Sub-acute dihydroxyacetone exposure alters mitochondria and induces cardiac strain

Abstract

Dihydroxyacetone (DHA), a carbohydrate detected in e-cigarette aerosols and used in sunless tanning products, is genotoxic and cytotoxic in various cell models. We previously demonstrated that inhaled DHA induced lung injury in A/J mice. Here, we evaluated cardiac effects of sub-acute DHA exposure in A/J mice and human Ac16 cardiomyocytes. Two weeks of DHA exposure significantly increased cardiac mitochondrial DNA (mtDNA) copy number and altered global longitudinal strain in both sexes. Ejection fraction and fractional shortening did not differ significantly after DHA exposure in either sex, though male mice showed increased ejection fraction and fractional shortening, while female mice decreased in both parameters. Male DHA-exposed mice also showed decreased body weight, while female mice body weight was unchanged. Plasma metabolic analyses, limited to female mice due to Flexivent analysis on the males, revealed significantly altered metabolic markers, including increased glucose levels and decreased resistin. To confirm mitochondrial DNA changes and DNA damage induction, we examined DHA exposure in the Ac16 cells sub-acutely exposed to DHA for 14 days. Sub-acute DHA exposure increased oxidative DNA lesions, without a significant increase in reactive oxygen species or mitochondrial content. mtDNA copy number was also increased in the sub-acutely exposed Ac16 cells. These results demonstrate that inhaled DHA alters cardiac cells and induces early cardiac tissue and cell changes, in addition to its lung effects, suggesting that the e-liquid oxidation products pose inhalation-related health risks.

Graphical Abstract

Highlights

• •Dihydroxyacetone (DHA) exposure increases mitochondrial DNA copy number in cardiomyocytes and cardiac tissue.
• •Sub-acute DHA exposure induces oxidative DNA damage in cardiomyocytes.
• •Sub-acute DHA exposures reveal sex specific differences in cardiac function and increased cardiac strain in both sexes.

1. Introduction

Electronic cigarettes (e-cigarettes) were introduced in the United States (US) in 2007 as an alternative to tobacco cigarette smoking. Their popularity grew from 2019 to 2023, with 6.5% of US adults using e-cigarettes, despite their growing increase over their safety [1], [2], [3], [4]. E-cigarette devices come in various shapes and sizes, including universal serial bus (USB)-powered and battery-operated versions, and they use various e-liquid compositions, including flavors and nicotine. E-liquids contain two main ingredients, propylene glycol and vegetable glycerol, which are heated to create an aerosol that a user inhales. A significant body of work has focused on the inhalation effects of nicotine and flavorings from e-cigarettes. A number of studies have examined e-liquids combustion without additional additives, demonstrating that propylene glycol and vegetable glycerol (PG:VG) increase inflammation and oxidative stress [5], [6], [7]. However, there is less work examining the effects of unique oxidation products produced by propylene glycol and glycerol [5], [7]. Large mass spectrometry experiments have identified complex chemical mixtures produced by e-liquid combustion, including caffeine, carbohydrates, terpenes, a pesticide, metals, carbonyls, free radicals, and phthalates [8], [9], [10], [11]. Production of these compounds depends on the type of device, their source, the e-liquid composition, and also the quality of e-liquid components. Many of the compounds produced are toxic or have unknown effects but are found in very low concentrations [8], [9], [10], [11]. There are also several chemicals produced at higher quantities with unknown toxicity, many of which have yet to be studied in animal or human exposures [8], [9], [10], [11].

One overlooked component of e-cigarette aerosol is dihydroxyacetone (DHA). DHA is a triose sugar produced by the free radical oxidation of glycerol. Up to 2.29 µg of DHA per puff in a 55 ml puff is produced, depending on device settings and heating [8]. DHA generation is significantly higher than that of some flavorings and additives, yet the effects of inhalation exposure to DHA are poorly understood. We initially characterized DHA exposure effects in systemic cell models of lung, heart, kidney, and liver [12], [13], [14], [15]. These in vitro studies demonstrated that DHA induces cytotoxicity, genotoxicity, metabolic imbalances, oxidative stress, and mitochondrial dysfunction [12], [13], [14], [15]. In human embryonic kidney cells (HEK293T), an acute but cytotoxic dose of DHA decreased mitochondrial function and ATP production within 24 h [15]. Additionally, an imbalance in redox cofactors nicotinamide adenine dinucleotide (NAD⁺/NADH) was observed in HEK293T, HepG3, and H9c2 cells [12], [14], [15]. In H9c2 cells, a rat cardiomyocyte cell line, a low, non-cytotoxic dose of DHA causes metabolic imbalances, including reactive oxygen species (ROS), mitochondrial energetics, and proliferation [12]. We also determined that DHA was genotoxic in lung, cardiac, and liver models, inducing DNA damage, mutations, and chromosomal aberrations [13], [16].

Following up on these findings, we then examined the inhalation exposure effects of DHA on the lungs of A/J mice using acute and sub-acute exposures [17]. We estimated DHA exposures from mass spectrometry studies and e-cigarette user studies. For cigarette-like devices, puff volumes vary from 96.81 to 133.92 ml per puff, while tank devices are 331.2–519.6 ml [18], [19]. Approximately 10 puffs per session have been reported by users, with vaping sessions occurring up to 24 times per day [18], [19]. Therefore, DHA exposures range from 38.4 µg/day to ∼ 1 mg/day, allowing a user to potentially inhale high micromolar to low millimolar doses daily. Therefore, we selected a low dose of 5 μg, approximating a single puff, and intermediate doses of 130 and 600 μg. The highest dose reflects the European Union Regulatory Commission’s calculated inhalation exposures to DHA from spray tanning booths [20]. 24 h after acute exposure to 5, 130, or 600 µg DHA, inflammatory responses increased and an increased cell count in bronchoalveolar lavage fluid (BALF) [17]. Histologically, we found significant alveolar damage and increased lung injury at the 130 and 600 μg doses [17]. Sub-acute exposure to 5 µg DHA for 2 weeks also showed increased BALF cell counts and protein levels, and a sex specific difference in cytokine response [17]. Male mice had reduced levels of IFN-γ and TNF-α and showed mixed obstructive and restrictive lung function. In female lungs, we observed significant lung damage, supported by decreased alveolar density and increased collagen deposition [17].

The exposure effects of DHA across different models show some of the same hallmarks observed in studies using PG:VG only exposures, suggesting that DHA is a contributor to the adverse exposure effects observed from e-cigarettes [5], [7], [21]. Given the findings that DHA exposure induced lung injury in A/J mice, we assessed the exposure effects on the hearts of the same mice after sub-acute DHA exposure [17]. We also examined sub-acute exposure effects of DHA in Ac16 cardiomyocytes to confirm cellular effects.

2. Materials and methods

2.1. In vitro exposure to DHA

The human cardiomyocytes (Ac16) were a gift from Dr Prasanna Krishnamurthy’s lab at the University of Alabama at Birmingham (UAB). Ac16 cells were cultured in Dulbecco’s Modified Eagle medium/F12 (DMEM/F12, Corning, Manassas, VA) supplemented with 10% FBS. For sub-acute dosing, cells were maintained at low confluency and cells were treated with 50 μl of no glucose DMEM media (labeled as control, Gibco, Grand Island, NY) with or without 0.2 mM DHA (PHR 1430, Sigma Aldrich, St. Louis, MO, US) for 14 days, using a schedule of 5 days on and 2 days off. This produced a mock-treated, passage matched control and sub-acutely dosed DHA Ac16 cells. The cells were screened monthly for mycoplasma contamination using the Lonza MycoAlert kit (Walkersville, MD, United States).

2.2. In vivo exposure to DHA

All animal procedures were approved by the University of Alabama at Birmingham Institutional Animal Care and Use Committee (IACUC) protocol 22789. A/J mice were purchased from Jackson Labs (Augusta, ME). Un-anesthetized A/J mice were exposed to DHA dissolved in saline or saline control using the Kent Scientific Aeroneb Lab Control Module with a nose-only animal holder (Kent Scientific, Torrington, CT, US). The 5 μg dose was selected based on the DHA production volume of ∼2.29 µg/puff for 55 ml puff volume, using an estimated single puff volume of 100 ml of cigarette like devices [8], [18], [22]. Male and female A/J mice were exposed to 5 µg of DHA for 6 days on and 1 day off for 2 weeks. The day off allowed for echocardiograms to be performed on the mice. After exposures, mice were returned to room air and monitored continuously for signs of distress. The experiment started with 6 animals per group. Two male mice in the DHA exposed group were lost during exposures, leaving 4 mice in the male DHA exposed group.

These animals were also used in a previous paper [17]. Flexivent analysis was performed on male mice in that study, so blood analyses were available only from female mouse samples. Sex specific results are indicated by color in all figures. Body weights were measured throughout the exposure period (n = 6 male vehicle, female vehicle, and female DHA exposed and n = 4 male DHA exposed). A linear mixed-effects model was used for longitudinal data, and a two-way ANOVA with Tukey’s multiple comparisons test was used as a post hoc test.

2.3. Repair assisted damage detection (RADD)

DNA damage was evaluated using RADD, which detects DNA lesions using a specific cocktail of DNA repair enzymes [23]. These enzymes remove the DNA lesion, and then the gap or strand break is tagged using a digoxigenin-labeled dUTP inserted with Klenow exo-, which lacks proofreading [23], [24]. We examined oxidative lesions using Fapy-DNA glycosylase (FPG, New England Biolabs (NEB), Ipswich, Massachusetts), Endonuclease IV (Endo IV, NEB), and Endonuclease VIII (Endo VIII, NEB) (oxRADD).

Ac16 cells sub-acutely exposed for 14 days were plated in 8-well chambers (Ibidi, Fitchburg, WI) at a density of 3000 cells per well. The chambers were allowed to adhere overnight (ON) in a 5% CO₂ incubator at 37 °C. After attachment, cells were dosed with 0.2 mM DHA for another 48 h to complete the 14 day dosing. After exposure, cells were fixed with 3.7% formaldehyde (Thermo Fisher, Waltham, MA) in phosphate-buffered saline (PBS, Hyclone, Logan, UT) for 10 min at room temperature (RT, ∼22°C) and washed three times with PBS. Then, cells were permeabilized using Biotium permeabilization buffer for 10 min at RT (Fremont, CA) and washed three times with PBS. The chambers were then incubated with the lesion removal cocktail and resuspended in 1× ThermoPol buffer (NEB) for 1 h at 37°C in a hybridization oven. Once the incubation was completed, the gap-filling mixture of Klenow exo- (Thermo Fisher) and digoxigenin-labeled dUTP (Sigma Aldrich, Burlington, MA) was directly added to the slides and placed again within the oven at 37°C for an additional 1 h. The slides are washed three times with PBS and blocked with 2% BSA (Jackson Immuno, West Grove, PA) for 30 min. After blocking, slides are incubated with primary antibody anti-digoxigenin (1:250 ab420, Abcam, Cambridge, UK) or anti-mouse IgG1 isotype control (1:625 5415, Cell Signaling, Danvers, MA) for 1 h at RT. IgG1 is used as a negative control in these cells. Once the primary antibody has finished incubating, the slides are washed three times with PBS and incubated with a secondary antibody anti-mouse Alexa Fluor 546 (1:400 Thermo Fisher) for 1 h at RT. The samples are then incubated with Hoechst solution (1:800, Thermo Fisher) for 15 min at RT and washed with PBS three times. Samples are then imaged using the Keyence BZ-X800 (Keyence, Itasca, IL) microscope with the 20 × objective (NA 0.75). For analysis, the Nikon Elements software created a region of interest (ROI) around the nucleus, and the total intensity for the RADD channel within the nucleus was recorded. The mean fluorescent intensity for all nuclei ± the standard error of the mean (SEM) was plotted in GraphPad Prism, and significance was assessed using Welch’s t-test.

2.4. Immunofluorescence

Immunofluorescent detection of mitochondria was performed on the control and sub-acutely DHA dosed Ac16 cells plated in 8-well chambers at a density of 3000 cells per well. The chambers were allowed to adhere ON in a 5% CO₂ incubator at 37 °C. After attachment, cells were treated with 0.2 mM DHA for 48 h to achieve a 14 day exposure. After exposure, cells were fixed with 3.7% formaldehyde in PBS for 10 min at RT and washed three times with PBS. Then, cells were permeabilized using Biotium permeabilization buffer for 10 min at RT and washed three times with PBS. Chambers were blocked for 30 min at RT in 2% BSA in PBS. The cells were incubated with α-tubulin for 1 h at RT. The cells were then washed 3 times with PBS and incubated with a secondary antibody anti-mouse Alexa Fluor 546 (1:400) for 1 h at RT. The cells were then washed 3 times with PBS and incubated with anti-mitochondria, clone 113–1 (Sigma Aldrich) labeled with Alexa 488 for 1 h at room temperature. 10 min before, Hoechst was added for nuclear staining (1:800). Finally, cells were washed 3 times with PBS and imaged. Imaging was performed on a Keyence (BZ-X800) using a 20 × objective. A minimum of 100 cells across multiple imaging fields were acquired for each condition, with three biological replicates. The Nikon Elements software was used to define a binary threshold for the fluorescence signal, then the fluorescent signal for each field was acquired and averaged over all biological replicates. The mean fluorescent intensity signal ± SEM was graphed using GraphPad Prism, and the significance was calculated using a Welch’s t-test.

Immunofluorescence was also performed on hearts from A/J male and female mice sub-acutely exposed to 5 µg DHA to evaluate structural changes (n = 6 male vehicle, female vehicle, and female DHA exposed and n = 4 male DHA exposed). Whole hearts were fixed in Bouin’s solution, then paraffin embedded. Tissues were sectioned at 5 µm for immunofluorescent analysis. Tissue sections were deparaffinized and rehydrated. Antigen retrieval was performed in antibody signal enhancer (ASE) buffer containing 10 mM glycine (Thermo Fisher), 1% Triton-X (Sigma Aldrich), and 0.05% Tween-20 (Thermo Fisher) [25]. The slides were immersed in ASE and heated three times: 18 s on and 5 min cooling off; 15 s on and 5 min cooling off; and 15 s on and 5 min cooling off. The slides were then cooled to RT, rinsed with Milli-Q H₂O, and dried off. The slides were blocked with 5% goat serum (Thermo Fisher) in PBS for 30 min, then incubated with the respective primary antibody (Table 1). The next day, the slides were washed three times for 5 min each with 1 × Tris-buffered saline (TBS) (MP Biomedical, Solon, OH) with 0.1% Tween-20 (TBST). The slides were dried and incubated for 1 h at RT with the corresponding secondary antibody Alexa Fluor 546 (1:400) diluted in 5% goat serum. Nuclear staining was performed using Hoechst (1:800) for 15 min. After incubation, the slides were washed three times with 1 × TBST for 5 min. The slides were then mounted using Prolong Gold (Life Technologies, Eugene, Oregon) and allowed to dry ON.

Table 1.

Antibodies used in this work.

Antibody	Dilution	Source
Myosin	1:10	Cell Signaling 8824
Anti-mitochondria Alexa fluor 488 clone 113–1	1:250	Sigma Aldrich Mab1273A4
Alexa 546 Secondary	1:400	Invitrogen A11010
Mouse (G3A1) mAb IgG1 Isotype Control #5415	1:625	Cell Signaling Technologies 5415S
Anti-mouse Alexa Fluor 647	1:400	Invitrogen A21235

The whole heart slides were imaged using the Keyence with a 10X objective (0.40 NA). The Nikon Elements software was used to perform the analysis. The fluorescent intensity was measured using a binary threshold across the entire tissue with ROIs randomly selected throughout the image. The values obtained for each ROI, with at least 10 ROIs per heart, were averaged for each animal. The average fluorescent intensity per mouse heart was graphed in GraphPad Prism and presented as mean ± SEM for each treatment group and sex. Significance was calculated using a two-way ANOVA with a Tukey multiple comparison correction with a family-wise alpha of 0.05 and a 95% confidence interval.

2.5. Cellular reactive oxygen species (ROS) measurements

CM-H2DCFDA (DCFDA, Invitrogen, Eugene, Oregon) was performed on the control and sub-acutely DHA dosed Ac16 cells plated in 8-well chambers at a density of 3000 cells per chamber. The chambers were allowed to adhere ON in a 5% CO₂ incubator at 37 °C. After attachment, cells were dosed with 0.2 mM DHA daily for 48 h to achieve a 14 day exposure. The cells were transferred from media to 1 × PBS for fluorescent indicator addition. Cells were also dosed with tert-Butyl hydroperoxide (TBHP, Thermo Fisher) in PBS for 1 h, as a positive reactive oxygen species (ROS) control. After 30 min, the DCFDA indicator dye was added to the positive control and to the control and DHA-dosed cells. Cells were immediately imaged using the Keyence with the 20x objective. A minimum of 100 cells across at least four fields of view were imaged for each condition, with three biological replicates. We used Nikon Elements to create a binary threshold based on cell-specific fluorescence intensity. We extracted the fluorescence intensity for cells within each field. We averaged the intensities across multiple fields of view and biological replicates. We calculated the mean fluorescence intensity for each condition. GraphPad Prism was used to graph the mean fluorescent intensity ± SEM, and significance was calculated using a Welch’s t-test.

MitoSOX Red mitochondrial superoxide indicator (Thermo Fisher) was used to measure mitochondrial ROS formation. Control and DHA-dosed Ac16 cells were transferred from media to 1X PBS for experiments. As a positive control, cells were also dosed with rotenone for 30 min. After 30 min of rotenone, MitoSOX was added to control, DHA-exposed, and rotenone-exposed cells for 30 min. Cells were immediately imaged using the Keyence with a 20 × objective. We imaged at least 100 cells across at least 4 fields of view for each condition, with 3 biological replicates. We used Nikon Elements to create a binary threshold based on cell-specific fluorescence intensity. We extracted the fluorescence intensity for cells within each field. We averaged intensities across multiple fields of view and biological replicates. We calculated the mean fluorescence intensity for each condition. GraphPad Prism was used to graph the mean fluorescent intensity ± SEM, and significance was calculated using a Welch’s t-test.

2.6. NADP(H) measurements

Intracellular NAD(P)H autofluorescence was measured using two-photon excitation microscopy as described in [12]. Briefly, 30,000 control and DHA dosed Ac16 cells were seeded in fluorodishes (World Precision Instruments, Sarasota, FL) for 3 days. The cell media was changed to DMEM/F12 phenol-free media 24 h before imaging. Cells were imaged with a Nikon A1R Multiphoton Microscope with a 25X objective. We imaged at least 100 cells across at least 5 fields for each condition, with 3 biological replicates. We used Nikon Elements to create a binary threshold based on cell-specific fluorescence intensity. We extracted the fluorescence intensity for cells within each field and averaged intensities across multiple fields of view. Each biological replicate was normalized to control then averaged to determine the fold change in NAD(P)H content. GraphPad Prism was used to graph the normalized fluorescent intensity change ± SEM, and significance was calculated using a Welch’s t-test

2.7. Echocardiograms

Echocardiography was performed before treatment with DHA, on day 7, and on day 14 during treatment. For the final echocardiography, n = 5 male vehicle mice with one mouse unable to be analyzed due to an erratic heart rate; n = 4 male DHA exposed mice; n = 6 female vehicle and female DHA exposed mice. Mice were anesthetized by 2% isoflurane inhalation in an induction chamber and then transferred to a 37 °C platform with 1.0–1.5% isoflurane to maintain ≈ 400 beats/min. Cardiac function was assessed by echocardiography using a Vevo 3100 ultrasound system (FUJIFILM VisualSonics, Inc., Toronto, CAN). A minimum of 3 cardiac cycles were collected per imaging plane for all measurements. Ejection fraction was determined using the left ventricle (LV) trace method in B-mode. End-diastolic and end-systolic frames were automatically identified and calculated after tracing the endocardial regions using ECG gating. The Vevo software was then used to measure the ejection fraction (EF). Fractional shortening (FS) was measured from M-mode images acquired from the papillary muscles. The LV trace tool was used to trace the endocardial and epicardial borders. From these tracings, Vevo LAB identified the end-diastolic diameter (LVIDd) and the end-systolic diameter (LVIDs), and FS was calculated.

Myocardial Strain was measured using the VevoStrain speckle tracking software. For strain analysis, the endocardial contour was traced to initialize tracking and speckle tracking, as visually verified. VevoStrain quantified radial strain by reflecting the thickening and thinning of the myocardial wall perpendicular to the long axis. Longitudinal strain was quantified by reflecting the shortening and lengthening along the long axis of the ventricle. Global longitudinal strain was calculated as the average peak across all the myocardial segments. Global radial strain as calculated as the average peak radial strain across all myocardial segments. For this analysis, n = 4 male vehicle mice with one mouse’s echocardiographs failing to yield a strain calculation; n = 4 male DHA exposed mice; n = 6 female vehicle and female DHA exposed mice. The values obtained are displayed as mean ± SD. Significance was calculated using a two-way ANOVA with a Tukey multiple comparison correction with a family-wise alpha of 0.05 and a 95% confidence interval.

2.8. Transmission electron microscopy

Transmission electron microscopy (TEM) was performed on a subset of hearts from sub-acutely exposed A/J male and female mice to evaluate cardiomyocyte ultrastructure (n = 3 mice per group). Tissues were processed, stained, and imaged by EM labs. Mitochondria counts and lipid droplet counts were quantified using ImageJ. Values obtained are displayed as mean ± SEM. Significance was calculated using a two-way ANOVA with a Tukey multiple comparison correction with a family-wise alpha of 0.05 and a 95% confidence interval.

2.9. Mitochondria copy number assay

10 µg of cardiac tissue from the vehicle and DHA-exposed A/J mice were measured out and homogenized (n = 6 male vehicle, female vehicle, and female DHA exposed and n = 4 male DHA exposed). The genomic DNA was extracted and isolated using the Purelink Genomic DNA Mini Kit (Thermo Fisher) per the manufacturer’s instructions. The quantity of gDNA was determined using the AccuBlue Broad Range dsDNA Quantitation Kit (Biotium). The DNA was used for qPCR for the mitochondrial gene primers, COX1 and ND1 (Table 2). The normalization control was actin (ACTB, Table 2).

Table 2.

Primers used in this work.

Cox1	Human Forward: TTCGCCGACCGTTGACTATTCTCT Human Reverse: AAGATTATTACAAATGCATGGGC Mouse Forward: AGCCGGAAACCTAGCCCATGC Mouse Reverse: CACCCCGGCTAGGTGGAGGG
ND1	Human Forward: CCTAGCCGTTTACTCAATCCT Human Reverse: TGATGGCTAGGGTGACTTCAT Mouse Forward: TCCCTCACAACACACCCCCT Mouse Reverse: TGGAGCTTCCACGTGGGCTT
β-actin	Human Forward: CATCTCTTGCTCGAAGTCCA Human Reverse: ATCATGTTTGAGACCTTCAACA Mouse Forward: CCTAGACTTCGAGCAAGAGA Mouse Reverse: GATGCCACACGATTCCATAC

The qPCR contained a total of 20 µl reaction mixture with 2 µl of 10 µM forward and reverse primer mix, 15 ng of genomic DNA, 10 µl SsoAdvanced Universal SYBR Green Supermix (Bio-Rad, Hercules, CA), and DNA/RNA free water. Set up in the CFX96 Touch Real-Time PCR Detection System (Bio-Rad). Cycling conditions 2 min at 50°C, 2 min at 95°C, 40 cycles for 15 s at 95°C, and 1 min at 66°C. Three technical replicates were performed for each experiment, and outliers within the technical runs were assessed by the ROUT method and removed. ΔC_T was calculated as mtDNA C_T - nucDNA C_T, with β-actin serving as the nuclear control. The relative change in mtDNA copy number for treated and untreated samples was calculated for COX1 and ND1. This relative change was graphed in GraphPad Prism, and mean ± SEM are shown. Significance was calculated using a two-way ANOVA with a Tukey multiple comparison correction with a family-wise alpha of 0.05 and a 95% confidence interval.

2.10. Immunohistochemistry

Collagen deposition was evaluated on heart sections from mice exposed sub-acutely to 5 µg DHA for 2 weeks as described (n = 6 male vehicle, female vehicle, and female DHA exposed and n = 4 male DHA exposed). The tissues were fixed in Bouin’s solution, processed, embedded, and stained with Masson’s Trichrome staining by UAB’s Comparative Pathology Lab. After staining, regions in the tissue slides were imaged using the Keyence microscope with a 10X objective. Collagen percentage was quantified using a binary mask threshold applied to the whole image in ImageJ. The values obtained are displayed as mean ± SEM. Significance was calculated using a two-way ANOVA with a Tukey multiple comparison correction with a family-wise alpha of 0.05 and a 95% confidence interval.

2.11. Metabolic marker analysis

As noted, we used the male mice for Flexivent analysis of lung function; therefore, we could not perform blood analysis from these mice. Metabolic markers and gas analysis was only performed on female mice (n = 5 female vehicle treated, one mouse failed to generated detectable metabolites and n = 6 female DHA treated).

An Arterial Blood Gas (ABG) analysis was performed at the end of the treatment period on female mice to measure blood pH and essential metabolite composition, including glucose and lactate. 1 ml of blood was collected from A/J female mice sub-acutely exposed to DHA for 2 weeks. We used 100 μl of blood in the EPOC blood analysis test card and reader (Siemens Healthineers, Malvern, Pennsylvania, USA) to measure arterial blood gas and metabolites. The remaining blood was added to K2E (K₂EDTA) blood collection microtainer tubes (BD Microtainer Becton Dickinson, Franklin Lakes, NJ, USA), and plasma was collected. To the plasma, a final concentration of 50 µM of the metabolic inhibitor DPPIV (Millipore Sigma, Franklin Lakes, NJ, USA) and a 1X protease cocktail were added for metabolic analysis. Plasma was sent to EVE Technologies for a metabolic hormone discovery assay to be performed. The metabolic panel measured amylin, connecting peptide 2 (C-peptide 2), and total and active levels of glucose-dependent insulinotropic polypeptide (GIP), glucagon-like peptide 1 (GLP-1), ghrelin, glucagon, insulin, leptin, pancreatic polypeptide (PP), peptide tyrosine tyrosine (PYY), resistin, and secretin. Values obtained from EVE technologies were graphed as a heatmap displaying the median values. Graphs displaying individual values are shown as the mean ± standard deviation (SD) in GraphPad Prism. Significance was calculated using a two-way ANOVA with a Benjamini, Krieger, and Yekutieli two-stage step-up false discovery rate (FDR) multiple comparison correction with an FDR of 0.05. The individual graphs report the adjusted p value for this correction.

2.12. Statistical analysis

The biological unit for in vivo analyses was the individual mouse. We have included the mouse number per experiment in the figures legends and methods. For tissue imaging assays, multiple regions of interest or image fields obtained from the same heart were averaged to generate one value per mouse before group level statistical analysis. Image fields were not treated as independent biological replicates. For cell-based assays, n represents independent biological experiments performed using separately seeded and independently treated cultures. Multiple cells or image fields within an experiment were averaged before analysis.

For endpoints measured in both sexes, we used two-way analysis of variance with sex and treatment as fixed factors and included the sex-by-treatment interaction. When the overall model or a relevant interaction was significant, pre-specified pairwise comparisons were performed using Tukey’s multiple-comparisons procedure. Repeated longitudinal body-weight measurements were analyzed using a linear mixed-effects model with fixed effects for treatment, sex, time, and relevant interactions, and mouse-specific random intercepts. For female-only plasma analytes, comparisons were performed using two-way analysis of variance, and multiplicity was controlled using the Benjamini–Krieger–Yekutieli two-stage step-up false-discovery-rate procedure at q = 0.05.

All tests were two-sided, and adjusted p < 0.05 was considered statistically significant. Exact p-values are reported where feasible. Analyses were performed using GraphPad Prism Version 11.0.

3. Results

3.1. In vitro sub-acute exposure to DHA induces oxidative DNA damage and increased mitochondrial DNA

Using the cardiomyocyte cell line Ac16, we previously assessed the cytotoxicity of acute DHA exposures. Fig. 1A shows the dose-response curve for cell growth following a single acute exposure to increasing DHA concentrations. The IC₅₀ dose was determined to be 2.0 ± 0.28 mM [13]. For subacute exposure, we selected 0.2 mM DHA (3.6 µg or 40nmol), which is within physiologically relevant ranges and is non-cytotoxic. Cells were dosed with DHA for 5 days on and 2 days off. After 2 weeks of exposure, we assessed DNA damage using the RADD assay [23]. Our previous studies showed that oxidative DNA lesions were significantly elevated after acute IC₉₀ doses of DHA [13]. Therefore, we examined oxidative lesions using a cocktail of glycosylases specific to oxidative DNA damage (oxRADD). A significant increase in oxidative lesions was observed after 2 weeks of exposure to 0.2 mM DHA (Fig. 1B). We then examined ROS generation in cells using the general oxidative stress indicator CM-H2DCFDA after 2 weeks of exposure (Fig. 1C). Consistent with our previous studies in other cell models, no increase in total cellular ROS was observed after DHA exposure [13].

Fig. 1.

Sub-acute dosing effects of 0.2 mM DHA in Ac16 cardiomyocytes. A. Sub-acute DHA exposure increased DNA damage and mitochondrial alterations in Ac16 cells [14]. Ac16 were exposed to 0.2 mM DHA for 5 days on and 2 days off for 2 weeks. B. Oxidative DNA lesions were assessed by Repair Assisted Damage Detection (oxRADD). C. Cellular reactive oxygen species (ROS) were measured by DCFDA production. D. Mitochondrial ROS was measured using MitoSOX. E. NAD(P)H autofluorescence was measured using two-photon microscopy. F. Mitochondrial DNA (mtDNA) copy number was measured using qPCR and normalized to the β-actin gene. G. Mitochondrial content was measured by immunofluorescence using an antibody specific for the mitochondrial membrane (anti-mitochondria). For all graphs, the mean values are displayed ± the standard error of the mean (SEM). Three biological replicates were used for each experiment (n = 3). Statistical significance was determined by Welch’s t-test; *p < 0.05, ****p < 0.0001. For immunofluorescent images, the scale bar is 10 μm.

We then examined mitochondrial ROS using MitoSOX, because elevated mitochondrial ROS was previously observed in H9c2 cardiomyocytes after acute low doses of DHA [12]. After sub-acute exposure to 0.2 mM DHA for 2 weeks, only a slight, non-significant increase in mitochondrial superoxide formation or other reactive species was observed (Fig. 1D). In our previous work, we noted an increase in antioxidant potential through increased NAD(P)H after DHA exposure [12], [15]. We therefore examined if the ROS generation was being mitigated by the production of NAD(P)H by DHA exposure. Using two-photon microscopy to excite NADH and NAD(P)H within the Ac16 cells, we measured an increase in NAD(P)H autofluorescence, which was borderline significant (p = 0.056, Fig. 1E).

Given the slight increase in mitochondrial ROS and NAD(P)H, we examined changes in mitochondrial copy number and content to determine if DHA was causing other mitochondrial effects. Using qPCR, we examined the relative gene expression of COX1 and ND1 compared to nuclear β-actin after two weeks of exposure. mtDNA copy number was elevated for both genes (p = 0.021 for COX1 and p = 0.020 for ND1) after exposure to DHA (Fig. 1F). Despite the increase in mitochondrial DNA copy number, immunofluorescence assessment of mitochondria revealed no change in mitochondrial content in DHA-dosed Ac16 cells compared with passage matched controls (Fig. 1G). These data demonstrate that sub-acute doses of DHA induce oxidative DNA lesions, increase mitochondrial DNA copy number, but do not significantly induce cellular and mitochondrial oxidative stress.

3.2. In vivo sub-acute exposure to DHA promotes altered body weight in male mice

We previously exposed A/J mice to 5 μg of DHA over 14 days to evaluate lung injury [17]. Additionally, we performed echocardiograms and examined the hearts from these same mice. During DHA exposure, we monitored the animals' body weights at 6, 12 and 14 days after dosing and observed a significant decrease in male body weight over the exposure period (Fig. 2). There was no change in body weight for female mice over the same dosing period.

Fig. 2.

Sub-acute inhaled DHA exposure caused a significant decrease in male body weights. Male and female A/J mice were exposed to 5 μg inhaled DHA for 6 days on, 1 day off. Body weights were measured before dosing and at 6, 12, and 14 days after dosing. Values are displayed as mean ± SEM. Statistical significance was determined by a linear mixed-effects model; **p < 0.01. For males, n = 6 vehicle and n = 4 DHA treated. For females, n = 6 for both groups.

3.3. Sub-acute exposure to DHA promotes sex-specific cardiac function changes and increased strain

We then assessed cardiac function using echocardiograms (ECHOs) after 2 weeks of DHA exposure in both sexes. Males showed a non-significant increase in EF 41.5 ± 19.8% in vehicle treated (mean ± SD) compared to 61.6 ± 5.9% in DHA treated males (Fig. 3B), and a non-significant increase FS (20.4 ± 11.7% vs. 32.2 ± 4.3%, Fig. 3C). Females showed a non-significant decrease in EF (73.4 ± 14.7% vs. 60.5 ± 17%, Fig. 3B) and FS (43.1 ± 13.8% vs. 32.6 ± 12.9%, Fig. 3C). While changes in the mean values for EF and FS were observed for both sexes, DHA exposure did not significantly change EF or FS. The mixed model analysis only detected a significant difference in EF and FS between the two sexes (p = 0.049 and p = 0.047, respectively).

Fig. 3.

Echocardiograms revealed sex specific differences after DHA exposure. Male and female A/J mice underwent echocardiography (ECHO) 2 weeks after exposure to measure ejection fraction (EF, B) and fractional shortening (FS, C). Mean values are displayed as mean ± SD. Significance was calculated using a two-way ANOVA with a Tukey multiple comparison correction with a family-wise alpha of 0.05 and a 95% confidence interval. For males, n = 5 vehicle and n = 4 DHA treated. For females, n = 6 for both groups.

Next, we used the ECHO measurements to calculate global longitudinal and radial cardiac strain after DHA exposure. Despite opposite EF and FS mean changes observed for males and females, both sexes show a significant shift in global longitudinal strain, toward zero, and reduced global radial strain (Fig. 4). In males, global longitudinal strain changed from −9.7 ± 5.9% (mean ± SD) in vehicle treated to −2.9 ± 2.1% in DHA treated hearts (p adjusted = 0.009, Fig. 4A). In males, global radial strain non-significantly decreased from 26.0 ± 24.9% (mean ± SD) in vehicle treated to 8.3 ± 2.9% in DHA treated hearts (Fig. 4B). In females, global longitudinal strain significantly changed from −7.8 ± 3.4% in vehicle treated to −2.7 ± 2.4% in DHA treated hearts (p adjusted = 0.009, Fig. 4A). While global radial strain non-significantly changed from 12.3 ± 5.5% (mean ± SD) in vehicle treated to 5.8 ± 3.4 in DHA treated hearts (Fig. 4B). Mixed model effects analysis showed a significant treatment effect for global longitudinal strain (p = 0.002) compared to global radial strain (p = 0.046). These results demonstrate that sub-acute exposures to DHA increased cardiac strain and began to alter cardiac function.

Fig. 4.

Sub-acute DHA exposure increased global longitudinal and global radial strain in male and female A/J mice. Mean values are displayed as mean ± SEM. Significance was calculated using a two-way ANOVA with a Tukey multiple comparison correction with a family-wise alpha of 0.05 and a 95% confidence interval; **p < 0.01. For males, n = 4 for both groups. For females, n = 6 for both groups.

3.4. Sub-acute DHA exposure altered mitochondria in the hearts

Sections of the mice hearts were examined by TEM to observe heart ultrastructure and mitochondrial content. Images were taken of the perinuclear, intrafibrillar, and sub-sarcolemma mitochondrial regions. Fig. 5A and B show perinuclear and intrafibrillar regions.

Fig. 5.

Sub-acute DHA exposure increased variation in mitochondrial number. Male and female A/J mice hearts were fixed for TEM. Mitochondria were counted across three, 3400 × field from the perinuclear, intrafibrillar, and sub-sarcolemma regions (one field each). Mitochondrial counts per field were averaged for each mouse to provide a total average number. Mean mitochondrial numbers are displayed as mean ± SEM. Significance was calculated using a two-way ANOVA with a Tukey multiple comparison correction with a family-wise alpha of 0.05 and a 95% confidence interval. N = 3 mice per treatment group and sex.

From all images, we counted the number of mitochondria per field and averaged across the three fields to obtain a mean value per mouse. No significant increase in mitochondrial number was observed in both male and female mice after 2 weeks of DHA exposure (Fig. 5C). The mixed model analysis did show a significant effect with treatment (p = 0.02), though the individual adjusted p values for vehicle compared to DHA treatment for each sex were only 0.08.

Additionally, the DHA-exposed hearts were darker and more intensely stained. As a result, sarcomere structures appeared altered in the images, but this may be due to a chemical reaction between DHA and proteins or amine-containing structures within the heart tissue, or to chemical enhancement of the TEM process. DHA undergoes a Maillard reaction, producing a browning effect, which is why it is the active ingredient in sunless tanning products [26], [27]. We therefore only conducted an analysis of the mitochondrial number to avoid changes which may be induced by coloration in the sarcomere or other structures.

We then measured mitochondrial DNA copy numbers for ND1 and COX1 using qPCR. Both sexes showed significant increases in COX1 and ND1, with females showing a greater elevation than males (p < 0.0001, Fig. 6). Sub-acute inhaled DHA exposure altered mitochondrial DNA copy number.

Fig. 6.

Sub-acute DHA exposure increased mitochondrial DNA (mtDNA) copy number. Male and female A/J mouse hearts were used to perform qPCR for mitochondrial genome markers COX1 (A) and ND1 (B). The mtDNA copy number for all genes relative to β-actin was determined. Changes were graphed in GraphPad Prism, and mean ± SEM are shown. Significance was calculated using a two-way ANOVA with a Tukey multiple comparison correction with a family-wise alpha of 0.05 and a 95% confidence interval; ****p < 0.0001. For males, n = 6 vehicle and n = 4 DHA treated. For females, n = 6 for both groups.

3.5. In vivo sub-acute exposure to DHA altered collagen and myosin levels within the heart

Given the mitochondrial DNA changes and increased strain, we used immunohistochemistry and immunofluorescence to examine the heart's structural architecture. Using Masson’s trichrome staining, we measured changes in the collagen content of the hearts (Fig. 7). Male mice showed a slight, but not significant, increase in collagen content, while female mice showed no change. Mixed model analysis revealed the most significant difference between sexes (p = 0.007) and only significant differences in the intensity values for vehicle and DHA treatment between the males and females, respectively (p adjusted = 0.032, Fig. 7B).

Fig. 7.

Male mice have a slight increase in collagen with DHA exposure. Masson’s trichrome staining was performed on hearts exposed to saline vehicle or DHA sub-acutely for 2 weeks to measure collagen deposition. Mean values are displayed as mean ± SEM. Significance was calculated using a two-way ANOVA with a Tukey multiple comparison correction with a family-wise alpha of 0.05 and a 95% confidence interval; *p < 0.05. For males, n = 6 vehicle and n = 4 DHA treated. For females, n = 6 for both groups.

We found no significant change in myosin content in male or female mice (Fig. 8). Early structural changes appear to occur but are not significant after only 2 weeks of exposure.

Fig. 8.

Male A/J mice have a decrease in myosin after DHA exposure. Immunofluorescence was performed on heart tissue to detect myosin. Mean values are displayed as mean ± SEM. Significance was calculated using a two-way ANOVA with a Tukey multiple comparison correction with a family-wise alpha of 0.05 and a 95% confidence interval. For males, n = 6 vehicle and n = 4 DHA treated. For females, n = 6 for both groups.

3.6. In vivo sub-acute exposure to DHA altered metabolic markers in female mice

Finally, we conducted an analysis of metabolic markers in female mice exposed to vehicle and DHA (Fig. 9). Male mice underwent Flexivent analysis for lung function, so their blood gases and plasma metabolites could not be evaluated. The full effect mixed model showed significant interaction between treatment groups and the molecules assessed (p = 0.003). Of the 18 molecules examined (Fig. 9A–C), only glucose showed a significant increase after DHA treatment (p adjusted < 0.0001, Fig. 9C).

Fig. 9.

Female mice had changes in metabolic markers after DHA exposure. Blood gas analysis was used to measure the oxygen, carbon dioxide and pH of the blood. Metabolic hormones in the female plasma also show slight changes after DHA exposure. Values are displayed as a heat map of median values. Resistin changes are displayed as mean ± SEM. Graphs displaying individual values are shown as the mean ± standard deviation (SD) in GraphPad Prism. Significance was calculated using a two-way ANOVA with a Benjamini, Krieger, and Yekutieli two-stage step-up false discovery rate (FDR) multiple comparison correction of 0.05. The individual graphs report the adjusted p value for this correction. Vehicle treated females n = 5 and DHA treated females n = 6.

The full effect mixed model of the metabolites analyzed in the plasma of female mice showed a significant interaction between treatment and metabolites assessed (p = 0.006, Fig. 9D). Of the 12 metabolites assessed, only resistin showed a significant decrease (p adjusted < 0.0001, Fig. 9E).

4. Discussion

DHA is the active ingredient in sunless tanning agents, and the FDA has approved it only for topical use. The discovery that DHA is also contained in e-cigarette aerosol raises significant concerns about inhalation exposures. E-cigarette users are exposed to significant concentrations of DHA [8]. Additionally, spray tanning booths and at home spray tanning applicators can also expose individuals to inhaled DHA at unknown concentrations. While the FDA has warned spray tanning users against inhalation and mucous membrane exposures [2], [28] the effects of inhaled DHA are poorly understood, particularly repeated exposure from e-cigarette use.

Most DHA studies have focused on skin with exposure based on sunless tanning exposure estimates in the high millimolar range [29], [30], [31]. One study used normal human primary tracheobronchial epithelial cells in air-liquid-interface cultures to examine the exposure effects of 0.2, 0.4, and 1.0 M applied acutely or exposed once every 7 days for up to 5 weeks to mimic spray tanning exposures [32]. Cilia beating frequency and MUC5AC (mucin 5AC) secretion initially decreased after exposure [32]. No significant morphological changes were observed in DHA-treated cultures after 5 weekly exposures, and the authors concluded that sporadic exposures to DHA had transient toxic effects on human airway cultures [32].

With e-cigarette exposures occurring more frequently and chronically, the importance of understanding DHA inhalation exposure effects has only grown. While we lack precise dosimetry of DHA exposure via e-cigarettes due to variations in devices and e-liquids, we have employed estimates based on e-cigarette aerosol quantification and smoking behaviors [8], [9], [10], [18], [22], [33]. Using these estimates, we characterized the exposure effects in the lungs of A/J mice exposed acutely to 5, 130, and 600 μg DHA and sub-acutely to 5 μg of DHA for 2 weeks [17]. Inhaled DHA increased inflammatory markers, fibrosis, and altered lung function [17].

Beyond the impact on the lungs, e-cigarettes have been shown to induce cardiac effects ranging from elevated heart rates to heightened sympathetic nervous system activity to dangerous, irregular heart rhythms like ventricular arrhythmias and conduction defects [34], [35], [36]. While nicotine may be a factor in some of these studies, studies on PG:VG alone have also demonstrated cardiac effects, including endothelial dysfunction. A recent systematic analysis of nicotine-free e-cigarette use determined that in human studies, vapers experienced endothelial dysfunction, oxidative stress, increased arterial stiffness, and transient blood pressure elevation [37]. Notable findings included flow-mediated dilation, increased aortic stiffness or pulse-wave measures in some MRI-based studies, and increased central blood pressure [37]. Effects on peripheral blood pressure and heart rates were more mixed across human studies, with fewer reports of arrhythmia.

A review of animal studies using nicotine-free devices showed similar effects, with mitochondrial injury and developmental cardiotoxicity also being observed [37]. PG/VG aerosols and their oxidation products were associated with mitochondrial oxidative stress and electrophysiological disturbances, including ventricular ectopy, slower conduction, altered repolarization, and altered myocardial conduction [37].

Finally, a 2026 study by Rafiei et al. identified sex specific differences in cardiac effects in rats exposed to nicotine-free e-cigarette aerosol [38]. Male rats exposed to nicotine-free aerosol experienced a significant impairment in the balance between oxygen delivery and oxygen demand. Female rats showed a significant increase in myocardial oxygen demand, but they also showed an increase in oxygen delivery, preserving the overall balance [38].

While these effects are often attributed to breakdown products of PG:VG, like acrolein, these aerosols also contain significant amounts of DHA, which can induce oxidative DNA damage, protein damage, and metabolic reprogramming [12], [13], [14], [16]. In examining low, acute exposures to DHA in H9c2 rat cardiomyocytes, we found that non-cytotoxic doses of DHA induce metabolic reprogramming and mitochondrial dysfunction [12].

Therefore, we sought to characterize sub-acute exposures in human cardiomyocytes and in the hearts of the A/J mice exposed sub-acutely to inhaled DHA. For cell dosing, we selected 0.2 mM, which is a fold below the IC₅₀ (2 mM) and corresponds to ∼3.6 μg, which is near the estimated maximum DHA exposure for a 100 ml puff volume [8]. We then dosed the cells for 2 weeks to approximate the inhaled sub-acute exposures conducted in A/J mice. Sub-acutely dosed Ac16 cells showed significantly elevated oxidative DNA damage, indicating that DHA is entering and altering mitochondrial metabolism via triose kinase FMN cyclase (TKFC)-mediated phosphorylation of DHA to DHAP, which then enters multiple metabolic pathways, including glycolysis [27], [39], [40]. These metabolic changes are consistent with the metabolic changes observed after acute exposures to low dose DHA in H9c2 cells [12]. Paradoxically, we did not observe a significant increase in cellular- or mitochondria-specific ROS, although antioxidant molecules NAD(P)H increased (Fig. 1E). We previously noted a lack of cellular oxidative stress in other systemic models, with increases in NAD(P)H likely buffering oxidative stress in the cytosol and nucleus [12], [13], [14], [15], [16]. Additionally, in the Ac16s, we observed a significant increase in mitochondrial DNA. This change may indicate compensatory mtDNA amplification that warrants further investigation. We have consistently observed DHA-mediated stress in mitochondria after acute doses [12], [14], [15].

More importantly, in the hearts of mice sub-acutely exposed to inhaled DHA, we observed mitochondrial and metabolic alterations with sex specific differences. Male A/J mice after 2 weeks of DHA exposure showed significantly reduced body weight. Female A/J mice showed no changes in body weight after exposure (Fig. 2). Body weight reduction in males can influence cardiac dimensions, EF, metabolism, and mitochondrial content. Analysis of metabolites in the blood of female mice showed significant elevation of glucose and resistin (Fig. 9C and E). The male mice were used for Flexivent analysis, so we could not analyze their blood metabolites. However, these results suggest repeated exposure to inhaled DHA induces systemic metabolic changes.

After two weeks of exposure, echocardiograms showed changes in the mean values for EF and FS for males and females, though the overall changes were not significant. Males generally showed increased mean EF and FS after DHA exposure, while females showed decreased mean EF and FS (Fig. 3). While neither group showed significant changes, after only 2 weeks of exposure to 5 μg of DHA, we observed alterations in cardiac function parameters that may reflect early cardiac remodeling and warrant investigation in longer duration exposure studies, especially between the sexes. More significant effects were confirmed by strain analysis; both sexes showed increased strain after sub-acute exposure to DHA, with significant increases in global longitudinal strain observed for both sexes (Fig. 4A).

We then examined the heart tissues for structural changes using TEM and immunohistochemistry. Analyzing mitochondrial content in TEM images, we observed a slight increase in mitochondrial numbers in males and females after DHA exposure (Fig. 5). These changes were confirmed by increased mitochondrial copy number, with significant increases in ND1 and COX1 in both sexes (Fig. 6). Similar to the exposed Ac16, these data suggest DHA is incorporated into metabolic pathways and alters mitochondrial metabolism.

Immunohistochemistry and immunofluorescence also showed subtle changes in collagen levels (Fig. 7). Male A/J mice showed a slight increase in collagen levels, while females were unchanged [41], [42].

Taken together, these in vivo and in vitro findings suggest that low dose, sub-acute exposures to DHA induce early changes in cardiac tissue consistent with reports of e-cigarette aerosol exposures [34], [35], [36], [43]. This study has several limitations. First is the estimation of DHA exposure doses. Due to the lack of quantitative data on the inhalation of DHA from spray tanning and e-cigarette use, we have tried to model exposures based on measured ranges of DHA from mass spectrometry studies, self-reported vaping behaviors, and puff topography reports [8], [9], [18], [22], [44]. We conservatively selected the 5 μg DHA dose to cover the lowest potential exposure for spray tanning and e-cigarette exposure. While these estimates may have significant errors, particularly due to self-report bias, they are consistent with the European Union estimates for one-time exposures from spray tanning booths of 0.21–0.6 mg and represent a reasonable starting point [20]. To study the exposure effects of DHA alone, we exposed animals to DHA in saline, which is another limitation. The produced DHA would be delivered with unreacted PG:VG and other additives in e-cigarettes. We studied DHA in isolation without other carrier molecules, which may alter its absorption or distribution.

Another limitation was the availability of blood for analysis in this work. We are lacking some blood characteristics in the male mice due to their use in Flexivent experiments in our previous report [17]. We have sought to minimize animal exposure by using the available tissue and blood for analysis. Given the changes observed at this low dose after 2 weeks of exposure to 5 μg, future research should focus on longer, longitudinal studies with potentially higher doses.

The final limitation is the use of Ac16 cardiomyocyte cells to confirm specific exposure effects and potential mechanisms in vitro. These are immortalized cells and lack a contractile, beating phenotype. We previously characterized H9c2 cells, which do have a contractile phenotype, and the results observed here are consistent with that low dose acute study showing DHA induced metabolic effects and mitochondrial changes [12].

Despite these limitations, sub-acute DHA exposure was associated with increased mtDNA copy number in cardiac tissue and altered global longitudinal strain in the hearts of A/J mice, alongside oxidative DNA damage and increased mtDNA copy number in Ac16 cells. DHA may be significant but overlooked contributor to the cardiac effects observed in population and animal studies using PG:VG e-liquids and potentially complete e-cigarette aerosol. These findings have important implications for human health because DHA is present in all e-cigarette aerosols using glycerol as an e-liquid, regardless of whether it contains nicotine or other additives. These findings are also important because DHA is also used extensively in commercial sunless tanning products, including aerosolized salon and in home spray tans. While the present study evaluated only a short-term exposure period, the observed increases in cardiac strain and mitochondrial DNA abundance suggest that repeated inhalation exposure may affect cardiac metabolism and function before overt structural disease becomes detectable. Given the widespread use of vaping products and spray tanning applications, longer-term studies examining chronic exposure, cumulative cardiac injury, and potential susceptibility factors are warranted.

CRediT authorship contribution statement

Natalie R. Gassman: Writing – review & editing, Writing – original draft, Visualization, Supervision, Resources, Project administration, Funding acquisition, Formal analysis, Data curation, Conceptualization. Saurabh Aggarwal: Writing – review & editing, Supervision, Resources, Funding acquisition, Conceptualization. Jenna Hedlich-Dwyer: Writing – review & editing, Methodology, Investigation. Arlet Hernandez: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Hailey J. Levi: Writing – review & editing, Writing – original draft, Visualization, Methodology, Investigation, Formal analysis, Data curation.

Funding

NIH/NIEHS R01 ES032450 supported this study. HL was supported by R01 ES032450-04S1, and AH was supported by CCTS TL1TR003106. SA is supported by U01 ES033265.

Competing interests

The authors declare no competing interests. This research was funded in part by the National Institute of Health. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Handling Editor: Dr. L.H. Lash

Data availability

Data will be made available on request.