Reactive oxygen species-mediated epidermal stress in human keratinocytes under dual exposure to blue light and heated tobacco product

Abstract

Introduction

Heated tobacco products (HTPs) are increasingly used among young people and adolescents, while their long-term toxicological effects remain insufficiently characterized. In parallel, frequent use of electronic devices has led to prolonged exposure to blue light (BL), a visible radiation that penetrates deeply into the skin and may exert phototoxic effects. Although the individual impact of these exposures has been studied, their combined effects on human skin are poorly understood.

Methods

In this study, we investigated the dual toxicity of HTP-derived total particulate matter (TPM) and BL irradiation (4.32 and 17.28 J/cm², respectively) in HaCaT human keratinocyte cell line. Cell viability via MTT test, intracellular reactive oxygen species (ROS), inflammatory markers, and matrix remodeling factors via ELISA, and autophagosome formation via western blotting were assessed following single and combined exposures.

Results

Our findings demonstrated that BL and TPM co-exposure significantly reduced cell viability in a time-dependent manner and augmented intracellular ROS production. Co-exposure also up-regulated MMP-1 and IL-6 levels while increasing LC3β-II expression in human keratinocytes, suggesting the implication of oxidative stress, inflammatory signaling, and autophagy in the observed dual cytotoxicity.

Discussion

These findings indicate that concurrent exposure to BL and HTPs may exacerbate epidermal damage beyond the effects of each stressor alone. The results highlight potential health concerns associated with modern lifestyle habits, particularly regarding skin toxicity, and support the need for further toxicological evaluation of combined environmental exposures.

Graphical Abstract

Introduction

Exposure to ultraviolet (UV) rays can cause oxidative damage, inflammation and pigment changes in the skin and is closely associated with the increasing incidence of skin cancer. ¹ Blue light (BL) is second only to UV light in terms of energy and is reported to penetrate the skin more than UV light. It can be emitted from any electronic device with an LED screen, such as smartphones, tablets, and computers, and the intensity of BL typically increases in the following order: mobile phone, computer or tablet, and television. ² BL exposure has become increasingly common compared to UV exposure, largely due to the widespread use of electronic devices, which have become an integral part of our daily lives. BL in our environment, emitted particularly from the sun and nowadays from artificial sources, may lead to detrimental effects on health. ³ Despite its widespread exposure, the phototoxic ⁴ potential of BL on the skin remains under-investigated. Since BL has high energy, it may accelerate photochemical reactions in several target organs such as the eye and skin^3–5 in addition to its circadian rhythm-disruptive potential.^6–9 Furthermore, current findings are associated with health problems such as decreased physical activity and loss of visual function in individuals, decreased sleep quality, and decreased crystalline lens transparency, focal length, and pupil diameter in infants due to the frequency and intensity of BL exposure ⁷ ; decreased cognitive capacity due to decreased sleep duration, especially in adolescents and young adults, ¹⁰ increased obesity, and susceptibility to depression.^6,8 In addition to all these target organs mentioned, the skin is one of the largest and most important target organs for BL exposure. Current studies frequently focus on skin aging and UV exposure, whereas research on BL exposure and its potential photobiological and phototoxic effects remains relatively limited. It has been shown that BL caused an increase in oxidative stress in human skin fibroblasts ¹¹ ; increased mitochondrial and lysosomal damage in human keratinocytes by stimulating photoaging, and dose-dependent increased cytotoxic response. ¹² On the other hand, some studies have demonstrated that specifically, the BL spectrum (400–500 nm) can generate oxidative stress through production of reactive oxygen species (ROS) and also stimulation of the MMP production. Given the limited research available, BL exposure could pose a toxicological risk to young individuals, who are increasingly using electronic devices, especially following the surge in online activities after the COVID-19 pandemic.

In addition to the exposure to BL, new-generation tobacco product (NGTP) exposure is remarkably increasing over the last decade among the young population. Most popularly, e-cigarettes and heated tobacco products (HTPs) are favorable among these populations due to marketing strategy. Briefly, HTPs can be defined as a new smoking system through a battery-powered tobacco stick that does not burn tobacco. ¹³ Due to the use of these devices, some undesirable effects of tobacco can also be seen on the skin. It is known that tobacco use is associated with many chronic dermatosis cases. Since nicotinic subtypes of acetylcholine receptors, including the α7 receptor, are expressed especially in skin cells, it is also thought that nicotine is the main component responsible for the undesirable effects on the skin. ¹⁴ Studies suggest that traditional cigarette smoke, and its residues as well, ¹⁵ plays a crucial role in skin aging by increasing matrix metalloproteinase (MMP) activation in the skin and mediating inflammatory response and oxidative damage, and affects the biophysical parameters in the skin, especially the thickness and density of the dermis, epidermis and nasolabial folds.^16,17 However, due to its different chemical dynamics and aromatic content, the effects of HTP exposure, mostly its total particulate matter (TPM) fraction, on the skin have not been fully elucidated. Despite the increasing use of HTPs, most toxicological assessments have focused on the gas-phase emissions or systemic effects, leaving the cutaneous consequences of TPM largely underexplored. Hence, BL and HTP-derived TPM represent relevant co-exposures with the potential to exert cumulative skin toxicity, particularly in young individuals. However, current literature lacks sufficient studies addressing the chronic nature of both BL and TPM exposures in daily life on skin integrity. It has been reported that simultaneous traditional cigarette smoke and UVA1 light exposure increased TGF-β release and MMP-1 activity in in vitro fibroblast and keratinocyte cells and significantly decreased collagen III-IV protein expression levels, causing premature skin aging. ¹⁸ In this context, both BL and HTP-derived TPM can act as environmental stressors capable of inducing molecular responses such as ROS production, inflammatory signaling, and matrix degradation. Yet, the mechanisms underlying their potential effects on the epidermis remain poorly understood. Importantly, HaCaT keratinocytes serve as a validated model for exploring skin toxicity ¹⁵ and photoaging, allowing the elucidation of pathways involving oxidative damage, MMP-1 expression, autophagy regulation, and inflammatory mediators such as IL-6. This study aimed to characterize, for the first time, the toxicological profile of combined BL and HTP exposures in human keratinocytes, focusing on oxidative stress, inflammation, autophagy, and matrix remodeling within the framework of New Approach Methodologies (NAMs).

Material and methods

Extraction of HTP-derived particulate matter (TPM)

The extraction of HTP was performed according to the previous methods.^13,19 Briefly, one non-flavored heat stick was heated separately using its own heating device according to the ISO-Health Canada Intense (HCI) Puff Regimen protocol, which includes a puff volume of 55 mL, a puff interval of 2 s and a puff cycle of 30 s (ISO, 2018). For the isolation of TPM, the apparatus depicted in Figure 1 was used, in which a piece of Whatman filter paper was placed at the tip of the syringe. This assembly provided the puff control and the collection of TPM on previously tared Whatman filter paper isolated as described before. ²⁰ The prepared TPM was then extracted by incubating the final concentration of 12 mg/mL in dimethylsulfoxide (DMSO) by the modification of the method of the previous study ²¹ at 37°C for 24 h with shaking and centrifugation. The prepared samples were labeled as TPM and stored at −80°C for up to 1 month to prevent stability loss before use in in vitro studies. Moreover, TPM yields per heat stick, solvent use for the extraction, and final DMSO concentrations for each independent extraction were given in Supplemental Table S1.

Figure 1.

Experimental setup for BL exposure of HaCaT cells. An opaque acrylic enclosure was used to minimize ambient light interference and ensure homogeneous BL delivery to the culture plates, simulating a controlled phototoxic environment. (Created in BioRender. Reis, R. (2026) https://BioRender.com/euuo3rw).

Nicotine content of TPM

Quantitative analysis was performed using an LC-20 series high-performance liquid chromatography (HPLC) system (Shimadzu, Japan) equipped with a diode array detector (DAD) set at 260 nm. Chromatographic separation was achieved on a Chromolith® High Resolution RP18e monolithic column (100 × 4.6 mm) maintained at 25°C. The mobile phase consisted of 10 mM phosphate buffer (pH 7) containing 0.3% triethylamine and acetonitrile in a ratio of 87.5:12.5 (v/v), delivered isocratically at a flow rate of 1.0 mL/min. The injection volume was 10 µL. The extract samples were diluted at a ratio of 1:10 with the corresponding solvent and injected directly into the system without further pretreatment. Quantification was carried out by comparing the peak areas with those of external standards, and the results were expressed accordingly.

Cell culture and viability assay

For in vitro studies, HaCaT human keratinocyte cell line obtained from Cell Line Service (CLS), Eppelheim, Germany, Cat# 300493) was cultured in Dulbecco’s Modified Culture Medium (DMEM) containing 10% fetal bovine serum (FBS; w/v) (Invitrogen, Waltham, MA, USA) and 1% Penicillin and Streptomycin (Penstrep, w/v) as stated previously. Cells were grown at 37°C and 5% CO₂ and incubated until becoming 80% confluent. ¹⁵ HaCaT cell line includes spontaneously immortalized human keratinocytes which are the primary cells of the human epidermis offering a relatively indefinite lifespan and also elimination of limitations such as the need for growth conditions, culture diversity, and donor-related variety and, in addition to providing epidermal barrier function, are involved in the inflammatory response of the skin and the re-epithelialization step, wound repair and partial in the aging processes. ²² Hence, this study did not involve human participants or live animals, and all experiments were conducted using an established immortalised human keratinocyte cell line without requirement of an ethics committee approval. To determine the dose and time-response relationship of TPM and TPM + BL exposure, relative cell viability was assessed through (3-(4,5-Dimethylthiazol-2-yl)-2,5-Diphenyltetrazolium Bromide) (MTT) assay. Briefly, MTT assay used to detect cytotoxicity through mitochondrial dehydrogenase activity, and a widespread technique due to its sensitivity, reproducibility. After exposure of seeded cells (10⁵ cells/mL) to the TPM and BL alone and dual exposures, purple-colored formazan was quantified spectrophotometrically at 570 nm in independent triplicates for each group, according to the previously described method. ¹⁵ All experimental assays were conducted on healthy, exponentially growing cells between passages #16-26.

Simulation of BL exposure

HaCaT cells were exposed to BL using a portable light emitting diode (LED) irradiator (λmax = 470 nm) in phenol red-free medium with and without TPM extract, modifying the previous method. ²³ The light dose was determined as 4.32 J/cm² (for 0.5 h); 17.28 J/cm² (for 2 h), and light intensity was defined as 2.4 mW/cm² with 30% dim, fixed during the study. BL exposure was carried out for 0.5 h for modeling acute/short-term exposure, and 2 h for repeated/long-term exposure in a closed mechanism at 37°C to determine the dose-dependent effect. The TPM + BL exposure setup was depicted in Figure 1.

Autophagosome formation

Autophagy is a natural and tightly regulated physiological mechanism responsible for the degradation of damaged organelles and the recycling of cytoplasmic materials involved in various pathophysiological processes. Recent studies suggested that this mechanism might play a role in photoaging and epidermal differentiation. ¹ Hence, the protein expression of the microtubule-associated protein 1A/1B-light chain 3-phosphatidylethanolamine conjugate (LC3β-II/I), which is involved in autophagosome formation, was evaluated. Briefly, isolated exposed cell pellets, as described in our previous study, were lysed in RIPA lysis buffer (10 mM Tris-HCl, pH 8, 0.32 mM sucrose, 5 mM EDTA, 2 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, and 1% Triton X-100), defined as whole-cell lysates. The isolated proteins were loaded onto a 14% sodium dodecyl sulfate–polyacrylamide gel (SDS-PAGE) for electrophoresis. Separated proteins were then transferred onto polyvinylidene difluoride (PVDF) membranes and blocked for 1 h at room temperature in 5% non-fat dry milk diluted in TBST (10 mM Tris-HCl, pH 8.0, 137 mM NaCl, 0.05% Tween-20). Subsequently, membranes were incubated overnight at 4°C with primary antibodies against LC3β (1:3000; Boster, USA) in TBST containing 5% non-fat milk. The next day, membranes were incubated with horseradish peroxidase-conjugated secondary antibodies (1:5000 in TBST; Santa Cruz Biotechnology, CA, USA). Protein quantification was performed using the Pierce bicinchoninic acid (BCA) assay kit (Thermo Scientific™, 23227), with 20 µg of protein loaded per sample. ¹⁹ All experiments were conducted in triplicate, and band intensities were analyzed as relative density values normalized to β-actin expression.

BL and TPM induced-ROS

The intracellular total ROS formation was assessed based on our previous report. ¹⁹ Briefly, a fluorescent probe sensitive to the detection of hydroxyl, peroxyl, and other ROS, 2′,7′-dichlorofluorescin diacetate (DCFDA), was used to determine the change in ROS levels in HaCaT cells exposed to the conditions described in viability assay and simulation of BL exposure. The culture medium was removed following the exposure, and cells were washed with PBS. Subsequently, 100 µL DCFDA solution (Sigma-Aldrich, D6883) was added to each well and incubated for 30 min at 37°C in the dark under 5% CO₂. As a positive control, 100 µM hydrogen peroxide (H₂O₂) was applied to the cells 4 h before DCFDA staining. The fluorescent signal was measured using a fluorometric plate reader (CLARIOstar, BMG Labtech) at 495 nm excitation and 529 nm emission wavelengths. Each experimental group was assessed in triplicate, and the results were expressed as the percentage of relative ROS levels compared to the positive control. In a parallel plate, stained cells were imaged via Axiovert (Primovert, Zeiss, Germany) under 20× magnification.

Inflammation and epithelial matrix remodeling

Inflammatory response induced by the dual exposure to BL and TPM was assessed with Human IL-6 ELISA kit from the cell supernatants according to the manufacturer’s instructions (Elabscience, E-EL-H6156). A key enzyme responsible for the degradation of collagen and the loss of mechanical integrity in photoaged skin, the level of matrix metalloproteinase-1 (MMP-1) was detected via the Human MMP-1 ELISA Kit (Elabscience, E-EL-H6073) as described previously. ¹⁵ Each sample was analyzed in duplicate in the ELISA assays.

Statistical analysis

The data were analyzed using GraphPad Prism version 9.0 (GraphPad Software, San Diego, CA, USA). For comparisons among more than two groups, one-way or two-way ANOVA followed by Tukey’s post-hoc test was applied when data were normally distributed, whereas the Kruskal–Wallis test followed by Dunn’s multiple comparisons test was used for non-parametric data. A p-value <0.05 was considered statistically significant for all analyses.

Results

Chemical characterization of TPM: Detectable levels of nicotine

In this work, the nicotine concentration in the TPM sample was determined to be 498.15 ± 3.19 µg/mL using an HPLC-DAD method. The linearity of the method was evaluated in a concentration range of 10-150 µg/mL by constructing a five-point calibration curve (10, 20, 50, 100, and 150 µg/mL), with each point analyzed in triplicate. The resulting calibration curve was defined by the linear regression equation y = 8220.9 × −3464.8, exhibiting an R² value of greater than 0.9999, which confirmed the excellent linearity of the method across the specified analytical range. The limits of detection (LOD) and quantification (LOQ) for nicotine were statistically derived from the parameters of the calibration curve. Based on the standard deviation of the y-intercept (σ) and the slope of the curve (S), the LOD and LOQ were calculated as 3.3 × (σ/S) and 10 × (σ/S), respectively, yielding concentrations of 0.68 µg/mL and 2.06 µg/mL. The specificity of the method was verified through peak purity analysis via DAD. Peak purity values obtained from real sample analyses were greater than 0.9999 for nicotine, demonstrating that the method is specific for the analyte without interference from matrix components. The calibration curve and chromatograms of the samples and nicotine standard are presented in Supplemental Figure S1.

Co-exposure to TPM and BL induced cytotoxicity and morphological alterations in human keratinocytes

According to the MTT analysis, TPM exposure significantly decreased human keratinocyte cell viability between 3.75 and 60 µg/mL concentrations compared to the control group (Figure 2(A)), which was noted as a concentration-independent response. In addition, when the cells were exposed to the BL along with TPM for 0.5 and 2 h, a significant decline in cell viability was observed in a time-dependent manner, especially in the 2-h exposure group (p < 0.01) (Figure 2(A)). Since no significant dose-response relationship was observed in the studied TPM samples, as a result of the cytotoxicity analysis. Although statistically insignificant, a slight increase in the viability of keratinocytes was observed at higher doses of TPM under BL co-exposure, which was likely attributable to the complex chemicodynamic properties of TPM. In addition, since the selected dose was the closest to the half-maximal inhibitory concentration (IC₅₀) in HaCaT cells, further analyses in the study were performed using the experimental dose of 3.75 µg/mL. In the phase-contrast microscopic examination performed in parallel with the quantitative analysis of cytotoxicity, BL exposure significantly decreased cell density depending on time. This effect appears to increase when 0.5 and 2-h exposures are applied simultaneously with TPM (Figure 2(B)).

Figure 2.

Cytotoxicity of TPM w or w/o BL exposure in HaCaT cells for 24 h. (a) MTT assay results of HaCaT cells exposed to TPM and BL Data are presented as mean ± SD (n = 3). (b) Morphological examination of HaCaT cells exposed to TPM (3.75 µg/mL) and different durations of BL (Magnification rate 10×). Ctrl: DMSO 0.5% (v/v); BL-:w/o blue light; BL+ (0.5 h): BL exposure for 0.5 h; BL+ (2 h): BL exposure for 2 h. Statistical differences between groups and Ctrl (BL-) *p < 0.005; **p < 0.0001; differences between groups compared to their respective BL− groups as ^ap< 0.0001, ^bp < 0.0001, ^cp < 0.0005, ^dp < 0.0001, ^ep < 0.0005.

Prolonged co-exposure to BL with TPM impaired keratinocyte autophagic flux through the conversion of LC3β-I to LC3β-II

According to the results of western blotting, BL exposure alone caused a time-dependent change in LC3β expression (Figure 3(A)–(C)). After exposure to 0.5 h of BL, LC3β-II expression showed notable increase compared to the control (BL -), indicating a raise in autophagic response (p < 0.05). In addition, BL- 2 h group exhibited a robust increase in LC3β-II and a higher LC3β -II/I ratio relative to control (BL-) cells (p < 0.01). Hence, it might be suggested that BL exposure alone potentially increased autophagic activity through autophagosome formation, the most notably in prolonged exposures in HaCaT cells. On the other hand, TPM markedly up-regulated of conversion of LC3β-I to LC3β-II in HaCaT cells (Figure 3(A)). Moreover, co-exposure of keratinocyte cells to TPM + BL showed a slight increase in the conversion of LC3β-I to LC3β-II, which was more prominent in prolonged BL exposure (Figure 3(a)). However, TPM + BL of keratinocyte cells did not show a diminished autophagic activity against dual stress compared to the corresponding BL exposure alone significantly. Although BL alone appeared to significantly induce autophagy activation, co-exposure with TPM further enhanced the conversion of LC3β-I to LC3β-II in a BL exposure duration-dependent manner (Figure 3(A)). The slight increase with TPM co-exposure suggested that dual exposure to those environmental stressors may possibly initiate autophagy-dependent cell death as an early defensive response in human keratinocytes, thereby boosting adaptive cellular responses against phototoxic stress. Furthermore, it might be suggested that BL alone markedly increased LC3β-II expression and LC3β-II/I ratio, co-exposure with TPM did not further elevate LC3β-II levels (Figure 3(B)).

Figure 3.

Modulation of autophagy in HaCaT cells. (a) Densitometric analysis of LC3β-II/I ratio following exposure to TPM (−/+) (3.75 µg/mL) and BL for 0.5 or 2 h; (b) Densitometric analysis of LC3β-II expression (c) Representative Western blot images showing LC3β-I (18 kDa), LC3β-II (16 kDa), and β-actin (50 kDa) protein levels. Data are presented as mean ± SD (n = 3). Statistical differences between selected groups were defined with ^* p < 0.05; differences between Ctrl (w/o TPM) versus groups were defined with ^#p < 0.05; ^##p < 0.01.

TPM and BL co-exposure up-regulated pro-inflammatory response and epithelial matrix remodeling in human keratinocytes

According to ELISA analysis, pro-inflammatory cytokine release through IL-6 was shown in Figure 4(A). Our findings revealed that IL-6 levels were elevated in a time-dependent manner following BL exposure (p < 0.01). Moreover, TPM exposure alone appeared to enhance IL-6-mediated inflammatory signaling in keratinocytes. Consistent with the increase IL-6 level, MMP-1 which was used as a marker of collagen degradation and loss of mechanical integrity in the skin, did not significantly increase in human keratinocytes via TPM exposure alone, whereas the co-exposure to TPM and BL resulted in a time-dependent increase in MMP-1 expression (Figure 4(B)).

Figure 4.

IL-6 and MMP-1 levels in HaCaT cells following TPM and BL exposure. (a) IL-6 level (pg/mL) measured by ELISA in HaCaT cell culture supernatants following 24-h exposure to TPM with or without blue light (BL) exposure for 0.5 or 2 h; (b) MMP-1 level (pg/mL) under the same treatment conditions. Cells were grouped as Ctrl (0.5% DMSO), TPM (3.75 µg/mL), and exposed to BL for either 0.5 h or 2 h. Data are presented as mean ± SD (n = 2). BL−: no blue light; BL+ (0.5 h): blue light exposure for 0.5 h; BL+ (2 h): blue light exposure for 2 h. Statistical differences between indicated groups were as ^*p < 0.05, ^**p < 0.005.

BL and TPM co-exposure amplifies oxidative stress in human keratinocytes in a time-dependent manner

Oxidative damage is one of the molecular pathways that play a key role in the phototoxicity process; thus, intracellular ROS levels were fluorometrically measured using DCFDA to assess oxidative stress induced by TPM and BL exposure in HaCaT cells. Relative ROS levels increased in all treatment groups compared to the negative control (medium-treated), with the highest ROS generation observed in the BL (2 h) + TPM group (Figure 5(A)–(B)). Moreover, this elevated level of ROS was found to be dependent on the duration of BL exposure as well (p < 0.01).

Figure 5.

Intracellular ROS formation in HaCaT cells following TPM and BL exposure. (a) Quantitative assessment of intracellular reactive oxygen species (ROS) levels, expressed as relative % compared to the positive control (PC), in HaCaT cells exposed to TPM (3.75 µg/mL),/mL) and/or blue light (BL) for 0.5 or 2 h (n = 3); (b) Representative fluorescence microscopy images of DCFDA-stained cells under the same exposure conditions. Data are presented as mean ± SD (n = 3). Ctrl: 0.5% (v/v) DMSO; BL−: no blue light exposure; BL+ (0.5 h): blue light exposure for 0.5 h; BL+ (2 h): blue light exposure for 2 h; PC: Positive control treated with H₂O₂. Statistical differences between indicated groups versus Ctrl (BL−) as ^*p < 0.05, ^**p < 0.005; statistical differences between TPM (BL−) and TPM (BL+) groups were indicated as ^#p < 0.05.

Discussion

In the present study, to our knowledge, the first demonstration highlighted the potential epidermal stress induced via HTP-particulate fraction and BL co-exposure in vitro. HaCaT keratinocytes were selected as a human-relevant in vitro model due to their well-characterized responsiveness to oxidative and phototoxic stress. Their reproducibility, barrier-relevant properties, and capacity to activate key stress signaling pathways such as AP-1 (activating protein-1), NF-κB (Nuclear factor kappa B), and autophagy, make this cell line a valuable tool for evaluating the cutaneous effects of environmental stressors. Although TPM is primarily considered in the scope of inhalation, it can deposit on skin or environmental surfaces, and enabling dermal contact. Given the lipophilic nature of many TPM components, mainly nicotine, their percutaneous absorption is plausible and reported in our previous study. ¹⁵ Hence, due to its semi-volatile, lipophilic properties and well-known transdermal permeability of compounds like nicotine, TPM poses a realistic concern for cutaneous absorption, particularly in contact areas such as the face and hands. In the present study, TPM exposure (3.75–60 µg/mL) did not exhibit a clear dose-dependent decline in the viability of HaCaT cells thus, 3.75 µg/mL as IC₅₀ in HaCaT cells was used as a representative dose to mimic moderate toxicity. It is known that HTPs generally emit fewer particles than cigarettes, but their aerosols still leave behind measurable residues. In addition, not only inhalation of HTPs but also finger-to-face transfer as well as dermal contact with HTP contaminated surfaces might represent a critical hazard for TPM exposure. However, given the chemically complex nature of TPM, extrapolating in vitro conditions to real-life exposure is inherently challenging. According to a report, surfaces contaminated with heavy cigarette smoke can accumulate TPM in the 0.1–1 µg/cm² range. ²⁴ Moreover, surface loading on the order of 1 µg of particulate/cm² of skin (assuming a few cm² of cell monolayer area/mL of cell medium) might represent a micro-scale exposure that one might encounter through everyday contact with tobacco aerosol residues. On the other hand, fine particles can directly settle on exposed skin and with continuous exposure, airborne deposition could achieve a coating in the same µg/cm² range as our in vitro dose. Similarly, human volunteer studies demonstrated that simply being in a smoke-filled room led to significant dermal uptake of nicotine, approximately ∼570 µg, through their skin within 3 h. ²⁵ From that perpective, it might be suggested that 3.75 µg/mL TPM is conservative in comparison thus, it represents a tiny fraction of the dermal burden one might get in a smoke-heavy environment, but it is highly relevant as a plausible incidental dose from routine contact with HTP emissions.

Our findings demonstrated that co-exposure to BL and TPM enhanced cytotoxic effect in human keratinocytes, driven by oxidative stress and inflammation. Mechanistically, the combined stress elevated intracellular ROS levels and triggered pro-inflammatory IL-6 signaling alongside upregulation of MMP-1 enzyme levels. Moreover, this dual exposure time-dependently altered autophagosome formation through up-regulated LC3β lipidation.

While previous studies emphasized gas-phase emissions, we demonstrated that TPM, a stable particulate fraction enriched in nicotine, aldehydes, and semi-volatile compounds, induces significant epidermal toxicity. According to the previous studies, this phase contains concentrated toxicants such as nicotine, aldehydes, PAHs, phenols, heavy metals, and semi-volatile organic compounds, as the major drivers of toxicity, and in addition, mimics surface or skin deposition.^15,20 Hence, TPM extraction allowed for standardized assessment of HTP-induced oxidative and inflammatory responses in human keratinocytes, aligning with previous in vitro toxicity models. Although nicotine was the only quantified constituent in our study, it is important to acknowledge that TPM contains a broad mixture of toxicologically relevant compounds. Independent analytical studies have shown that HTP aerosol contains water-soluble organic compounds, carbonyls, VOCs, PAHs, and residual metals, albeit generally at lower levels than in combustible cigarette smoke.^26–28 These constituents may interact synergistically to influence oxidative stress, inflammatory signaling, and protein homeostasis in keratinocytes. Therefore, toxicity responses observed via TPM in our study cannot be solely attributed to nicotine and may instead reflect the combined effects of multiple TPM constituents with diverse chemicodynamic properties. Further analysis on the full chemical profiling of TPM samples will enable clearer attribution of the mechanistic aspect underlying keratinocyte stress responses. On the other hand, BL irradiation is known to induce ROS production, DNA damage ²⁹ ; endoplasmic reticulum stress and apoptosis ¹¹ ; skin aging, ³⁰ either alone or combined with mainstream tobacco smoke, based on recent reports. Our experimental approach used in this study was informed by prior in vitro BL studies. Hettwer and colleagues (2017) exposed HaCaT keratinocytes to 13 J/cm² of 470 nm BL (1.8 mW/cm² over 2 h) and observed a significant rise in intracellular ROS and oxidative damage in keratinocytes. ²³ Therefore, we selected a comparable irradiance (2.4 mW/cm²) and extended the exposure up to 17 J/cm² to ensure a biologically relevant dose that can elicit cellular responses. Notably, BL-induced oxidative stress was detected in keratinocytes at even “moderate” doses according to Hettwer and colleagues’ findings. Hence, our acute exposure model, 4.3 J/cm² exposure, serves as a low-to-moderate BL dose, while the 17.3 J/cm² exposure approaches the range where measurable oxidative effects in keratinocytes have been reported in literature. Moreover, the selection criterion allowed us to probe dose-dependent effects of BL under controlled conditions and then extrapolate the findings to plausible real-world simulations, such as the effect of several hours of strong sunlight or prolonged nightly screen use on skin biology in vitro. In addition to the in vitro simulation examples, it is critical to contextualize these doses in terms of real-world BL exposure. Notably, direct full sunlight can emit roughly 1.5 mW/cm² of blue wavelengths in the 425–465 nm range, which might be suggested as the selected experimental BL intensity is slightly higher than the peak noontime sun in the blue range. Consequently, the “acute” 0.5 h exposure (4.3 J/cm²) in our model is roughly equivalent to about 45–50 min of midday sun in terms of BL dose, whereas the 2 h exposure (17.3 J/cm²) corresponds to roughly 3–4 h of midday summer sun exposure on the skin. ³¹ These comparisons might indicate that while our BL doses are significant, they mimic realistic environmental exposure on the order of an hour of strong sunlight (0.5 h) or a few hours of sunlight (2 h). According to the recent reports, high-dose BL irradiation (200 J/cm², 66 min) is reported to induce ROS generation in keratinocytes, which is accompanied by the up-regulation of pro-inflammatory cytokines such as IL-6, IL-8, and IL-1α. ²⁹ In parallel, numerous literature data reported that traditional cigarette smoke impairs the antioxidant defense mechanism of human skin via triggering the release of inflammatory mediators and destabilization of the NLRP3 inflammasome.^32,33 In our study, the combined BL + TPM exposure led to ROS accumulation, depleting the HaCaT antioxidant capacity. In keratinocytes, oxidative stress activates inflammation and matrix remodeling pathways such as MAPK/AP-1 and NF-κB. In addition, BL irradiation has been shown to activate transient receptor potential (TRP) channels, leading to AP-1 and NF-κB activation, resulting in up-regulation of TNF-α. ²⁹ The elevation in the inflammatory response and intracellular ROS levels were prominent in BL and TPM co-exposure, confirming that oxidative damage has a central role in the keratinocyte damage pathway, and long-term BL exposure produces a phototoxic microenvironment in keratinocytes together with TPM exposure. Notably, MMP-1 was significantly elevated only under BL + TPM, suggesting that BL potentiates matrix remodeling in the presence of TPM. According to a study in reconstructed skin, UVA exposure upregulated the production of MMP-1, mainly by dermal fibroblasts. ³⁴ In addition, increased MMP-1 levels was reported under BL irradiation in human keratinocytes, similar to the interpreted findings in the present study. ³⁵ Hence, our findings highlight the time-dependent potency of BL-irradiation in promoting the degradation of extracellular matrix components in vitro. When comparing to combined exposures, previous data strongly support that co-exposure to tobacco-pollutants and BL amplifies skin damage. In a 3D human skin model study, repeated exposure to cigarette smoke extract (CSE) and solar-simulated light (UVA, visible, and IR), a notable increase in MMP-1 activity was reported only under the CSE + light exposure, in addition to a decrease in collagen III and IV in 3D model as well. Moreover, it was noteworthy that the dual exposure hyper-activated other key signaling pathways such as ERK1/2 and JNK, involved in keratinocyte proliferation, differentiation, wound-healing and inflammation against environmental stressors. ¹⁸ Similarly, a recent study reported that co-exposure to UVA and cigarette sidestream smoke led to MMP-1 upregulation in human dermal fibroblast, which was attributed to the global histone H3 acetylation. ³⁶ Moreover, when HaCaT cells were co-exposed to TPM and BL, the conversion of LC3β-I to LC3β-II showed a further, although not statistically significant, increase compared with BL alone. This trend might suggests that dual exposure may enhance the autophagic response in a BL-exposure time-dependently. Excessive or sustained autophagy under such combined stress could reflect an adaptive defense mechanism or, if prolonged, a transition toward autophagy-associated cell death. Similar biphasic autophagy patterns have been reported under oxidative and phototoxic stress, where moderate activation is cytoprotective but prolonged stimulation was previously reported as detrimental in keratinocytes. Consistent with a previous study in HaCaT cells, PM2.5 (±UVB) activated autophagy in HaCaT cells, and pharmacological autophagy inhibition increased apoptosis, supporting a cytoprotective role for autophagy during PM and UV-induced injury. Our BL-only condition showed an increased LC3β-II and an elevated LC3β-II/I ratio, in line with an early, adaptive autophagic response. ³⁷ Under BL + TPM co-exposure, LC3β expressions were altered relative to BL alone, a divergence that may reflect phase and exposure-dependent autophagy dynamics reported for TPM. Specifically, cigarette smoke was reported to elicit biphasic autophagy, and an initial beneficial phase followed by a phase with impaired autophagosome-lysosome fusion and progression toward apoptosis.^38,39 Based on the previous findings, it might be suggested that TPM constituents may shift autophagy from adaptive to dysregulated states depending on dose and exposure time in keratinocytes, as well. Moreover, the related pattern is aligned with a broader synthesis indicating that short-term PM exposure generally induces autophagy, whereas prolonged or pathological contexts can suppress autophagy or block flux. However, as LC3β-II accumulation alone does not establish changes in autophagic flux, our findings should be interpreted in detail. From another perspective, our results might reflect that autophagy was already maximally activated by BL, and additional TPM stress might not promote further autophagosome accumulation. This unchanged LC3β-II expression under dual exposure may reflect a balance between autophagosome formation and/or degradation, suggesting that autophagic flux was not substantially altered in human keratinocytes. As a limitation of the present study, further confirmation through p62/SQSTM1 or lysosomal inhibition assays would be required to verify whether enhanced LC3β-II reflects an increased autophagic activity or a slowdown in autophagosome turnover. On the other hand, even though apoptotic cell death is a well-recognized downstream outcome of oxidative stress-driven cellular damage, specific markers were not assessed in the study as our primary objective was to characterize the ROS-mediated stress response and early autophagy modulation in keratinocytes under the aforementioned exposure conditions. Future studies within a similar model will incorporate Annexin V/PI-based apoptosis profiling and downstream caspase signaling assays to further clarify the transition from oxidative and autophagy-related responses to regulated keratinocyte cell death under dual environmental stress. Based on previous data, BL irradiation has been shown to increase TRPV1 expression and inhibit the EGFR (epidermal growth factor receptor)/PI3K/AKT/FoxO3a pathway in HaCaT cells, ultimately leading to EGFR degradation. Consequently, suppression of the downstream AKT signaling cascade induces pro-apoptotic signaling and promotes apoptosis. ⁴ The overall data might suggest that concurrent BL + TPM exposure depleted antioxidant defense, and possibly altered the autophagic activity in a BL exposure duration-dependent manner, thus aggravating epidermal injury in vitro. Taken together, these findings imply that prolonged exposure to both BL and TPM may exceed the compensatory thresholds of keratinocytes, modulating cytoprotective processes such as autophagy and antioxidant defense, which in turn exacerbates cellular damage.

Study limitations

In the present study, there are several points that need to be addressed as limitations. First, the experiments were conducted with the immortalized HaCaT cell line, which, although widely accepted as a robust in vitro model for epidermal toxicity and phototoxicity studies, does not fully represent the structural and cellular complexity of human skin. Hence, future studies employing dermal fibroblasts as well as 3D reconstructed human skin models may provide more physiologically relevant insights into epidermal–dermal interactions under dual exposure scenarios. Second, given the complex content of TPM, including VOCs, PAHs, and trace metals, the observed biological effects cannot be attributed to a single constituent. Although the current study only identified nicotine content to confirm TPM consistency, additional chemical profiling techniques would help to elucidate the contribution of specific toxicants to the observed epidermal stress responses. Third, the autophagy response was limited to LC3β-I/II expression analysis. While autophagosome formation through LC3β-II expression is a common marker of autophagy, it does not provide definitive information on autophagic flux. Therefore, the lack of complementary markers such as p62/SQSTM1 or lysosomal inhibition assays limits the interpretation of toxicity responses that reflect autophagy induction or impaired autophagosome formation. Moreover, downstream cell death pathways, including apoptosis and necrosis, were not directly evaluated. Although oxidative stress, inflammation, and autophagy modulation represent early and mechanistically relevant responses to combined environmental stressors, the lack of apoptosis-specific markers precludes definitive conclusions regarding the terminal fate of keratinocytes under prolonged co-exposure. In future investigations, the incorporation of Annexin V/PI staining, caspase activation assays, and mitochondrial dysfunction analyses would further strengthen the mechanistic understanding of epidermal stress. Despite these limitations, the study provides a novel and mechanistically informative framework for elucidating the combined effects of BL and HTP-derived TPM on human keratinocytes, highlighting the importance of integrated exposure models within the context of lifestyle-related environmental toxicology.

Conclusion

These findings provide novel insights into the detrimental impacts of simultaneous environmental exposures, particularly accumulative phases of new-generation tobacco products and BL sources on skin health. Future dermatotoxicological risk assessments should incorporate such combined exposure models to better reflect real-life scenarios. Moreover, the results underscore the importance of incorporating comprehensive environmental exposure models in toxicological risk assessment strategies.

Supplemental material

Supplemental material - Reactive oxygen species-mediated epidermal stress in human keratinocytes under dual exposure to blue light and heated tobacco product

Supplemental material for Reactive oxygen species-mediated epidermal stress in human keratinocytes under dual exposure to blue light and heated tobacco product by Rengin Reis, Kubra Kolci, Melis Cetin, Sercan Yildirim in Human & Experimental Toxicology.

Supplemental material

Supplemental material - Reactive oxygen species-mediated epidermal stress in human keratinocytes under dual exposure to blue light and heated tobacco product

Supplemental material for Reactive oxygen species-mediated epidermal stress in human keratinocytes under dual exposure to blue light and heated tobacco product by Rengin Reis, Kubra Kolci, Melis Cetin, Sercan Yildirim in Human & Experimental Toxicology.