DMEM/F12 and K-SFM media, 1% (v/v) penicillin-streptomycin, 10% (v/v) fetal bovine serum (FBS), serum extracts, epidermal growth factor (EGF), and pre-stained protein MW marker 26616 were purchased from Thermo Fisher Scientific (Waltham, MA). Human IL-6 and IL-8 ELISA kits were purchased from Ray Biotech (Norcross, GA). Cell Titer^® 96 AQueous One Solution Reagent was purchased from Promega (Madison, WI). Dichlorodihydrofluorescin diacetate (DCFH-DA) was obtained from ABP Biosciences (Rockville, MD). Polyvinylidene fluoride membrane was acquired from Millipore (Billerica, MA). Anti-Caspase 3 (ab13847), anti-Caspase 8 (ab25901), anti-Caspase 9 (ab286147), anti-Bax (ab32503), anti-Bad (ab32445), anti-Bcl-2 (ab32124), anti-NF-κB-p65 (ab16502), PI3K/AKT signalling pathway panel (ab283852) and anti-β-actin (ab8227) were purchased from Abcam (Cambridge, United Kingdom). PE Annexin V Apoptosis Detection Kit I was obtained from BD Biosciences (San Diego, CA). IRDye 800 conjugated with affinity purified anti-Rabbit IgG antibody was purchased from Rockland (Philadelphia, PA).

The human primary keratinocytes (HSK) and dermal fibroblasts (HSF) were purchased from National Intrastructure of Cell Line Resource (NICR) in Beijing, China. Human primary dermal fibroblasts were cultured as a monolayer in DMEM/F12 medium supplemented with 1% (v/v) penicillin-streptomycin (100 μg/mL penicillin, 100 μg/mL streptomycin), and 10% (v/v) FBS. Human primary keratinocytes were cultured in K-SFM medium supplemented with 1% (v/v) penicillin-streptomycin, 500 μg/mL serum extracts, and 5 ng/mL EGF. Cells were grown in tissue culture flasks of size 25 cm² at 37°C under an atmosphere containing 5% CO₂ in a humidified incubator and passaged every 5–7 days.

The cocultured skin cell model was constructed as follows: Human keratinocytes were inoculated at a density of 5 × 10⁴ cells/well in a 6-well plate and cultured at 37°C under an atmosphere of 5% CO₂ in a humidified incubator. After the keratinocytes adhered, 1.25 × 10⁴ dermal fibroblasts per well were subsequently seeded into the same wells. Similarly, human keratinocytes were seeded at a density of 2 × 10³ cells/well in 96-well plates. After cell adhesion, 0.5 × 10³ dermal fibroblasts were seeded into the same wells. Cell culture was performed using a mixed medium of K-SFM/DMEM/F12 at a ratio of 3:1:1.

The proliferation of cocultured skin cell model influenced by PAVA was assessed using the 3-(4, 5-dimethylthiazol-2-yl)-5-(3-carboxy-methoxyphenyl)-2-(4-sulphophenyl)-2H-tetrazolium (MTS) assay. After coculturing in a 96-well plate for 24 h, the cells were washed with PBS (0.01 M, pH 7.4) to remove unattached cells and treated with various concentrations of PAVA (12.5 μM, 50 μM, 100 μM, and 200 μM) for different durations (5 h, 24 h, and 48 h). Subsequently, Cell Titer^® 96 AQueous One Solution Reagent (20 μL) was added to each well at at time points of interest (0 h, 5 h, 24 h, and 48 h), followed by incubation for an additional hour. The absorbance at an optical density of 490 nm was measured using a multi-detection microplate reader (BIO-TEK Instruments Inc., Winooski, VT). Mean values and standard errors of mean (SEM) were calculated for each concentration. Cell proliferation and inhibition rates were determined using the formula “Cell proliferation = O D P A V A − O D B l a n k / O D C o n t r o l − O D B l a n k ″ and “Cell inhibition = ( 1 − Cell proliferation ) × 100%”. The half maximal inhibitory concentration (IC₅₀) was analyzed using GraphPad Prism software version 8.0.1.

The intracellular level of reactive oxygen species (ROS) in cocultured skin cell model stimulated by PAVA was detected using the dichlorofluorescein (DCF) method. After being cocultured in a 6-well plate for 24 h, the cells were treated with PAVA at concentrations of 100 μM for durations of 5 h and 24 h. Subsequently, they were washed twice with PBS (0.01 M, pH7.4), incubated with 15 μM DCFH-DA for 30 min at 37°C, and then washed three times with HBSS (Thermo Fisher Scientific, United States) before being incubated with DMFM/F12 medium at 37°C for 10 min. Finally, the fluorescent signal was measured using a Cytation cell imaging multi-mode reader (BIO-TEK Instruments Inc., Winooski, VT) with an excitation wavelength set to 485 nm and an emission wavelength set to 590 nm.

The type of cell death was investigated using flow cytometry. HSK and HSF were co-cultured in a 6-well plate and treated with PAVA for 5 h and 24 h. After collection and rinsing with PBS (0.01 M, pH 7.4), the cells were resuspended with 100 μL binding buffer (10 mM HEPES pH 7.4, 140 mM NaCl, and 2.5 mM CaCl₂) containing 5 μL 7-ADD and 5 μL Annexin V-PE, followed by incubation in the dark at room temperature for 15 min. Subsequently, 400 μL of binding buffer was added to the sample tubes, which were then analyzed using FACS Calibur (Becton Dicknson, Franklin Lakes, NJ). At least 10,000 events were measured for each sample in both the FL1 channel (orange-red fluorescence: 575/25 nm) and the FL2 channel (far-red fluorescence: 670/30 nm). Data analysis was performed using FlowJo™ software (version 10.0.6, Tree Star, Ashland, OR).

The apoptosis-related proteins were assayed using Western blot according to the established protocol. Briefly, cocultured cells treated with 100 μM PAVA for 5 h, 24 h, and 48 h were harvested. Subsequently, the proteins were resolved on a 15% SDS-PAGE gel and transferred onto polyvinylidene fluoride membranes. These membranes were then incubated overnight at 4°C with polyclonal rabbit antibodies against anti-Caspase 3, anti-Caspase 8, anti-Caspase 9, anti-NF-κB-p65, anti-Bax, anti-Bad, anti-Bcl-2 and anti-β-actin (Abcam, Cambridge, United Kingdom). Afterward, the membranes were incubated with IRDye800-conjugated affinity purified anti-Rabbit IgG antibody for 1.5 h at room temperature. Finally, the membranes were scanned using the Odyssey Infrared Imaging System (LI-COR Bioscience, Lincoln, NE).

The secretion levels of interleukins (IL-6 and IL-8) from cocultured skin cell model were measured using the enzyme linked immunosorbent assay (ELISA). Following the standard operating procedures of Human IL-6 and IL-8 ELISA kit, supernatants from 4 to 5 × 10⁵ cocultured skin cell model treated with 100 μM PAVA for 5 h, 24 h and 48 h were collected for ELISA analysis to determine the secretion of IL-6 and IL-8. Absorbance was measured at an optical density of 450 nm using the multi-detection microplate reader (BIO-TEK Instruments Inc., Winooski, VT).

Statistical analysis was conducted using GraphPad Prism 8.0.1 software. Differences between groups were analyzed using One-Way ANOVA followed by Dunnett’s post hoc test for multiple comparisons to compare each treatment group with the control group. For comparisons between two groups, an unpaired t-test was used. All in vitro experiments were performed in triplicate and repeated three times, with the results presented as the Mean ± SEM. P value <0.05 is considered statistically significant. *P <0.05; **P <0.01; ***P <0.001; ****P <0.0001.

To investigate the impact of PAVA treatment on cell proliferation in a constructed cocultured skin cell model, this study applied different concentrations of PAVA (50 μM, 100 μM, and 200 μM) to this model for varying durations (5h, 24h, and 48 h). While no significant effect on cell proliferation was observed with 50 μM PAVA compared to the control group, both 100 μM and 200 μM PAVA exhibited varying degrees of impact on cell proliferation, with the most pronounced effect observed at a concentration of 200 μM (Figure 1A). The MTS assay results (Figure 1A) demonstrated that a dose-dependent inhibitory effect on cell proliferation in the cocultured skin cell model after a treatment duration of 24 h with different concentrations of PAVA ranging from 12.5 μM to 200 μM. Compared to the control group, inhibition rates ranged from 0.021% ± 0.036% (12.5 μM) to 0.85% ± 0.014% (200 μM) for cocultured skin cell model which were treated with various concentrations of PAVA between 12.5–200 μM (Figure 1B). Furthermore, the IC₅₀ of PAVA against cocultured skin cell model was determined to be 101.1 μM ± 0.84 μM (approximately 29.6 mg/L).

The morphological impact of PAVA (100 μM) on cells in cocultured skin cell model was observed using inverted microscopy. In the control group, both cell types exhibited morphology consistent with their characteristic features (fibroblasts: long spindle-shaped, full and stretched cells with clear cell outline; keratinocytes: cobblestone-like appearance with defined borders), with a uniform distribution and clear cell boundaries, indicating a healthy state (Figure 2). Treatment with 100 μM PAVA for 5 h induced morphological changes, including slight cell swelling, less distinct cell borders, and minimal alterations in intercellular spacing, although the overall integrity of the cells remained essentially unchanged. After 24 h of treatment with 100 μM PAVA, more pronounced morphological changes were observed, including irregular cell shapes, loss of clear cell contours in many cells, a reduction in cell size with rounding and loss of characteristic features in numerous cells, and clear signs of apoptosis (Figure 2).

The generation of ROS in cocultured skin cell model was evaluated using fluorescence microscopy under various conditions (Figure 3A). Minimal green fluorescence was observed in the control group exhibited, indicating low levels of ROS and negligible oxidative stress. Treatment with 100 μM PAVA for 5 h resulted in an increase in green fluorescence spots, suggesting that PAVA treatment induced oxidative stress and enhanced intracellular ROS generation. A low-magnification view (4×) of the visual field is provided in supplementary materials Supplementary Figure S1 to further illustrate the observed effects. After 24 h of exposure to the same concentration of PAVA, the intensity of green fluorescence significantly intensified compared to the 5-h treatment group, demonstrating cumulative effects of oxidative stress over time (Figure 3B).

The level of apoptosis in cells following PAVA treatment was quantitatively analyzed using flow cytometry results (Figure 4). In the control group, minimal apoptosis was observed, with the majority of cells (93.40%) located in the live cell quadrant (Q4). After treatment with 100 µM PAVA for 5 h, a slight increase in early apoptotic cells (Q3: 6.31%) was observed. By 24 h, both early apoptotic cells (Q3: 7.95%) and late apoptotic cells (Q2: 5.03%) exhibited an increase. Compared to the control group, the overall apoptotic rate in cocultured skin cell model treated with PAVA for 5 and 24 h was notably elevated, indicating that the apoptotic effects accumulate over time. Specific cell gating strategies are provided in the supplementary materials (Supplementary Figure S2).

Cell apoptosis is a complex process involving intricate interplay among multiple pathways. Combining results from oxidative stress and flow cytometry analysis, this study employed Western blot protein immunoblotting to examine apoptosis-related protein molecules. The results revealed significant upregulation of activated Caspase-3 and Caspase-9 expression levels at 5 h, 24 h, and 48 h following treatment with 100 μM PAVA. Moreover, an increase in the levels of pro-apoptotic proteins Bad and Bax indicated activation of the mitochondrial pathway of apoptosis (Figure 5). With prolonged exposure to PAVA, the expression of Bad predominantly decreased; however, the expression of pro-apoptotic protein Bax remained significantly elevated along with a relative increase in the Bax/Bcl-2 ratio. Furthermore, after 5 h treatment with 100 μM PAVA, there was a notable increase in Caspase-8 expression levels (Figure 6), suggesting that PAVA may also induce apoptosis through the Caspase-8-dependent pathway, another crucial apoptotic signaling molecule.

To gain a comprehensive understanding of cellular responses under PAVA stimulation, we also assessed key molecules in the PI3K/AKT signaling pathway and NF-κB signaling pathway, which are crucial for cell survival pathways. The results revealed a significant increase in the expression levels of phosphorylated AKT (p-AKT), phosphorylated PI3K (p-PI3K), and NF-κB after 5 h of treatment with 100 μM PAVA. However, as the duration of treatment increased, the levels of p-AKT, p-PI3K, and NF-κB decreased (Figure 6). These findings suggest that early PAVA treatment (5 h) initiated cell apoptosis and pro-inflammatory responses through activation of apoptotic factors and NF-κB signaling. By 24 h, the mitochondrial pathway became the predominant apoptotic pathway due to upregulation of Caspase-3/9 and Bax. Concurrently, pro-inflammatory responses persisted, although there was a decline in NF-κB expression possibly due to sustained stimulation leading to negative feedback regulation. In the late stage of PAVA treatment (48 h), apoptotic signals persisted, with cell apoptosis becoming the dominant response. Meanwhile, NF-κB signaling further decreased, indicating an impaired ability of cells to sustain survival mechanisms.

Given the pronounced upregulation of NF-κB observed in early stages of PAVA treatment, it is postulated that PAVA-induced cellular apoptosis may ultimately be associated with sustained inflammatory responses. In this study, sandwich ELISA was employed to detect secreted pro-inflammatory cytokines (IL-6 and IL-8). The results revealed a significant upregulation in the secretion of IL-6 (Figure 7A) and IL-8 (Figure 7B) following PAVA treatment, suggesting the potential capacity of PAVA to stimulate the production and release of pro-inflammatory cytokines by cells.

Considering all experimental findings, it is evident that the effects of PAVA treatment on cells exhibit a time-dependent nature. Initially, PAVA triggers apoptotic signals and pro-inflammatory responses through multiple pathways. Subsequently, the mitochondrial pathway gradually emerges as the predominant apoptotic pathway, while survival signaling pathways are progressively attenuated.