Acute and Persistent Cardiovascular Effects of Menthol E‐Cigarettes in Mice

Abstract

Background

Although e‐cigarettes provide an alternative to conventional smoking, the cardiovascular impacts of e‐cigarette use are unresolved. The popularity of menthol e‐cigarettes has surged recently and may escalate further with bans on combustible menthol cigarettes and e‐cigarette flavors other than menthol and tobacco. Despite recent evidence in mice that menthol e‐cigarettes acutely induce cardiac arrhythmias, the impacts of repeated menthol e‐cigarette use on cardiovascular function and the cardiac proteome remain unclear. We therefore investigated the acute and persistent cardiovascular effects of menthol e‐cigarettes in a mouse model.

Methods and Results

Adult C57BL/6J mice with ECG and blood pressure radiotransmitters were exposed to e‐cigarette aerosols (180–270 puffs/day; n=4–8/group). One‐day exposures to nicotine‐containing e‐cigarette aerosols depressed heart rate variability regardless of flavor, but menthol e‐cigarette aerosols uniquely increased heart rate and urine epinephrine and elicited spontaneous ventricular premature beats. Menthol e‐cigarette aerosols consistently increased blood pressure acutely, and this effect recurred throughout the 20‐day regimen. Pretreatment with atenolol abolished e‐cigarette–induced arrhythmias, suggesting the involvement of β1‐adrenoceptors. After 4 weeks of exposure to JUUL Menthol aerosol, mice had basal sinus bradycardia that persisted up to 3 weeks after exposure cessation. After cessation, e‐cigarette–exposed mice also exhibited an altered chronotropic response to restraint stress and prolonged ventricular repolarization (corrected QT interval). Integrated proteomic and phosphoproteomic analysis of cardiac tissue harvested from mice exposed to menthol e‐cigarette aerosols for 5 and 20 days revealed molecular signatures of dilated and arrhythmogenic cardiomyopathy.

Conclusions

Exposure to menthol e‐cigarette aerosols induces persistent cardiovascular autonomic imbalance in vivo. These findings raise the possibility of similar effects in humans using mentholated e‐cigarettes.

Nonstandard Abbreviations and Acronyms

HR

heart rate

HRV

heart rate variability

PG

propylene glycol

TNE

total nicotine equivalent

VG

vegetable glycerin

VPB

ventricular premature beat

Research Perspective.

What Is New?

• This preclinical study shows that menthol e‐cigarettes elicit cardiac autonomic imbalance that persists after cessation of e‐cigarette exposure in rodents.

What Question Should be Addressed Next?

• Proof‐of‐concept and epidemiological studies are needed to evaluate the long‐term toxicity of menthol e‐cigarettes in humans.

E‐cigarettes remain popular among adolescents and young adults. Although youth vaping rates have declined recently, the 2024 National Youth Tobacco Survey found that >1.6 million high school and middle school students in the United States still reported current e‐cigarette use. ¹ Nearly 88% of current youth users use flavored e‐cigarettes despite regulatory efforts to stifle marketing and decrease accessibility of flavored e‐cigarettes among adolescents. ¹ , ² Among youth users of flavored e‐cigarettes, 31% use menthol‐flavored products. ² Efforts by the US Food and Drug Administration to ban e‐cigarettes with flavors other than tobacco or menthol could drastically heighten the popularity of menthol e‐cigarettes. Thus, clear understanding of the potential harms of menthol e‐cigarettes is urgently needed to guide policies and inform users and health professionals.

Although e‐cigarettes are perceived as safer alternatives to conventional cigarettes, several studies have now shown that e‐cigarettes induce cardiopulmonary injury and dysfunction. ³ Observational short‐term studies in human users revealed that nicotine‐containing e‐cigarettes acutely increase heart rate (HR), blood pressure, and sympathetic tone. ⁴ , ⁵ In addition, following acute use, e‐cigarettes increased augmentation index and pulse‐wave velocity, consistent with impaired endothelium‐dependent vasodilation. ⁶ , ⁷ Animal studies have shown a causal relationship between e‐cigarette exposure and acute changes in HR, HR variability (HRV), blood pressure, impaired vasodilation, and cardiovascular oxidative stress, depending upon the presence and the concentration of nicotine and nonnicotine constituents. ⁸ , ⁹ , ¹⁰ , ¹¹ , ¹² , ¹³ Several studies also reported that habitual e‐cigarette use may lead to cardiac autonomic remodeling and increased susceptibility to cardiac arrhythmias. Of interest, exposure to vanilla custard–flavored e‐cigarette aerosol for 10 weeks led to sympathovagal imbalance and heightened ventricular tachycardia inducibility in mice. ¹⁴ Similarly, in rats, 8‐week exposure to tobacco‐flavored e‐cigarette (JUUL) aerosol increased vulnerability to atrial and ventricular tachyarrhythmias in concert with cardiac neural remodeling indicative of sympathetic hyperinnervation. ¹⁵ However, it is unclear whether these observations are unique to specific constituents, flavors or a general response to all nicotine‐containing e‐cigarettes. We have recently reported that menthol (but not tobacco‐flavored) e‐cigarettes uniquely evoke spontaneous ventricular arrhythmias and alter early ventricular repolarization in conscious mice. ⁸ Also, we have found that in mice, inhalation of e‐cigarette aerosols containing nicotine salt, which is common in newer flavored “pod” devices, promotes sympathetic dominance, relative tachycardia, and increased ventricular arrhythmias, in contrast to free‐base nicotine, which is present predominantly in older generation e‐cigarettes. ¹⁶ Despite these observations, the long‐term cardiovascular effects of menthol e‐cigarettes remain unclear.

In this study, we sought to discern both the acute and the persistent cardiovascular effects of menthol e‐cigarette exposures in vivo. Using ECG and blood pressure radiotelemetry in conscious mice, we demonstrate that menthol e‐cigarette aerosols transiently evoke arrhythmia via sympathetic dominance and persistently induce cardio‐autonomic imbalance up to 3 weeks after cessation of a 20‐day exposure. Our results reveal a critical need for evaluating the long‐term cardiovascular toxicity of menthol e‐cigarettes in humans.

Methods

This study adheres to the Animal Research: Reporting of In‐Vivo Experiments guidelines. The data that support the findings of this study are available from the corresponding author upon reasonable request.

Animals

C57BL6/J mice (male and female, 8‐ to 12‐week‐old) were obtained from Jackson Laboratories (Bar Harbor, ME) and housed under pathogen‐free conditions (temperature: 22–24 °C, humidity: 40%–60%, 12‐hour dark/light cycle) in the University of Louisville Comparative Medicine Research Units facilities with ad libitum access to standard rodent chow (Rodent Diet 5010, 4.5% fat by weight; LabDiet, St. Louis, MO) and water. All procedures involving animals were approved by the University of Louisville Institutional Animal Care and Use Committee (Protocol No. 20764) and conform to the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

E‐Cigarette Aerosol Generation and Exposure

E‐cigarette aerosol was generated using a JUUL lithium‐ion polymer battery–powered device (≈8 W) attached to an inExpose system with the flexiWare 6 software–controlled electronic nicotine delivery system extension (SCIREQ, Montreal, Canada). For preliminary studies on acute cardiac effects of e‐cigarettes, e‐cigarette aerosols were generated from JUUL pods (3% nicotine) that were emptied, cleaned with ethanol (70% in deionized water), and refilled with propylene glycol (PG):vegetable glycerin (VG) (30/70)±nicotine benzoate (59 mg/mL) as nicotine‐free and nicotine‐containing controls. All other experiments were performed using commercially purchased JUUL pods (JUUL Menthol or JUUL Virginia Tobacco, 5% nicotine). Mice were exposed to 180 to 270 puffs/daily over the course of 3 hours (4‐second puffs, 91 mL puff, 2 puffs/min, 18 puffs/session, 9‐minute washout between sessions, 10 sessions total). Aerosols were delivered to the mice, which were housed in 5‐L chambers during exposures, at 3 L/min bias flow. ⁸ Air controls received filtered room air for the entire duration of the experiment.

Telemetry

Radiotelemetry devices (ETA‐F10 for ECG only and HD‐X11 for blood pressure and ECG; Data Sciences International, St. Paul, MN) were surgically implanted in mice under isoflurane (2% inhaled) anesthesia. ⁸ , ¹⁷ All mice received ketoprofen (5 mg/kg SC) for analgesia up to 3 days after surgery. All animals were allowed to recover for 10 days before starting exposure.

ECG Analysis

ECG signals were continuously sampled at 1 kHz for analysis of HR, HRV, ECG morphology, and arrhythmia incidence using ecgAUTO 3.5 (emka Technologies, Paris, France). ⁸ , ¹⁶ We used automated detection of QR complexes to analyze RR intervals for time‐domain HRV (root mean square of successive differences, and SD of normal interbeat intervals configured for mice). ECG morphology was assessed by automatically analyzing waveforms with a library of ≥50 manually marked representative beats, deriving intervals (PR, JT, QT, and a custom corrected QT), P duration, and amplitudes (S, J, and mean from 15 to 0 ms before Q‐begin serving as the isoelectric line). We excluded motion artifacts, and 1‐minute parameter means coinciding with ≥1 second of tachycardia, that is, >750 bpm from HRV and ECG morphology analyses. Waveforms with mean baseline QT <30 ms or >50 ms were inspected and excluded from QT and JT estimates if T was mismarked. Mice with extremely poor ECG signals (unrecognizable QRS complexes or abnormally low R amplitude) and a high ventricular premature beat (VPB) count before e‐cigarette aerosol exposure (≥24 events/24 hours) were deemed outliers. A VPB is an ectopic QRS complex with at least 3 of the following 4 features: (1) a lengthened QRS duration; (2) a premature occurrence and a subsequent compensatory pause, with the R‐VPB‐R interval greater than or equal to the sum of the prior 2 normal RR intervals; (3) no visible P or an overtly shortened PR; and (4) abnormal R, S, or J wave morphology (amplitudes and areas). ¹⁶ Any VPB occurring within an episode of nonsustained ventricular tachycardia was quantified as an individual arrhythmia. ¹⁶

Blood Pressure Analysis

Standard approaches for arterial blood pressure were used on all pressure waveforms with typical shape and discernable systolic peaks and diastolic nadirs using ecgAUTO. ¹⁷ Beat‐to‐beat series of systolic blood pressure (SBP) and RR interval were examined for spontaneous sequences of increases (up sequences) or decreases (down sequences) in SBP associated with parallel changes in RR interval as previously described. ¹⁷ The length of the sequences included in the analysis was 3 consecutive beats. Sections of ECG traces exhibiting arrhythmias or artifacts were excluded from analysis.

Urinalysis

Mice were housed in urine collection chambers immediately after e‐cigarette exposure with ad libitum access to 0.3% glucose and 0.125% saccharin in drinking water. Urine was collected for 3 hours and was stored at −80 °C until analysis. Levels of catecholamines, nicotine, cotinine, and 3‐hydroxycotinine as well as volatile organic compound metabolites were quantified using ultra‐performance liquid chromatography–tandem mass spectrometry. ¹⁸ , ¹⁹ , ²⁰

Proteomic and Phosphoproteomic Analysis

Mice were euthanized at 5‐ or 20‐day end points with sodium pentobarbital (50 mg/kg IP) and exsanguination. Hearts were perfused with phosphate‐buffered saline containing protease and phosphatase inhibitor cocktail. Apical sections were cut and homogenized in radioimmunoprecipitation assay lysis buffer containing protease and phosphatase inhibitor cocktail. Total proteome and phosphoproteomic analyses of tissue lysates were conducted as recently described. ²¹ RAW data files were searched in PeaksXpro (Bioinformatics Solutions Inc., Waterloo, ON, Canada) using the Denovo, PeaksDB, and PeaksPTM algorithms against the UniprotKB Mus musculus reviewed, canonical protein sequences (proteome ID UP000000589) downloaded on August 9, 2022. The Reporter Ion Quantification algorithm was used with the PeaksPTM results using a mass error tolerance of 15 ppm for the quantification algorithm. ²¹

Statistical Analysis

All data are presented as mean±SEM. Apart from bioinformatic analyses, all other data were analyzed using Prism 10 (GraphPad Software, La Jolla, CA). Unpaired Student's t tests or mixed‐effect analyses with repeated measures were used to analyze differences between groups unless otherwise indicated. ⁸ The Tukey (equal variances assumed) or Sidak post hoc test was used for multiple comparisons. Statistical analyses of the cardiac proteome and phosphoproteome were performed using logarithmized intensities for values that were quantifiable in any experimental condition in Perseus statistical platform. ²² Total proteome and phosphoproteome intensities were normalized by subtracting the median intensity of each sample. ²² To identify significantly modulated phosphopeptides across 2 different conditions, we performed a Student's t test with a P value cutoff of 0.05. ²² Functional enrichment analysis for Kyoto Encyclopedia of Gene and Genomes was performed using STRING version 12.0. ²³ Ingenuity Pathway Analysis (Qiagen, Venlo, the Netherlands) was used to determine overrepresented canonical pathways and toxicology pathways. ²⁴ For all instances, statistical significance was considered at P<0.05.

Results

E‐Cigarette Constituents Influence Acute Cardiac Effects of E‐Cigarettes

To determine the acute cardiac effects of e‐cigarettes, we first conducted a preliminary study in mice instrumented with ECG transmitters (n=4). Given our previous observations that second‐generation menthol e‐cigarettes disproportionately elicit spontaneous arrhythmias and cardiac conduction defects in mice, ⁸ we sought to resolve whether menthol e‐cigarette aerosols affect cardiac electrophysiology when generated from fourth‐generation pod devices containing nicotine salt formulations that augment nicotine intake. ²⁵ For this, mice were acutely exposed to various e‐cigarette aerosols generated from JUUL pods containing PG:VG (30/70)±nicotine benzoate (59 mg/mL), or JUUL formulations of VT (Virginia tobacco) or Menthol e‐liquids with 59 mg/mL nicotine benzoate. Analysis of real‐time changes in HR over 15 e‐cigarette puff cycles (each cycle consisted of a 9‐minute puff session followed by 9 minutes of ambient air for washout) revealed that only menthol e‐cigarettes robustly increased HR, especially during washout, relative to both air and PG:VG exposures (Figure 1A). Interestingly, tobacco‐flavored e‐cigarette (JUUL VT) failed to induce a significant HR elevation relative to air controls despite having the same nicotine content as JUUL Menthol (Figure 1A, Table S1). HRV analysis revealed that the menthol e‐cigarette alone markedly reduced SD of normal interbeat intervals during the washout phases (P<0.05 versus midexposure; Table S1). However, root mean square of successive differences, a short‐term HRV metric, was significantly reduced during washout relative to the midexposure phase in all nicotine‐containing e‐cigarette exposures (Table S1), indicative of nicotine‐mediated sympathetic excitation.

Figure 1. Acute cardiac effects of flavored JUUL e‐cigarettes.

A, Changes in HR across 15 cycles of puff sessions with various e‐cigarette aerosols, relative to baseline HR of the respective exposure. P<0.05 vs air^a, PG:VG^b, PG:VG+N^c, JUUL VT^d, or P<0.10 vs air^e using mixed‐effects analysis for n=4/exposure. B, Incidence rate of VPBs during various e‐cigarette exposures for n=4/exposure. C, TNE measured in urine after various e‐cigarette exposures. P values were determined using mixed‐effects analysis for n=4/exposure. D, Regression analysis between urine TNE and VPBs in all exposures. HR indicates heart rate; N, nicotine; PG, propylene glycol; TNE, total nicotine equivalent; VG, vegetable glycerin; VPB, ventricular premature beats; and VT, Virginia Tobacco.

In addition to the significant changes in HR and HRV, menthol e‐cigarettes uniquely elicited a greater number of spontaneous VPBs than the air and PG:VG exposures (Figure 1B). This was accompanied by a shortening of the PR segment and PR interval, indicative of accelerated atrioventricular conduction (both P<0.05 versus PG:VG; Table S2). The PR interval was also shortened in mice exposed to PG:VG+nicotine (P<0.05 versus PG:VG). These effects parallel our observations that cotinine correlates inversely with PR segment in smokers and that smoking corresponds with an increased risk of abnormally short PR. ²⁶ We also observed a tendency for an increase in J amplitude in all nicotine‐containing exposures (Table S2) consistent with our previous work. ⁸ Because the J wave results from early repolarization heterogeneity and is associated with a high HR in mice, ²⁷ the increase in J amplitude is likely due to nicotine‐induced sympathoexcitation. To determine whether nicotine intake varied across these exposures, we performed analyses of nicotine and volatile organic compound metabolites in urine collected from mice for 3 hours immediately after each exposure. Compared with air and PG:VG, all nicotine‐containing e‐cigarette exposures increased urine levels of nicotine, cotinine, 3‐hydroxycotinine and total nicotine equivalents (TNEs) (Figure 1C, Table S3).

Although JUUL VT exposure resulted in ≈40% lower TNEs than JUUL Menthol, this was not statistically significant. We also noted that TNE for the nicotine‐free PG:VG exposure was one tenth that for the nicotine‐containing exposures; these trace levels are likely due to incomplete removal of nicotine from the JUUL pods that were cleaned and used for the exposures. We noted that urine TNE positively correlated with VPBs (Figure 1D). Apart from nicotine and its metabolites, all the other urinary volatile organic compound metabolites measured in this study were not robustly affected relative to the air exposure (Table S3). Parallel urinary catecholamine analysis using these samples revealed that all nicotine‐containing e‐cigarettes tended to reduce excretion of dopamine and its metabolite, 3‐methoxytyramine (Table S4). JUUL Menthol also tended to increase urinary epinephrine in all relative to the air control and PG:VG alone (Table S4).

Menthol E‐Cigarette Acutely Elevates Heart Rate and Blood Pressure and Evokes Arrhythmias

We next used naïve mice (n=8) instrumented with transmitters capable of measuring both ECG and blood pressure to better understand the acute cardiovascular effects of menthol e‐cigarettes. We subjected these mice to 10 e‐cigarette puff cycles generated from JUUL Menthol over the course of 3 hours, while a separate control group of mice (n=7), which exhibited no baseline differences in any of the cardiovascular parameters that were studied, was concurrently exposed to filtered ambient air in a separate 5‐L chamber (Table S5). The menthol e‐cigarette exposure induced a robust increase in HR immediately at the onset of the first puff session, which was sustained for 18 minutes until the end of the following washout session (Figure 2A). After this first cycle, the menthol e‐cigarette aerosol consistently induced bradycardia during puffs and relative tachycardia during washout, resembling the biphasic response frequently attributed to respiratory irritant‐induced cardiopulmonary afferent reflexes combined with the sympathomimetic effects of nicotine, as seen in our preliminary studies above. Menthol e‐cigarette aerosol also induced a sustained elevation in SBP relative to the air control during the puff sessions, which was slightly attenuated during the intervening washouts (Figure 2B). To account for the phasic responses and dynamic changes in HR and SBP, we analyzed the mean differences in these parameters relative to the air control by phase (ie, midexposure and washout) across the 10 puff sessions (Figure 2C). Compared with the air control, menthol e‐cigarettes significantly elevated HR during washout but not during puff sessions, although there was a dramatic increase in HR during the first puff session. Conversely, SBP was significantly increased during midexposure but showed only a tendency to increase during washout that was slightly blunted compared with midexposure (Figure 2C). Menthol e‐cigarette aerosol also led to a significantly higher VPB incidence rate compared with the air control (Figure 3A), and half of the e‐cigarette–exposed animals (4/8), but zero mice in the air control, had a VPB frequency >1. This suggests that prior exposure to nicotine‐free and nicotine‐containing e‐cigarette aerosols was not necessary to sensitize these mice to menthol e‐cigarette–induced arrhythmias. We have previously reported that biological sex may influence the arrhythmogenicity of e‐cigarettes. ⁸ Thus, we exposed a separate group of age‐matched male and female mice (n=4/sex) to JUUL Menthol. Interestingly, we observed a heterogeneous distribution of VPBs in the females, although there was a similar trend of increased VPBs and incidence (50%) in the e‐cigarette–exposed mice (Figure 3B). Notably, 1 female mouse presented with 3 bouts of nonsustained ventricular tachycardia (<30 seconds), while the others had VPB rates comparable with those in males. Representative VPBs and nonsustained ventricular tachycardia in mice exposed to menthol e‐cigarettes are shown in Figure 3C.

Figure 2. Cardiovascular effects of a 10‐cycle exposure to JUUL Menthol aerosol.

Changes in (A) HR and (B) SBP across 10 cycles of exposure to JUUL Menthol e‐cigarette aerosols, relative to baseline values. P<0.05 vs air^a using mixed‐effect analysis, n=7–8/group for HR and n=6/group for SBP, respectively. C, Phase‐specific changes in HR and SBP relative to the air controls. P values were determined using Student's t test. Gray vertical rectangles indicate 9‐minute puff phases. HR indicates heart rate; and SBP, systolic blood pressure.

Figure 3. Arrhythmogenic effects of JUUL Menthol.

A, Incidence rate of ventricular premature beats during a single 10‐cycle exposure to JUUL Menthol in ECG+BP telemetered male mice. P‐value determined via independent t test for n=7–8/group. B, Incidence rate of ventricular premature beats during a 10‐cycle exposure to JUUL Menthol aerosol in ECG‐only telemetered male and female mice (n=4/group). C, Representative ECG traces depicting 2 spontaneous ventricular premature beats (upper) and nonsustained ventricular tachycardia in mice exposed to JUUL Menthol e‐cigarette aerosol. BP indicates blood pressure.

Autonomic Imbalance Underlies Acute Cardiac Effects of Menthol E‐Cigarettes

Autonomic imbalance has been implicated in the cardiovascular effects of e‐cigarettes and may underlie e‐cigarette–evoked arrhythmias. ⁵ Consistent with our observation in mice exposed serially to different e‐cigarette aerosols (Tables S1 and S2), there was a marked shift in RR distribution toward shorter RR intervals during acute exposure to menthol e‐cigarette aerosol relative to the air group (Figure 4A). Compared with air controls, there was also a trend for increased HRV midexposure as indicated by both SD of normal interbeat intervals and root mean square of successive differences, but this was not statistically significant (Table S5). In contrast, both SD of normal interbeat intervals and root mean square of successive differences were significantly reduced during the washout session in mice exposed to e‐cigarettes relative to air control (Figure 4B). Separate subsets of mice (n=4/group) were used for urine analysis immediately after exposure and revealed that menthol e‐cigarette significantly reduced urinary dopamine and robustly increased urinary epinephrine relative to air controls (Figure 4C). Overall, the shift in RR distribution, in concert with reduced HRV and increased urinary epinephrine indicates sympathetic hyperactivity in mice exposed to menthol e‐cigarettes. To determine whether sympathetic hyperactivity contributes to menthol e‐cigarette–evoked arrhythmias, we then pretreated naïve ECG‐telemetered mice (n=4) with a β₁‐adrenergic blocker (atenolol) in drinking water (0.1 g/L) for 3 days and exposed them acutely to menthol e‐cigarette aerosols (Figure 4D). Separate age‐matched control animals (n=4) received normal drinking water and served as vehicle controls. Based on the daily water consumption (Figure S1), we estimated the mice in the treatment group received 20.9±1.5 mg/kg atenolol daily. Atenolol treatment for 3 days led to a significant reduction in resting heart rate (−49.6±10.0 bpm) relative to baseline (Figure S1). As observed in previous experiments, mice exposed to menthol e‐cigarette aerosol had higher VPB incidence rates than those in the air control (Figure 4D). However, atenolol significantly prevented e‐cigarette–induced VPBs, suggesting that menthol e‐cigarettes evoke arrhythmias via sympathetic stimulation of β₁‐adrenergic receptors.

Figure 4. Role of autonomic imbalance in menthol e‐cigarette–evoked spontaneous arrhythmias.

A, Distribution of RR in mice exposed to JUUL Menthol. B, Change in SDNN and RMSSD during the washout phase in mice exposed to JUUL Menthol. P values were determined using an independent t test (n=7–8/group). C, Heat map (left) and bar graphs (right) representing changes in urinary catecholamine levels after a 10‐cycle exposure to JUUL Menthol. P values were determined using a paired t test (n=4/group). D, Incidence rate of ventricular premature beats during a single JUUL Menthol exposure in ECG‐telemetered male mice pretreated with atenolol for 3 d compared with mice pretreated with normal drinking water. P values were determined using mixed‐effects analysis (n=4). RMSSD indicates root mean square of successive differences; and SDNN, SD of normal interbeat intervals.

Persistent Cardiovascular Effects of Menthol E‐Cigarettes in Mice

To determine whether menthol e‐cigarette aerosols induce persistent cardiovascular effects, we studied ECG and blood pressure changes in mice instrumented with HD‐X11 implants across 4 weeks of exposure (5 days/week, 20 exposure days). Figure 5 shows acute changes in cardiovascular parameters relative to baseline on respective days in response to air or JUUL Menthol exposures. Although menthol e‐cigarettes significantly elevated HR on the first day, this elevation dampened over time, and no significant difference was observed between air and e‐cigarette–exposed mice toward end of exposure (Figure 5A). Conversely, menthol e‐cigarette–induced elevations in blood pressure, especially SBP (and diastolic blood pressure to a lesser extent), persisted across all time points, indicating that the pressor effect of nicotine is intact even after 4 weeks of exposure (Figure 5B and 5C). There were no significant differences in HR and diastolic blood pressure at all time points compared with the initial control exposure (baseline); however, SBP in e‐cigarette–exposed mice was significantly lower than baseline on day 15 and were slightly lower than the air control toward the end of exposure (Figure S2). The average maximum upslope of aortic pressure (dP/dt+) can be used as a crude marker of contractility. Although menthol e‐cigarette aerosols increased arterial dP/dt+ on day 1 during washout (Table S5), we observed a trend for dP/dt+ reduction in e‐cigarette–exposed mice relative to air controls starting at day 10, during preexposure, puff sessions, and washouts, suggesting that menthol e‐cigarettes modestly impaired contractility (Figure 5D). Nonetheless, e‐cigarette exposure has little to modest impact on the pulse pressure (Figure 5E), although there was a tendency for reduction in week 4.

Figure 5. Persistent cardiovascular effects of JUUL Menthol.

Changes in (A) HR, (B) SBP, (C) DBP, (D) dP/dt+, and (E) PP from baseline air exposure and (F) absolute ventricular premature beat incidence rate across 20 days of exposure to JUUL Menthol e‐cigarette aerosols. P<0.05 vs air^a or P<0.10 vs air^b using mixed‐effects analysis, n=7–8/group for HR and n=6/group for SBP, respectively. DBP indicates diastolic blood pressure; HR, heart rate; PP, pulse pressure; and SBP, systolic blood pressure.

Given the multiple variables that influence this parameter, we can only speculate that these observations might derive from negative inotropes upregulated by nicotine, such as adenosine ²⁸ or angiotensin II. ²⁹ Daily analysis of spontaneous VPBs in mice exposed to menthol e‐cigarettes revealed that e‐cigarette–evoked arrhythmias were completely blunted toward the end of the 20‐day exposure, although other acute sympathomimetic effects of nicotine such as increased SBP were preserved (Figure 5F, Figure S3). This is not surprising given that repeated exposure to menthol may abolish inhaled irritant‐evoked reflex tachyarrhythmias in rodents. ³⁰ Echocardiographic analysis was performed in a subset of animals (n=6/group) during week 4 of exposure (≈1 hour after exposure, under isoflurane anesthesia) to determine whether menthol e‐cigarettes induced any overt structural or functional changes in the heart (TableTable). We found that menthol e‐cigarettes only significantly decreased end‐systolic volume by 46% and increased midsystole posterior wall thickness by 29% (TableTable). There was also a trend for a slight increase in cardiac output that was still within the range of normally reported cardiac output in mice. The contrasting inotropic implications of aortic dP/dt+ during the latter e‐cigarette exposures and echocardiography shortly thereafter might be due to transient, phasic effects of e‐cigarette aerosol on contractility (like other parameters), or to isoflurane suppressing an exposure‐induced negative inotrope (ie, adenosine) ³¹ and thus freeing nicotine to enhance contraction.

Table 1.

Echocardiographic Data in C57BL/6J Mice After 4‐Week Exposure to Menthol E‐Cigarette Aerosol

Parameter	Air	JUUL Menthol	P value
Body weight, g	28.2±0.7	26.7±0.6	0.120
Heart rate, bpm	575±14	566±22	0.716
LVIDd, mm	3.5±0.1	3.2±0.1	0.102
LVIDs, mm	1.9±0.2	1.6±0.1	0.162
Fractional shortening, %	44.5±5.5	50.6±1.6	0.310
LVPWd, mm	0.9±0.1	1.1±0.1	0.148
LVPWs, mm	1.4±0.1	1.8±0.1	0.021
LVAWd, mm	1.1±0.2	1.0±0.1	0.505
LVAWs, mm	1.6±0.1	1.7±0.1	0.905
Ejection fraction, %	71.5±5.8	82.1±1.3	0.148
End‐diastolic volume, μL	37.6±4.5	33.5±0.8	0.400
End‐systolic volume, μL	13.1±0.6	7.1±1.7	0.001
Cardiac output, mL/min	14.8±0.8	17.2±0.7	0.057
IVRT, ms	15.9±0.7	14.5±0.6	0.170

Values are given as mean±SE for n=6/group. An unpaired Student's t test was used to compare differences between groups. IVRT indicates isovolumic relaxation time; LVAWd, left ventricular anterior wall thickness in diastole; LVAWs, left ventricular anterior wall thickness in systole; LVIDd, left ventricular internal diameter in diastole; LVIDs, left ventricular internal diameter in systole; LVPWd, left ventricular posterior wall thickness in diastole; and LVPWs, left ventricular posterior wall thickness in systole.

We performed a longitudinal analysis of HR, HRV, body temperature, and activity over 24 hours immediately after the 20‐day menthol e‐cigarette exposure (n=3 for air, n=4 for e‐cigarette) to determine whether persistent e‐cigarette exposure altered basal physiology in mice (Figure S4). Blood pressure data are not included because of poor signal quality. We did not observe significant differences in any of the parameters that were studied, although it is noteworthy that HR in e‐cigarette–exposed mice tended to be lower than that in the air control during the light phase (6:00 am to 6:00 pm), consistent with the preexposure HR during week 4 of exposure (Figure S2A). At this same time point, that is, the 24‐hour period immediately after the final exposure, we subjected the remaining animals (n=4 for air, n=4 for e‐cigarette) to an acute restraint stress test to unmask persistent cardiac autonomic imbalance and to test differences in stress‐inducible arrhythmias. Interestingly, we observed that mice exposed to menthol e‐cigarettes exhibited bradycardia at baseline during the light phase (1:00 pm), similar to our observations with the 24‐hour HR analysis (Figure S5A), which is indicative of underlying heart rhythm effects that only manifest during the low‐activity period (Figure S4E). In addition, the maximum HR attained during the restraint stress test was significantly lower in mice exposed to e‐cigarettes compared with that of the air control (Figure S5B). Yet, e‐cigarette–exposed mice achieved a greater change in HR during this challenge (Figure S5B). Acute restraint stress evoked VPBs in both groups (compared with 0 VPBs during the prestress period), but there was no statistically significant difference between these groups (Figure S5C).

E‐Cigarette–Induced Disturbance in Heart Rhythm Persists After Cessation

Because previous studies have shown that tobacco‐induced perturbations in heart rhythm persist after cessation, ³² we followed up with a second restraint stress challenge 3 weeks after the conclusion of inhalation exposures in these mice (n=7 for air, n=8 for e‐cigarette). At prestress baseline, e‐cigarette–exposed mice continued to exhibit sinus bradycardia and increased HRV without any changes in SBP relative to the air control (Figure 6A). These measurements were taken at the same time point as our first stress challenge. E‐cigarette–exposed mice again displayed impaired chronotropic responses to stress, indicating the persistence of cardiac autonomic imbalance (Figure 6B). Although this stress challenge induced a significant increase in SBP in both groups relative to their baselines (+22±6 mm Hg versus baseline in air and +33±15 mm Hg versus baseline in e‐cigarette), no significant difference was noted between these groups (Figure 6C). VPBs evoked by this stress challenge were not significantly different between groups (1.3±0.5 events/hour in e‐cigarette versus 1.6±0.4 events/hour in air; P=0.551). However, ECG morphology analysis revealed that e‐cigarette–exposed mice had a significant prolongation in rate‐corrected QT interval (Figure 6D and 6E), consistent with our recent epidemiological observation in smokers. ²⁶

Figure 6. Impact of 3‐week cessation on e‐cigarette–induced cardiac autonomic imbalance in mice.

(A) Baseline HR, SBP, SDNN, and RMSSD measured during steady state in mice 3 weeks after the end of the 20‐d JUUL Menthol exposure. P values were determined using the Student's t‐test for n=4–8/group. Changes in (B) HR and (C) SBP during an acute restraint stress challenge 3 weeks after the end of the 20‐d JUUL Menthol exposure. P<0.05 vs air^a or P<0.10 vs air^b using mixed‐effects analysis for n=4–8/group. D, Uncorrected QT and (E) rate‐corrected QT determined during baseline in mice 3 wks after the end of the 20‐d JUUL Menthol exposure. P values were determined using an independent t test for n=7 in air and n=8 in JUUL Menthol. HR indicates heart rate; PP, pulse pressure; R, duration of restraint stress; RMSSD, root mean square of successive differences; SBP, systolic blood pressure; SDNN, SD of normal interbeat intervals.

E‐Cigarette Induces Molecular Signatures of Dilated and Arrhythmogenic Cardiomyopathy

To identify the molecular signatures of menthol e‐cigarette–induced cardiovascular perturbations, we performed quantitative global proteomic and phosphoproteomic analysis in hearts collected from mice immediately after 5‐day or 20‐day exposures to menthol e‐cigarette aerosol (n=5/group per time point). Across the 2 pairwise comparisons (JUUL Menthol versus air 5‐day; JUUL Menthol versus air 20‐day), we quantified 1294 unique phosphorylation sites (phosphoproteome) and 3167 unique proteins (total proteome). The predominant phosphorylated residue was serine (92.04%), followed by threonine (7.49%) and tyrosine (0.48%). From these data sets, 130 and 209 proteins were differentially regulated (P<0.05) by 5‐day and 20‐day menthol e‐cigarette exposure, respectively (Data S1). By contrast, only 48 and 27 unique phosphorylation sites were differentially regulated (P<0.05) by 5‐day and 20‐day menthol e‐cigarette exposure, respectively (Data S2). The differences in the number of proteins and phosphorylation sites altered within each comparison are shown in Figure 7A. For the purposes of direct comparison between 2 time points, we then integrated data sets containing differentially regulated proteins and phosphosites from both comparisons (JUUL Menthol versus air 5‐day; JUUL Menthol versus air 20‐day) (Data S3). This integrative analysis revealed that 18 unique proteins were differentially regulated by e‐cigarettes at both time points (Table S6). To gain further insight into the molecular pathways altered by menthol e‐cigarettes, merged data sets of differentially regulated proteins and phosphosites for each time point were then independently analyzed for functional enrichment using STRING version 12 and Ingenuity Pathway Analysis. Kyoto Encyclopedia of Gene and Genomes enrichment analysis in STRING revealed that pathways related to dilated cardiomyopathy, and adrenergic signaling in cardiomyocytes were overrepresented in both time points (Figure S6). Canonical pathways identified by Ingenuity Pathway Analysis also indicated an overrepresentation of proteins associated with dilated cardiomyopathy and protein kinase A signaling–associated molecules in both comparisons (Figure S7). Considering the modest alterations in these proteins, small increase in posterior wall thickness from echocardiographic data, and lack of change in other classical markers of hypertrophy such as atrial natriuretic peptide and β myosin heavy chain (Data S1), we posit that these are early signatures of dilated cardiomyopathy. However, more in‐depth molecular analysis is necessary to verify differentially regulated proteins to validate Kyoto Encyclopedia of Gene and Genomes functional analysis. To provide insight into the potential long‐term toxicological impact of e‐cigarettes, we mapped differentially expressed proteins using the Ingenuity Pathway Analysis Tox list and found that e‐cigarette–induced molecular changes were associated with cardiac dilation, cardiac dysfunction, cardiac enlargement, cardiac arrhythmia, and, to a lesser extent, congenital heart anomaly (Figure 7B).

Figure 7. Proteomic and phosphoproteomic signatures of JUUL Menthol–induced persistent cardiac autonomic imbalance.

A, Volcano plots depict differentially regulated proteins and phosphosites following a 5‐day or a 20‐day JUUL Menthol e‐cigarette exposure. P values were determined using multiple t tests comparing log‐transformed raw intensities (n=5/group). B, Tox analysis of differentially regulated proteins at the 20‐day time point using Ingenuity Pathway Analysis.

Discussion

Major findings of this study are that the inhalation of menthol e‐cigarette aerosols acutely increases heart rate, systolic arterial pressure, and spontaneous ventricular arrhythmias, and the arrhythmogenesis occurs through β₁‐adrenoceptor stimulation. We found that e‐cigarette–evoked increases in heart rate and spontaneous arrhythmias dampen after the initial exposure day, whereas acute increases in systolic pressure remain undiminished over 20 exposure days. On the morning following each menthol e‐cigarette exposure, systolic pressure was unaffected, but in the fourth week of the regimen, basal heart rate was depressed, and this effect persisted for at least 3 weeks after exposure cessation. Mice exposed to e‐cigarette aerosols showed altered chronotropic response to acute stress after 20 exposure days even after 3 weeks of cessation indicative of persistent cardiovascular autonomic imbalance. Prolonged ventricular repolarization immediately before stress, concurrent with altered chronotropic responses to stress, further indicated underlying cardiac conduction defects.

Our findings parallel human studies demonstrating that nicotine‐containing e‐cigarettes increase HR and SBP acutely ⁴ , ⁵ and recapitulate our previous observation that menthol e‐cigarettes acutely evoke spontaneous arrhythmias in mice. ⁸ Some researchers have posited that nicotine, but not other constituents, mediates the acute cardiovascular effects of e‐cigarettes through sympathetic response. ⁴ , ³³ Indeed, we recently reported that changes in nicotine formulation influence e‐cigarette–induced acute sympathoexcitation in mice. ¹⁶ Here, we observed that menthol e‐cigarettes increased both TNE and epinephrine and reduced time‐domain HRV indices, reinforcing the notion that nicotine mediates e‐cigarette–induced sympathetic dominance. We also noted that β₁‐selective adrenergic blockade using atenolol abolishes e‐cigarette–evoked spontaneous arrhythmias, suggesting that nicotine‐mediated sympathetic dominance also underlies the arrhythmogenicity of menthol e‐cigarettes.

The acute arrhythmogenic and positive chronotropic impacts of menthol e‐cigarette aerosols exceeded those of tobacco‐flavored e‐cigarette aerosols but were dampened over the 20‐day exposure period despite consistent urinary TNE from week 1 to week 4 (Figure S8). Conversely, acute increases in SBP recurred until the exposure regimen's end, suggesting that the sympathomimetic effects of nicotine on blood pressure persist. The divergent timing of the chronotropic and pressor effects of e‐cigarette exposures may stem from several factors. For one, the aldehydes found within e‐cigarette aerosols, such as formaldehyde and acetaldehyde, and their metabolites, may interfere with heart rate regulation and acutely promote bradycardia. ¹¹ , ³⁴ , ³⁵ Alternatively, cardiac reflexes to inhaled irritants can diminish with desensitization of irritant receptors following repeated exposures ³⁶ , ³⁷ or menthol‐mediated abolishment of irritant reflexes, ³⁰ whereas e‐cigarettes could increase SBP independent of irritant reflexes via pressor actions of nicotine. ³⁸

Echocardiography at the end of the exposure regimen revealed that the 20‐day exposure to menthol e‐cigarette did not induce any overt structural and functional changes in the heart, except for reduction in end‐systolic volume and increased posterior wall thickness during systole. Because echocardiography was performed 1 hour after the exposure session, we believe these responses reflect the positive inotropic effects of circulating nicotine and catecholamines. The absence of overt structural and functional changes after exposure is not surprising, as several reports indicate e‐cigarette–induced cardiac remodeling and systolic dysfunction are not evident in mice before 6 months of repeated exposure. ⁹ , ³⁹ One recent study also suggested that the age at which e‐cigarette exposure begins determines susceptibility to e‐cigarette–induced reduction in cardiac function in mice. ⁴⁰ It was shown that among male mice, adolescents but not adults displayed reduced fractional shortening and diastolic dysfunction after 3 months of e‐cigarette exposure. ⁴⁰

A notable finding of this study is that e‐cigarette–exposed mice exhibited bradycardia after a 20‐day exposure, and this persisted up to 3 weeks after cessation. These mice also displayed altered chronotropic responses to stress challenge, suggesting a persistent impairment of HR regulation. Although this phenomenon has not been reported in the context of e‐cigarette exposures, one study has shown that female BALB/c mice displayed bradycardia, dynamic fluctuations in blood pressure, and disrupted physiological rhythms after just 1 week of daily cigarette smoke exposure, and these effects persisted across 24 weeks of exposure and 4 weeks following cessation. ³² Similarly, our observations could stem, at least in part, from circadian disruption with 4‐week e‐cigarette exposures due to specific aerosol constituents like nicotine or menthol. Nonetheless, the prolongation of heart rate–corrected QT interval indicates underlying conduction defects that promote arrhythmia. Accordingly, we found in a large cohort study that smoking associates with prolonged ventricular repolarization, and that this phenotype in smokers predicts the long‐term cardiovascular mortality rate. ²⁶ Future investigations into the impacts of e‐cigarette aerosols on circadian regulation of cardiac rhythm and electrophysiology may be informative given that circadian clocks govern cardiac repolarization and susceptibility to sympathetic‐evoked arrhythmias. ⁴¹ , ⁴²

Phosphoproteomic analysis revealed that the 20‐day menthol e‐cigarette exposure induced a molecular signature of dilated and arrhythmogenic cardiomyopathy in mice, and some facets of these signatures can be detected as early as 5 days despite our inability to detect gross cardiac structural changes with echocardiography on day 20. However, more long‐term studies are needed to determine how these processes are temporally regulated by e‐cigarette exposure, particularly considering that these changes are subtle at the time points studied herein and that tobacco‐induced diseases usually follow a long latency period before manifesting as pathophysiology. We also acknowledge that there is a need to validate these proteomic changes using a more rigorous approach, with validated antibodies and use of susceptible models that may further delineate the influence of menthol e‐cigarette aerosols on cardiac structure and function.

In this study, we demonstrated that repeated exposure to menthol e‐cigarette aerosols induces persistent cardio‐autonomic imbalance in mice; however, these findings may not extrapolate to humans due to obvious differences in physiological attributes (eg, HR, electrophysiology, tidal volume, and nicotine metabolism). Further, because we did not directly compare with nonmenthol e‐cigarettes in our repeated exposure study, only our acute exposure findings suggest menthol flavor promotes e‐cigarette–induced arrhythmias. In this simulated vaping experiment, mice were exposed to aerosols diluted in a 5‐L whole‐body chamber instead of nose‐only exposure. Although whole‐body simulated aerosol exposure has been extensively used to study cardiac effects of inhaled aerosols and avoids restraint stress‐induced changes in HR and HRV that may otherwise confound our observations, it can dilute the aerosol inhaled per puff by the rodents. Nonetheless, perhaps owing to the higher respiratory rate and greater lung surface to volume ratio in mice, our whole‐body approach with a comparable device wattage, nicotine salt formulation, and nicotine concentration renders plasma nicotine levels in mice comparable with human e‐cigarette users. ¹⁶ Although the aortic pressure data (dP/dt+) suggest that repeated menthol e‐cigarette exposures might have modestly suppressed left ventricular systolic function in conscious mice, these observations relied on a crude index of contractility and conflicted with echocardiographic measures in anesthetized mice. We also used only a single sex and strain (ie, male C57BL/6J) for the repeated exposures, although both sexes were used for acute studies due to heterogeneous e‐cigarette–induced arrhythmia susceptibility in female mice. Future studies in both sexes may reveal sex‐specific toxicities with greater generalizability and more nuanced public health implications. Additionally, we excluded several animals from blood pressure analysis due to a loss of signal quality over time, which diminished the power of our analysis to detect diurnal changes in blood pressure and measure nondipping blood pressure as a predictor of hypertension.

Conclusions

Taken together, the results of this study show that menthol e‐cigarettes induce persistent cardio‐autonomic imbalance that may promote the development of cardiomyopathy. This proof‐of‐concept study highlights the urgent need for evaluating the long‐term toxicity of flavored e‐cigarettes, especially menthol e‐cigarettes in users.

Sources of Funding

This work was supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health and the Food and Drug Administration Center for Tobacco Products (R01HL147353, R01HL163818, and U54HL120163), the National Institute of General Medical Sciences of the National Institutes of Health (P30GM127607, S10OD025178), and the National Institute of Environmental Health Sciences of the National Institutes of Health (P30ES030283). A.R.R. was supported by a Career Enhancement Core Fellowship from the American Heart Association Tobacco Regulation and Addiction Center 2.0. An ITEMFC Research Voucher from the University of Louisville Center for Integrative Environmental Health Sciences (P30ES030283) was used to support the phosphoproteomic analysis of heart samples. This content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health, Food and Drug Administration, or American Heart Association.

Disclosures

None.

Supporting information

Data S1–S3

Tables S1–S6

Figures S1–S8