The e-liquid flavoring cinnamaldehyde induces cellular stress responses in human proximal tubule (HK-2) kidney cells

Abstract

Flavored e-liquid use has become popular among e-cigarette users recently, but the effects of such products outside the lung are not well characterized. In this work, acute exposure to the popular flavoring cinnamaldehyde (CIN) was performed on human proximal tubule (HK-2) kidney cells. Cells were exposed to 0–100 μM CIN for 24–48 h and cellular stress responses were assessed. Mitochondrial viability via MTT assay was significantly decreased at 20 μM for 24 and 48 h exposure. Seahorse XFp analysis showed significantly decreased mitochondrial energy output at 20 μM by 24 h exposure, in addition to significantly reduced ATP Synthase expression. Seahorse analysis also revealed significantly decreased glycolytic function at 20 μM by 24 h exposure, suggesting inability of glycolytic processes to compensate for reduced mitochondrial energy output. Cleaved caspase-3 expression, a mediator of apoptosis, was significantly increased at the 24 h mark. C/EBP homologous protein (CHOP) expression, a mediator of ER-induced apoptosis, was induced by 48 h and subsequently lost at the highest concentration of 100 μM. This decrease was accompanied by a simultaneous decrease in its downstream target cleaved caspase-3 at the 48 h mark. The autophagy marker microtubule-associated protein 1 A/1B light chain 3 (LC3B-I and LC3B-II) expression was significantly increased at 100 μM by 24 h. Autophagy-related 7 (ATG7) protein and mitophagy-related proteins PTEN-induced putative kinase 1 (PINK1) and PARKIN expression were significantly reduced at 24 and 48 h exposure. These results indicate acute exposure to CIN in the kidney HK-2 model induces mitochondrial dysfunction and cellular stress responses.

1. Introduction

In 2023, the National Youth Tobacco Survey reported e-cigarettes as the most used tobacco product among middle (4.6 %) and high school (10.0 %) students [1]. These devices, also known as electronic nicotine delivery systems (ENDS), use e-liquids which allow for a non-combustible delivery of heated aerosol to the user. E-liquids often include nicotine, flavorings, and solvents such as propylene glycol (PG) and vegetable glycerin (VG) in order to produce a flavorful puff to the user. Although the Food and Drug Administration (FDA) enacted a ban on certain flavored e-cigarette cartridges in 2020 [2], approximately 1 in 4 high-school seniors have reported vaping use within the past 30 days according to the Monitoring the Future national survey, with approximately 1 in 12 reporting vaping just flavoring only with no nicotine [3]. Flavors in e-liquids have been reported to enhance the appeal of and willingness to vape, as well increase the likelihood of continued vaping and taking more puffs per vape long-term [4–6].

E-liquid ingredient safety is currently maintained under the “generally recognized as safe” (GRAS) classification for food additives. This presents a problem in determining the safety of e-liquid ingredients due to the inhalation route of administration. E-liquids which contain flavoring aldehydes such as cinnamaldehyde, vanillin, and ethyl vanillin can become chemically unstable, which may lead to acetal formation that can persist through inhalation and cause activation of the transient receptor potential (TRP) ion channels that act as receptors for irritant aldehydes in airway [7]. In addition, heating of e-liquids containing PG/VG has been shown to generate toxic compounds such as acrolein, acetaldehyde, and formaldehyde [8,9]. Individual flavors also present a unique problem when attempting to assess the safety of e-liquid products. Many cinnamon-flavored products contain cinnamaldehyde (CIN, Fig. 1), which is a reactive alpha, beta-unsaturated aldehyde. CIN has been shown to cause cytotoxicity and genotoxicity in human embryonic stem cells (hES) and human pulmonary fibroblasts (hPF), as well as destabilize growth, attachment, motility, and general cell morphology [10,11]. Lung models also show exposure to CIN can disrupt mitochondrial and ciliary function and impair innate immune cell function [12,13].

Fig. 1. Cinnamaldehyde chemical structure.

Cinnamaldehyde (CAS 104–55–2) is an α,β-unsaturated aldehyde used as a flavoring agent.

The lung is naturally well-researched in terms of the effects of e-liquids given that it is the first major organ to encounter the vapor from such products. Research into the systemic effects outside the lung, however, is not as robust and little is known about long-term effects. The kidney is responsible for filtration of the blood and receives a large amount of blood flow from the heart and should encounter many of the e-liquid components as they are being filtered out of the body. Yet very little is understood about how the kidney responds to e-liquid exposure. Intraperitoneal (i.p.) exposure in rats to flavored e-liquids both with and without nicotine has been shown to decrease renal oxidative stress defense enzymes while increasing renal proteins, thiol groups, and darkened/condensed nuclei with reduced cytoplasm within the collecting ducts [14]. Additionally, mice exposed to high-fat diet (HFD) in conjunction with vaped e-liquids show altered renal mitochondrial oxidative phosphorylation (OXPHOS) complex expression, and altered inflammatory, oxidative stress, and pro-fibrotic markers [15].

In order to address the gaps in knowledge regarding the effects of CIN on renal function, this study exposed human proximal tubule (HK-2 cells to CIN and determined changes on cellular stress markers and energy processes. Markers for apoptosis, lipid peroxidation-derived oxidative stress, endoplasmic reticulum (ER), and autophagy/mitophagy were evaluated by western blot. Glycolytic and mitochondrial energy processes were quantified using a Seahorse XFp Analyzer after exposure to CIN. Additionally, MTT assay was performed on cells to assess mitochondrial health. Finally, trypan blue exclusion was performed on samples after CIN exposure to determine cell viability and membrane integrity.

2. Materials and methods

2.1. Chemicals and reagents

Cinnamaldehyde was obtained from Sigma Aldrich, St. Louis, MO, Item No. W228613. Dimethyl sulfoxide (DMSO) was used as a vehicle to dilute CIN treatments and purchased from Fisher Scientific, Fair Lawn, NJ, Item No. D128–1.

2.2. Cell culture

HK-2 cells were purchased from American Type Culture Collection (ATCC, Manassas, VA, CRL-2190) and cultured according to ATCC guidelines. Cells were cultured in keratinocyte-free media with 50 μg/mL bovine pituitary extract and 5 ng/mL recombinant epithelial growth factor purchased from Thermo Fisher Scientific (Gibco, Carlsbad, CA, Item No. 17005–04). Penicillin-Streptomycin (0.5 %, 50 Units/mL) (Gibco, Grand Island, NY, Item No. 15140–122) was added to each flask to inhibit bacterial growth. Cells were grown in T75 tissue culture flasks (USA Scientific, Cyto One, Item No. CC7682–4875) at a concentration of 6.0 ×10⁶ cells per flask. Cells were grown in a warm, humidified incubator at a temperature of 37° C and 5 % CO₂.

2.3. Trypan blue exclusion

Trypan blue exclusion was used to assess cell viability and membrane integrity of HK-2 cells exposed to CIN. All experiments were conducted as 4 independent studies each with different cell passages. The vehicle control were treated with an equal volume of DMSO. The amount of DMSO added with or without cinnamaldehyde was 1 % of total volume in each well. Cells were treated with 0–100 μM CIN for 24 or 48 h and then collected. Collected cells were aliquoted and diluted in a 1:1 with 40 % w/v trypan blue solution (Sigma Aldrich, Item No. T6146). The diluted cell solution was gently pipetted and a 10 μL aliquot was transferred to a hemocytometer and the count was done using a Countess II FL cell counter (Thermo Fisher Scientific). Total, living, and dead cells were counted as a measure of cell viability.

2.4. MTT assay

3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) was purchased from Sigma Aldrich, St. Louis, MO, Item No. 1002120223. HK-2 cells were plated on 96-well culture plates (Fisher Scientific, Item No. FB012931) at a density of 5 ×10⁴ cellsmL and allowed to equilibarte for 48 h. Following equilibration, fresh media was added and cells were treated with 0–100 μM CIN for 24 or 48 h. After incubation with CIN, MTT (5 mg/mL) reagent was added to each well and cells were incubated in darkness for 4 h at room temperature with gentle rocking. After 4 h, all liquid volume was aspirated from each well and 100 μL DMSO was added. The MTT assay relies on conversion of the MTT dye to a purple-formazan color by NAD(P)H-dependent oxidoreductases. Plates were read using a BioTek plate reader (Gen5 software, version 3.14.03) at 570 nm.

2.5. Western blot analysis

Cells were plated onto 6-well plates (USA Scientific, Cyto One, Item No. 353046) at a density of 1.0 ×10⁶ cells per well and allowed to equilibrate for 48 h. Following equilibration, fresh media was added and cells were treated with 0–100 μM CIN for 24 or 48 h. After CIN treatment, cells were collected using trypsin (Gibco, Grand Island, NY, Item No. 25200–072), washed in Krebs solution (1.85 g NaCl, 96.3 mg KCl, 79.5 mg MgSO₄·7H₂O, 213 mg Na₂HPO₄, 37.3 mg CaCl₂·H₂O/250 mL dH₂O), and lysed (Cell Signaling Technology, Danvers, MA, Item No. 9803). Western blot analysis was done to assess the expression of the following: 4-hydroxynonenal (1:1000; 4HNE, EMD Millipore, Billerica, MA, Item No. ABN249), autophagy-related 7 (ATG7, 1:1000; Abcam, Waltham, MA, Item No. 133528), C/EPB homologous protein (CHOP, 1:1000; Cell Signaling Technology, Danvers, MA, Item No. 2895 S), microtubule-associated protein 1 A/1B light chain 3 (LC3B, 1:1000; Abcam, Waltham, MA, Item No.48394), mitochondrial oxidative phosphorylation complexes (OXPHOS, 1:1000; Abcam, Cambridge, MA, Item No. 110413), p62 (1:1000; Abcam, Item No. 56416), Parkin (1:500; Abcam, Item No. 77924), PINK1 (1:1000; Abcam, Waltham, MA, Item No. 300623; Novus Biologicals, Centennial, CO, BC100–494; Fisher Scientific, Invitrogen, Rockford, IL, PA1–16604), and pro-caspase 3 and cleaved caspase-3 (1:1000; Abcam, Cambridge, MA 136812). Bradford assay [16] was used to determine protein concentration in each sample prior to loading. A 35 μg aliquot of each sample was diluted to 20 μL in double-distilled H₂O (ddH₂O), followed by addition of reducing sample buffer (RSB). All samples except those used for OXPHOS analysis were boiled for 5 min prior to loading. Protein samples were separated on a 12.5 % polyacrylamide gel and transferred to nitrocellulose membrane (Bio-Rad, Hercules, CA, Item No. 1620097). Successful transfer and protein loading were verified using the Memcode Reversible Stain Kit (Fisher Scientific, Pierce Biotechnology, Rockford, IL, Item No. PI-24580). 4-HNE membranes were blocked for 1 h using 1 % w/v bovine serum albumin (BSA) in TBST (10 mM Tris-HCl, 150 mM NaCl, 0.1 % Tween-20; pH 8.0). Parkin membranes were blocked for 1 h using 3 % milk w/v in TBST. All other membranes were blocked for 1 h using 5 % milk w/v in TBST. Membranes were incubated with primary antibodies overnight in blocking solution at 4° C with gentle rocking. Membranes were then washed with TBST 4x for 5–10 min, followed by appropriate secondary antibody (goat anti-rabbit HRP, 1:2000; Cell Signaling Technology, Danvers, MA, 7074; goat anti-mouse HRP, 1:2000; Cell Signaling Technology, Danvers, MA, Item No.7076S) for 1–1.5 h according to manufacturer’s instructions. After secondary antibody, membranes were washed again 3x for 5–10 min with TBST and again 1x for 5–10 min in TBS (10 mM Tris-HCl, 150 mM NaCl, pH 8.0). Enhanced chemiluminescence (ECL) was used to develop membranes using a diluted 2:1:1 mixture of H₂O₂ (30 % v/v), coumaric acid (90 mM), and luminol (250 mM) in 1 M Tris, pH 8.5. Finally, membrane imaging and densitometry analysis was conducted using a Bio-Rad chemic-doc system (Image Lab version 6.0.1, Bio-Rad, Hercules, CA, Item No. 170–9690). All gel images were normalized to total protein scan for each lane from corresponding Memcode scan.

2.6. Seahorse XFp analysis

The Agilent Seahorse XFp analyzer allows for real-time measurement of oxygen consumption rate (OCR) and extracellular acidification (ECAR) in cultured cells. Glycolytic and mitochondrial function were measured using cell mito and glycolysis stress test kits (Agilent, Cedar Creek, TX, Item No. 103010–100 and 103017–100). HK-2 cells were cultured at a density of 1.5 ×10⁴ cells per well in XFp culture miniplates (Agilent Technologies, Cedar Creek, TX, Item No. 102984–100) and allowed to grow for 48 h. Following equilibration, fresh media was added and cells were treated with 0, 10, or 20 μM CIN for 24 h. Prior to the cell mito stress test assay, cells were washed 2x with pre-warmed assay media (Agilent Technologies, Santa Clara, CA, 103575–100.) supplemented with 2 mM glutamine, 1 mM pyruvate, and 10 mM glucose ((Agilent, Santa Clara, CA Item Nos. 103579–100, 103578–100, and 103577–100). For glycolysis stress tests, only 2 mM glutamine was added to the assay medium. All wells were incubated in 175 μL assay media for 1 h at 37° C in an incubator with 0 % CO₂. The cell mito stress test kit included the following reagents that are injected over the course of the assay to assess OCR and mitochondrial function: oligomycin (1.5 μM/well), carbonyl cyanide-4-(trifluoromethoxy)phenylhydrazone (FCCP, 0.5 μM/well), and rotenone/antimycin-A (0.5 μM/well). The glycolysis stress test kit included the following reagents that are injected over the course of the assay to assess ECAR and glycolytic function: glucose (10 mM), oligomycin (1.0 μM), and 2-deoxy-glucose (2-DG, 50 mM). Basal respiration and post-injection measurements of OCR and ECAR were taken every 3 min throughout the assay. Cells were incubated at 37° C in an incubator with 0 % CO₂ in media deprived of glucose and pyruvate for approximately 1 h prior to start of glycolysis stress test assay. After all assays, wells were gently washed with Krebs solution 2x and plates were frozen and stored at −80°C.

2.7. Cell normalization

Cell mito and glycolysis stress test assays were normalized using CyQUANT Direct Cell Proliferation Assay (Invitrogen Thermo Fisher Scientific, Life Technologies, Eugene, OR, Item No. C35011). Plates were removed from −80°C freezer and thawed to reach room temperature. In dark conditions, a working solution (WS) of cell lysis buffer (200 μL), ddH₂O (3790 μL) and CyQUANT GR dye (10 μL) was prepared and gently mixed. To each well, 200 μL of WS added and incubated for 5 min in darkness. After incubation, 175 μL of each sample well was transferred to a black-walled, clear-bottomed 96-well plate (Fisher Scientific, Greiner Bio-One, Item No. 655906). Fluorescence was measured at 480 and 520 nm using a BioTek plate reader (Gen5 software, version 3.14.03).

2.8. Statistical analysis

All values are represented as Mean ± SEM with a minimum of 4 independent experiments using different biological replicates. Differences between groups were determined using a one-way ANOVA followed by a Tukey post-hoc test with values of p < 0.05 considered as significant (GraphPad Prism Software, version 10.1.0).

3. Results

3.1. CIN effects on cell and mitochondrial health

MTT, 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, is converted to an insoluble formazan dye as an indicator of mitochondrial viability. Exposure to 0–100 μM CIN for 24 h and 48 h showed significantly reduced mitochondrial activity starting at 20 μM (p < 0.0001) and continued to decrease until the highest concentration (Fig. 2A, B). Following MTT assay, trypan blue exclusion was performed as an indicator of cell membrane integrity. Interestingly, no significant differences in cell viability were observed after trypan blue exclusion was performed (Fig. 2C, D).

Fig. 2. Cinnamaldehyde effects on cell and mitochondrial viability.

CIN diminished significantly mitochondrial activity at 24 (A) and 48 h (B) (p < 000.1) starting at 20 μM. Statistical difference from 0 μM (DMSO) vehicle control group is indicated by different superscript letters (a-g). Trypan blue exclusion showed no significant differences between control and treatment groups (C, D). One-way ANOVA was used for statistical analysis followed by a Tukey post-hoc test. Values represent mean ± SEM for a minimum of 4 independent experiments using different biological replicates.

3.2. CIN effects on mitochondrial function

Mitochondrial function was assessed using a Seahorse XFp assay machine and a cell mito stress test kit. This kit utilizes oligomycin, FCCP, and rotenone/antimycin-A in order to determine oxygen consumption rates over time. Oligomycin inhibits ATP synthase to determine ATP-linked oxygen consumption, while FCCP uncouples oxidative phosphorylation to induce maximum oxygen consumption. Rotenone and antimycin-A inhibit complexes I and III, respectively, which inhibits mitochondrial respiration to determine non-mitochondrial oxygen consumption. At 24 h, exposure 20 μM CIN significantly reduced ATP production, maximal respiration, and spare respiratory capacity compared to control (Fig. 3A–C). Basal respiration showed a decreasing trend (Fig. 3D), but the difference was not significant compared to control. Non-mitochondrial respiration, proton leak, and coupling efficiency were not significantly altered after 24 h (Fig. 3E–G). Fig. 3H shows a representative time course profile of the cell mito stress test after 24 h exposure to CIN.

Fig. 3. Cinnamaldehyde effects on mitochondrial function.

At 24 h, 20 μM CIN significantly decreased ATP production, maximal respiration, and spare respiratory capacity when compared to control (A-C). No significant differences in basal respiration, non-mitochondrial respiration, proton leak, or coupling efficiency were observed (D-G). Representative time course profile of a Seahorse cell mito stress test following exposure to CIN for 24 h (H). Total number of cells was measured using the CyQUANT Direct Cell Proliferation Assay and results were normalized to number of cells. Statistical differences denoted by an asterisk (* p < 0.05, ** p < 0.01, *** p < 0.001). Values represent mean ± SEM for a minimum of 4 independent experiments using different biological replicates.

3.3. CIN effects on glycolytic function

Glycolytic function was assessed using a Seahorse XFp assay machine and a glycolytic stress test kit. This kit includes glucose, oligomycin, and 2-DG in order to determine extracellular acidification rate (ECAR). Cells were deprived of glucose and pyruvate in assay medium prior to start of the assay. The first injection of glucose acts to stimulate glycolysis and generation of protons causing a rapid increase in ECAR, and this response is considered the rate of glycolysis at basal condition. The injection of oligomycin inhibits ATP production by the mitochondria, which shifts production of ATP to glycolysis and acts to demonstrate maximum glycolytic capacity. Finally, injecting 2-DG acts to inhibit glycolysis which decreases ECAR and allows for confirmation that ECAR produced during the assay is due to glycolysis. CIN significantly decreased glycolysis, glycolytic capacity, and glycolytic reserve after 24 h (Fig. 4A–C). Non-glycolytic acidification was reduced at 24 h, but not significantly (Fig. 4D). A representative time course profile of the glycolysis stress test is shown in Fig. 4E.

Fig. 4. Cinnamaldehyde effects on glycolytic function.

After 24 h, 20 μM CIN significantly decreased glycolysis, glycolytic capacity, and glycolytic reserve when compared to control (A-C). Non-glycolytic acidification was decreased, but not significantly (D). Representative time course profile of a Seahorse cell glycolysis stress test following exposure to CIN for 24 h (H). Total number of cells was measured using the CyQUANT Direct Cell Proliferation Assay and results were normalized to number of cells. Statistical differences denoted by an asterisk (** p < 0.01, *** p < 0.001, **** p < 0.0001). Values represent mean ± SEM for a minimum of 4 independent experiments using different biological replicates.

3.4. CIN effects on mitochondrial OXPHOS complex expression

Based on the results of Seahorse mitochondrial assays, expression of the mitochondrial oxidative phosphorylation complexes (OXPHOS) was probed using western blotting. After 24 and 48 h, exposure to CIN significantly reduced expression of OXPHOS complex V (ATP Synthase) (Fig. 5A–B). In addition, OXPHOS complexes I and II also were significantly reduced at 48 h, but only between non-control groups and 100 μM CIN. (Fig. 5C–D).

Fig. 5. Cinnamaldehyde effects on mitochondrial OXPHOS complexes.

CIN exposure significantly reduced complex V expression at 24 and 48 h (A-B). Expression of complexes I and II were significantly reduced only between non-control groups and 100 μM (C-D). Results were normalized to total protein concentration. The positive control was rat heart lysate. Statistical differences denoted by an asterisk (* p < 0.05, ** p < 0.01). One-way ANOVA was used for statistical analysis followed by a Tukey post-hoc test. Values represent mean ± SEM for a minimum of 4 independent experiments using different biological replicates.

3.5. CIN effects on apoptosis, oxidative stress, and endoplasmic reticulum stress markers

Several cellular stress pathway markers were also probed via western blotting to determine the effects of CIN on HK-2 cells. Oxidative stress was examined by probing for 4-hydroxynonenal (4-HNE), which is generated via lipid peroxidation and subsequently attaches to proteins within the cell. No significant 4-HNE formation was observed after 24 or 48 h exposure to CIN (data not shown). Endoplasmic reticulum (ER) stress was also examined by probing for C/EBP homologous protein (CHOP), which acts as a pro-apoptotic transcription factor within the ER. Damaged or unfolded proteins will initiate the unfolded protein response (UPR) pathway within the ER, however the damage is unable to corrected then CHOP expression is triggered to initiate the apoptotic pathway. After 24 h exposure to CIN, no CHOP signal was detected (data not shown, example image Fig. 6B). However, by 48 h CHOP expression was triggered and then significantly decreased at the highest concentration of 100 μM CIN (Fig. 6A). Fig. 6B shows the comparison of CHOP expression at 24 h vs. 48 h. In addition to CHOP, pro-caspase 3 and cleaved caspase-3 were probed in order to assess effects on the apoptosis pathway. No significant changes in pro-caspase 3 were observed (data not shown). However, cleaved caspase-3 was significantly increased after 24 h of CIN exposure (Fig. 6C). Interestingly, cleaved caspase-3 expression decreased after 48 h exposure to CIN (Fig. 6D), although the decrease was not considered significant (p = 0.07). Taken together, it could be possible that CIN induces the loss of CHOP expression by the 48 h time point, which leads to a decrease in its downstream target cleaved caspase-3.

Fig. 6. Cinnamaldehyde effects on ER stress and apoptotic pathway.

CHOP expression was triggered by 48 h exposure to CIN, and significantly decreased at 100 μM (A). Comparison of CHOP expression at 24 h vs. 48 h (B). Cleaved caspase-3 significantly increased after 24 h exposure to CIN (C), however this expression was lost by 48 h (D). Results were normalized to total protein concentration. The +CON for CHOP was C2C12 cells treated with thapsigargin. The +/−CON for cleave caspase 3 were HeLa cells treated with or without staurosporine. Statistical differences denoted by an asterisk (* p < 0.05, ** p < 0.01, *** p < 0.001). One-way ANOVA was used for statistical analysis followed by a Tukey post-hoc test. Values represent mean ± SEM for a minimum of 4 independent experiments using different biological replicates.

3.6. CIN effects on the autophagy-related protein expression

The effects of CIN on the autophagy pathway were also examined. Microtubule-associated protein 1 A/1B light chain 3 (LC3B) is often used as a marker of autophagy and is found in two forms within the cell. The cytosolic and soluble form, LC3B-I, is conjugated with phosphatidylethanolamine (PE) and attaches to the autophagosome membrane during autophagy. This conjugated form is known as LC3B-II and is required for autophagosome formation. For this reason, the ratio of LC3B-II to LC3B-I can be an indication of autophagy within the cell. After 24 and 48 h exposure to CIN, both LC3B-I and LC3B-II expression increased significantly (Fig. 7A–B, D–E). The ratio of LC3B-II to LC3B-I also significantly increased after 24 h exposure to CIN (Fig. 7C). While this ratio did increase after 48 h, it was not statistically different (Fig. 7F). Fig. 7G shows a comparison of LC3B-I and LC3B-II expression at 24 and 48 h.

Fig. 7. Cinnamaldehyde effects on autophagy marker LC3B.

CIN significantly increased LC3B-I and LC3B-II expression at 24 and 48 h (A-B, D-E). The ratio of LC3B II to LC3B-I was significantly increased at 24 h, and an increasing but non-significant trend was observed at 48 h (C, F). Comparison of LC3B-I and LC3B-II expression at 24 and 48 h (G). Results were normalized to total protein concentration. The −/+CON were HeLa cells treated with DMSO or staurosporine. Statistical differences denoted by an asterisk (* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001). One-way ANOVA was used for statistical analysis followed by a Tukey post-hoc test. Values represent mean ± SEM for a minimum of 4 independent experiments using different biological replicates.

An increase in LC3B ratio alone is often not enough to confirm the occurrence of autophagy within the cell. Because of this, both autophagy-related 7 (ATG7) and p62 proteins were also probed in order to confirm that autophagy processes were occurring after CIN treatment. ATG7 plays a role in the conjugation of PE to LC3B-I, generating LC3B-II. In addition, the p62 protein marks cell cargo waste destined for breakdown by the autophagic process. After 24 h, ATG7 expression significantly decreased only between non-control groups and the highest concentration of 100 μM CIN (Fig. 8A). A decreasing trend between control and 100 μM was observed, but it was not considered significant (p = 0.07). After 48 h exposure to CIN, ATG7 expression was significantly decreased between control and 100 μM CIN (Fig. 8B). No significant differences in p62 expression were observed at 24 or 48 h (data not shown).

Fig. 8. Cinnamaldehyde effects on autophagy-related 7 protein.

After 24 h exposure to CIN, expression of ATG7 significantly decreased between non-control groups and 100 μM CIN (A). ATG7 expression significantly decreased at 48 h between control and 100 μM (B). Results were normalized to total protein concentration. HeLa cell lysate from cells treated with staurosporine was used as the positive control. Statistical differences denoted by an asterisk (* p < 0.05, ** p < 0.01). One-way ANOVA was used for statistical analysis followed by a Tukey post-hoc test. Values represent mean ± SEM for a minimum of 4 independent experiments using different biological replicates.

3.7. CIN effects on mitophagy-related protein expression

Mitophagy is a form of autophagy that degrades damaged mitochondria. Given that CIN seems to cause deleterious effects within the mitochondria, two important mitophagy-related protein markers were probed. PTEN-induced putative kinase 1 (PINK1) is a 64 kDa serine-threonine kinase targeted to the mitochondrial membrane that is continuously cleaved under normal conditions within the cell into 60- and 52- kDa fragments. However, if the mitochondria is damaged, uncleaved PINK1 will begin to accumulate on the mitochondrial membrane and subsequently recruit PARKIN, which is a key mediator in the mitophagy process. Exposure to CIN for 24 and 48 h significantly reduced PINK1 and PARKIN expression when compared to control groups (Fig. 9A–D). It is also important to note that the cleaved fragments of PINK1 are still visible at the 24 h mark (Fig. 9A bottom), possibly indicating that normal function of the mitochondria is still occurring. However, by 48 h this cleavage is decreased (Fig. 9C bottom) which indicates a possible loss of mitochondrial function.

Fig. 9. Cinnamaldehyde effects on mitophagy-related proteins PINK1 and PARKIN.

Both 24 and 48 h exposure significantly reduced expression of PINK1 and PARKIN (A-D). PINK1 cleavage (A, bottom) at 24 h is decreased by 48 h mark (C, bottom). Results were normalized to total protein concentration. PINK1 positive control was HeLa cell lysate treated with staurosporine while mouse brain lysate was used as the positive control for Parkin Statistical differences denoted by an asterisk (* p < 0.05). One-way ANOVA was used for statistical analysis followed by a Tukey post-hoc test. Values represent mean ± SEM for a minimum of 4 independent experiments using different biological replicates.

4. Discussion

The results of these experiments indicate that acute exposure to CIN decreases mitochondrial function and energy output of HK-2 cells. In addition, expression of the pro-apoptotic marker cleaved caspase-3 is significantly increased as well by 24 h exposure. Expression of CHOP, an upstream signaler of cleaved caspase-3, is lost by the 48 h mark which also coincided with loss of cleaved caspase-3 expression. Finally, several markers of autophagy and mitophagy are altered after CIN exposure. These changes, however, did not translate into loss of cell membrane integrity based on trypan blue exclusion, indicating that the kidney HK-2 cell model was impaired by acute CIN exposure in the present study.

One of the most notable changes after CIN exposure was related to energy output and function in the mitochondria and extramitochondrial. The kidney is a highly dynamic organ that requires a substantial amount of ATP energy to function, with the majority of that energy being used for transport processes within the tubules [17,18]. Here we show a significant decrease in ATP output within the mitochondria that also corresponded to an observable significant decrease in ATP synthase expression. This also led to a decrease in overall maximal respiration and spare capacity, indicating that the HK-2 cells would have a decreased ATP energy pool to use should they require more energy. AKI is often associated with lower oxygen availability, and proximal tubule cells will rely more on glycolysis for energy purposes [19]. Our findings showed that glycolytic function was significantly decreased by the 24 h time point. This indicates that the glycolytic pathway was not able to compensate for loss of energy output by the mitochondria. While there is no definitive evidence to date that suggests flavored e-liquids can induce kidney injury, it is important to realize that long-term and frequent use of such products should be monitored and classified for possible hazardous health effects.

The unfolded protein response (UPR) is responsible for maintaining protein folding within the ER, but prolonged UPR response will eventually trigger the pro-apoptotic pathway via CHOP [20,21]. The expression of CHOP by the 48 h time point suggests that exposure to CIN induced the UPR, but prolonged activation eventually led to the pro-apoptotic pathway activation via CHOP itself. However, this activation of CHOP was lost at the highest concentration of 100 μM CIN. Cleaved caspase-3 is downstream target of CHOP that mediates apoptosis, and its expression also decreased at the highest CIN concentration. Taken together, this data suggests that loss of CHOP expression leads to a loss of cleaved caspase-3 expression.

In the present study, both proapoptotic and autophagy markers were detected after CIN exposure. It is important to remember that both autophagy and apoptosis are interconnected pathways in the cellular environment, with each process influencing the other [22,23]. In terms of kidney health, autophagy has been shown to have a protective effect in cisplatin-induced acute kidney injury (AKI) model in tubular cells [24]. In our experiment the autophagy marker LC3B did significantly increase by 24 h, however p62 and ATG7 did not change or decreased significantly, respectively. While autophagy may be initially activated after CIN exposure, the changes in expression of ATG7 suggest a possible loss of autophagic maintenance. Since these experiments were aimed at acute exposure, longer exposure may be necessary to determine more definitive effects on the autophagy pathway. Loss of PINK1 and PARKIN expression also suggests a decrease in mitophagy after CIN exposure. Based on these results, early activation of the autophagy pathway may occur followed by loss of autophagy and mitophagy-related mediators. CIN is a reactive alpha, beta-unsaturated aldehyde, and is relatively lipophilic. The reactive nature of CIN could potentially lead to interactions with these markers, altering their function.

We recognize that there are several limitations to the present study. First and foremost, direct application of the CIN flavoring to the HK-2 cells in culture does not examine modifications of CIN that may occur in other organs prior to delivery to the kidney. Also, as stated previously, stability on the shelf and heating of e-liquids does change their composition prior to inhalation. In addition, once inhalation of the vapor passes through the lung and into the bloodstream, it is likely that the flavor chemical profile is somewhat changed through biotransformation before it reaches the kidney. Therefore, the full concentration of the flavoring may not be reaching the kidney and instead its metabolites may be more abundant. CIN and some of its metabolites (methyl cinnamate and cinnamyl alcohol) have been found to have a half-life of 7 h in plasma after oral and intravenous exposure in rats [25], indicating possible metabolites to assess. One other major issues when classifying effects of flavorings in e-liquids is that they are rapidly produced among many manufacturers with different formulations. A previously cited study has found that CIN concentrations in just twenty e-liquid products ranged from 1.7×10⁻⁵ to 1.1 M [11]. Given that e-liquid products must only meet safety standards for ingestion as opposed to inhalation, it is difficult to assess which flavorings are most harmful after inhalation.

In conclusion, the present study characterized the effects of acute CIN exposure on a variety of cellular stress responses in kidney HK-2 cells. Most notably, effects on energy processes and energy output after exposure were affected significantly by CIN. Although this study demonstrates only short-term effects, it is necessary to consider the possible long-term health effects on people use regularly use flavored e-liquids. However, given the difficult nature of classifying and quantifying such effects of flavored e-liquid products, more research is necessary to determine their long-term impact on kidney health.

Data availability

Data will be made available on request.