresses constitutive activation of MAPK signaling pathway by inhibition of CRAF and MEK1/2. Signal Transduct Target Ther. 2023;8(1):96. pmid:36872366
- 69. Wang J, Wu H, Peng Y, Zhao Y, Qin Y, Zhang Y, et al. Hypoxia adipose stem cell-derived exosomes promote high-quality healing of diabetic wound involves activation of PI3K/Akt pathways. J Nanobiotechnology. 2021;19(1):202. pmid:34233694
- 70. Nakaoka A, Salah M, Ruixian Z, Nakahana M, Shimizu Y, Qin Q, et al. Radiation suppresses liver metastases via exosome secretion in pancreatic cancer cells. Sci Rep. 2025;15(1):23712. pmid:40610637
- 71. Kanwar IL, Haider T, Kumari A, Dubey S, Jain P, Soni V. Models for acne: A comprehensive study. Drug Discov Ther. 2018;12(6):329–40. pmid:30674767
- 72. Atwa A, Wadan A-HS, Hallool SI, Shati AA, Elbehairi SEI, Alfaifi MY, et al. Organoid Models of the Oral Mucosa for Precision Medicine. Tissue Eng Part B Rev. 2026;19373368261420588. pmid:41761920

관련 뉴스