E-cigarettes Induce Dysregulation of Autophagy Leading to Endothelial Dysfunction in Pulmonary Arterial Hypertension

Abstract

Given the increasing popularity of electronic cigarettes (e-cigs), it is imperative to evaluate the potential health risks of e-cigs, especially in users with preexisting health concerns such as pulmonary arterial hypertension (PAH). The aim of the present study was to investigate whether differential susceptibility exists between healthy and patients with PAH to e-cig exposure and the molecular mechanisms contributing to it. Patient-specific induced pluripotent stem cell-derived endothelial cells (iPSC-ECs) from healthy individuals and patients with PAH were used to investigate whether e-cig contributes to the pathophysiology of PAH and affects EC homeostasis in PAH. Our results showed that PAH iPSC-ECs showed a greater amount of damage than healthy iPSC-ECs upon e-cig exposure. Transcriptomic analyses revealed that differential expression of Akt3 may be responsible for increased autophagic flux impairment in PAH iPSC-ECs, which underlies increased susceptibility upon e-cig exposure. Moreover, knockdown of Akt3 in healthy iPSC-ECs significantly induced autophagic flux impairment and endothelial dysfunction, which further increased with e-cig treatment, thus mimicking the PAH cell phenotype after e-cig exposure. In addition, functional disruption of mTORC2 by knocking down Rictor in PAH iPSC-ECs caused autophagic flux impairment, which was mediated by downregulation of Akt3. Finally, pharmacological induction of autophagy via direct inhibition of mTORC1 and indirect activation of mTORC2 with rapamycin reverses e-cig-induced decreased Akt3 expression, endothelial dysfunction, autophagic flux impairment, and decreased cell viability, and migration in PAH iPSC-ECs. Taken together, these data suggest a potential link between autophagy and Akt3-mediated increased susceptibility to e-cig in PAH.

Graphical Abstract

Graphical Abstract.

Significance Statement.

PAH iPSC-ECs are more susceptible to e-cigs than healthy iPSC-ECs, supporting the possibility that e-cigs might have a particularly maladaptive effect on PAH iPSC-ECs. E-cig-induced impairment of autophagic flux is more pronounced in PAH iPSC-ECs. Transcriptomic and biochemical analyses indicate that the differential susceptibility of PAH iPSC-ECs to e-cig exposure is mediated by Akt3-mediated impaired autophagy.

Introduction

Electronic cigarettes (e-cigs), also called electronic nicotine delivery systems (ENDS), are battery-operated devices that mimic the look and feel of smoking by vaporizing e-liquids, mainly composed of nicotine, propylene glycol (PG) and/or vegetable glycerin (VG), and flavorings.¹ E-cig aerosol generally contains fewer toxins than combustible cigarettes and as such has been promoted as a safer replacement for cigarette smoking by not only assisting smoking cessation attempts but, perhaps more importantly, supporting long-term abstinence from tobacco cigarette smoking.² However, e-cig aerosols are not totally harmless, as users are potentially exposed to harmful substances known to have adverse health effects, including nicotine, ultrafine particles, heavy metals (eg, nickel, tin, and lead), and volatile organic compounds.³ Several recent studies have also reported that e-cig exposure induces acute vascular and lung inflammation and oxidative stress, affecting the metrics of lung and vascular health.^4,5

Pulmonary arterial hypertension (PAH) is a chronic disorder of the pulmonary vasculature characterized by a progressive increase in pulmonary vascular resistance causing right heart failure, ultimately leading to death or the need for transplantation.⁶ Although the pathogenesis of PAH is still incompletely understood, dysfunction of pulmonary arterial endothelial cells (PAECs) plays an important role in the initiation and progression of PAH.⁷ The initial detrimental stimulus leading to endothelial dysfunction in PAH is still undetermined; however, cigarette smoking is considered a major environmental factor associated with systemic pulmonary endothelial dysfunction in mice,^8,9 which may also apply to humans.^10,11 As expected, chronic inhalation of nicotine resulted in increased systemic systolic and diastolic blood pressure as well as pulmonary hypertension with pulmonary vascular and right ventricular remodeling.¹² Although cigarette smoking is common among patients with PAH, with over one-third of female and two-thirds of male patients with PAH having a previous smoking history,¹³ the impact of e-cig exposure as a risk factor that may aggravate PAH has never been investigated.

Autophagy is a conserved catabolic process that degrades and removes damaged or unneeded organelles by delivering cytoplasmic material to the lysosome.¹⁴ Although autophagy has traditionally been thought of as an adaptive response in maintaining intracellular homeostasis, including in ECs, conflicting roles of autophagy are starting to emerge in various diseases, such as PAH.¹⁵ While the dichotomous nature of autophagy in the development of PAH remains an open question, it appears to play an important role in vascular remodeling. For example, hypoxia-induced increased autophagy in PAECs was thought to be a protective mechanism, as LC3B-II knockout mice have exaggerated PAH in response to chronic hypoxia.¹⁶ A recent study implicated the detrimental role of increased autophagy in fetal angiogenesis in an experimental model of persistent PAH of fetal lambs.¹⁷ However, the association between e-cig and endothelial autophagy remains to be fully elucidated.

Recent studies have demonstrated that PAH iPSC-ECs are useful surrogates for native PAECs in PAH modeling.¹⁸ Here, we used patient-specific iPSC-ECs from patients with PAH to assess whether e-cig predisposes PAH iPSC-ECs to even higher susceptibility than healthy cells. We found that there was differential expression of Akt3 between healthy and PAH iPSC-ECs following e-cig exposure. Knocking down Akt3 in healthy iPSC-ECs mimicked the PAH cell phenotype following e-cig exposure, demonstrating that e-cig treatment significantly enhanced autophagic flux impairment as well as endothelial dysfunction. Finally, pharmacologic induction of autophagy via direct inhibition of mTORC1 and indirect activation of mTORC2 with rapamycin reverses e-cig-induced decreased Akt3 expression, endothelial dysfunction, autophagic flux impairment, and decreased cell viability and migration in PAH iPSC-ECs. These data delineate the involvement of the Akt3/mTOR-dependent pathway in hyperactivated autophagy in patients with PAH who can potentially be targeted for future PAH drug development.

Experimental Procedures

Detailed methods and supporting data are available in the Supplementary Materials.

Generation of E-cig Aqueous Aerosol Extract (AqE)

E-cig AqE was generated from a commercial e-liquid (Vape Dudes, menthol flavor, 24 mg/mL nicotine, 50% PG/50% VG) by puffing a pen-style e-cig (Innokin Endura T18 start kit, 5.5 V, 1.5 Ohm) on a vaping apparatus under the Cooperation Centre for Scientific Research Relative to Tobacco (CORESTA) Method #81 vaping regime of a square-wave puff profile, a 55 mL puff volume, 3 s puff duration, and 30 s intervals.¹⁹ A stock solution of 6 TPE (1 TPE = 1 puff/mL of culture media) was produced by bubbling 24 puffs of e-cig through 4 mL of EGM2 media using a glass impinger.²⁰ The AqE was then immediately passed through a 0.22 μm filter to remove possible contamination, and aliquots were either placed at 2-8 ºC and used within 3 h or stored at −80 ºC until testing.

Culture and Differentiation of iPSC-ECs

This study was approved by the University of Arizona Institutional Review Board. The 2 human PAH iPSC lines (SCVI-29 and SCVI-46) were obtained from the Stanford Cardiovascular Institute Biobank. Characterization of both cell lines was performed in a previous study,²¹ and in this study, abnormalities in PAH pulmonary arterial endothelial cell (PAEC) function were recapitulated in PAH iPSC-ECs, suggesting that PAH iPSC-ECs are useful surrogates and could be used for disease modeling. The control lines were derived from 2 healthy donors, and characterization of these lines was performed in our previous study.²² Differentiation of iPSC-ECs was performed using a protocol as described previously²³ (Supplementary Fig. S1). In brief, iPSCs (over passage 20) were plated at a low density with ROCK inhibitor (Y-27632). The next day, EC differentiation was initiated by N2B27 medium supplemented with 6 μM CHIR and 25 ng/mL BMP4, and the medium was left unchanged for 2 days. On day 2, the cells were cultured in StemPro-34 medium supplemented with 2 μM forskolin and 100 ng/mL VEGF, and the medium was replaced daily. The CD144-positive cells were then magnetically sorted on day 6 using a MicroBeads Kit (Miltenyi Biotec). iPSC-ECs were then cultured in EGM2 medium (Lonza) at 37 °C and 5% CO₂ in a humidified incubator with medium change every 2 days. In the present study, all experiments in this manuscript were carried out on iPSC-ECs between passages 2 and 5.

Statistical Analysis

SigmaPlot (Systat Software, Inc.) was used for statistical analyses. Data are presented as the mean ± SEM from 3 independent experiments, each measured in triplicate. Comparisons among the groups were determined using one-way ANOVA followed by Dunnett’s post hoc test. A value of P < .05 was considered statistically significant.

Results

PAH iPSC-ECs Show Increased Susceptibility to E-cig-Induced Vascular Toxicity

To test whether e-cig predisposes the familial form of PAH (FPAH) iPSC-ECs (hereafter referred to as PAH) to even higher susceptibility than healthy cells, iPSC-ECs differentiated from 2 healthy individuals and 2 PAH patient-derived iPSCs (Supplementary Fig. S1) were first exposed to aqueous aerosol extract (AqE) generated from e-cig aerosols (hereafter referred to as e-cig) at varying concentrations (0, 0.2, 0.6, 0.8, 1, and 2 total puff equivalents (TPE)²⁴) for 48 h. Following exposure to e-cig, a dose-dependent decrease in cell viability was observed in PAH iPSC-ECs, whereas no statistically significant effect of e-cig on healthy iPSC-ECs was detected except at higher concentrations of 1 and 2 TPE (Fig. 1A). In addition, the potential of e-cig to induce intracellular reactive oxygen species (ROS) production was evaluated by measuring H₂O₂ levels in cells after exposure to increasing doses of e-cig. As shown in Fig. 1B, we observed significantly higher intracellular ROS levels in PAH iPSC-ECs than in healthy iPSC-ECs even without e-cig treatment, which was further heightened by 48 h of e-cig treatment with 0.6 or 1 TPE, while only the highest e-cig dose (1 TPE) affected healthy iPSC-ECs. Based on these results, the optimal concentration for e-cig was set to 0.6 TPE for subsequent experiments, as this concentration showed approximately 85% and 70% viability of PAH iPSC-ECs after 24 (Supplementary Fig. S2A) and 48 h of treatment (Fig. 1A), respectively. As migration of ECs is a crucial step of angiogenesis, we then assessed the influence of e-cig on EC functions. Compared with healthy iPSC-ECs, cell migration (Fig. 1C) was similarly impaired in PAH iPSC-ECs. We also observed significantly higher apoptotic cell populations (annexin V positive/PI positive and annexin V positive/PI negative) in PAH iPSC-ECs than in healthy iPSC-ECs even without e-cig treatment, which was further heightened by 24 h of e-cig exposure (Fig. 1D).

Figure 1.

PAH iPSC-ECs show more severe endothelial dysfunction than healthy iPSC-ECs following exposure to e-cig. The effects of e-cig on (A) cell viability and (B) intracellular ROS levels after 48 h of treatment (n = 5). Data are represented as the mean ± SEM. ^*P < .05 vs. control within each group and ^#P < .05 vs. healthy iPSC-ECs. (C) Migration assays with representative images and quantification data (n = 3). Data are expressed as the mean ± SEM. ^*P < .05 vs. 0 h within each group and ^#P < .05 vs. healthy controls at 12 h. (D) Representative images of the percentage of apoptosis or necrosis in healthy or PAH iPSC-ECs treated with either vehicle or e-cig analyzed by Annexin V/PI staining (n = 3). ^*P < .05.

PAH iPSC-ECs Have Distinct Transcriptomic Profiles Upon Exposure to E-cig

To illuminate the mechanisms underlying the increased susceptibility of PAH iPSC-ECs compared to healthy cells upon exposure to e-cig, we examined their respective transcriptomes by carrying out RNA sequencing (RNA-seq) on both healthy and PAH iPSC-ECs treated with e-cig for 24 h. Unsupervised analysis using principal component analysis (PCA; Fig. 2A) and ggplot2 line plot (Fig. 2B) revealed a distinct profile change between patients with PAH and healthy groups. In Fig. 2B, we attempted to identify the genes that had differential changes between disease condition (PAH) and e-cig treatment (T: treated) simultaneously. E-cig treatment further contributed to increased cluster separation, where cluster 1 indicates the e-cig-induced downregulated genes and cluster 2 presents upregulated genes in PAH iPSC-ECs. In addition, we found a significant opposing regulation of a total of 66 differentially expressed genes (DEGs), including 23 upregulated DEGs and 43 downregulated DEGs, in PAH iPSC-ECs after exposure to e-cig compared to healthy iPSC-ECs (Fig. 2C and Supplementary Table S1). Interestingly, among these DEGs, Akt3, which is involved in the angiogenic response and autophagy,²⁵ was found to be unaffected in healthy cells but downregulated in PAH ECs upon e-cig exposure, and these RNA-seq results were confirmed by real-time RT-PCR (Fig. 2D). Consistent with these results, Western blotting (Fig. 2E) further confirmed that Akt3 levels were markedly decreased in PAH iPSC-ECs following e-cig exposure, while Akt1 and Akt2 remained unaffected by e-cig (Supplementary Fig. S2B), suggesting that e-cig-induced Akt-dependent signaling is driven primarily by Akt3. In addition, Reactome pathway (Fig. 2F) and gene ontology (GO) enrichment analyses (Supplementary Fig. S3) of DEGs allowed us to understand e-cig-associated regulation of gene expression between the 2 groups. The downregulated genes in PAH iPSC-ECs treated with e-cig (cluster 1) were intrinsically linked with signaling in viral protein R (Vpr)-mediated induction of apoptosis by mitochondrial outer membrane permeabilization, cysteine formation from homocysteine, intracellular oxygen transport, immune system, influenza virus-induced apoptosis, and Akt-mediated inactivation of FOXO1A, whereas the e-cig-induced upregulated genes in PAH iPSC-ECs (cluster 2) were associated with cytosolic tRNA aminoacylation, gene and protein expression by JAK-STAT signaling after interleukin-12 stimulation, Fanconi anemia pathway, and tRNA aminoacylation (Fig. 2F). Recent reports raise the possibility that FOXO protein family members such as FOXO1 and FOXO3 are involved in promoting autophagy,^26,27 suggesting that autophagy might be affected in PAH iPSC-ECs after e-cig treatment.

Figure 2.

Transcriptomic profiling reveals differential regulation of Akt3 signaling between healthy and PAH iPSC-ECs exposed to e-cig. (A) Principal component analysis of gene expression across 2 healthy and 2 PAH iPSC-ECs treated with either vehicle or e-cig. H: healthy iPSC-ECs with vehicle treatment, HT: healthy iPSC-ECs with e-cig treatment; P: PAH iPSC-ECs with vehicle treatment, PT: PAH iPSC-EC with e-cig treatment. (B) Line plots and (C) heatmap of differentially expressed gene (DEG) clusters between vehicle and e-cig treatments in healthy and PAH iPSC-ECs. T: e-cig treatment. U: vehicle-treated controls. (D and E) Healthy and PAH iPSC-ECs were exposed to e-cig for 24 h. Quantitative real-time RT–PCR and Western blotting were performed to evaluate the expression level of Akt3 (n = 3). Data are expressed as the mean ± SEM. ^*P < .05. (F) Pathway analysis of DEGs in PAH iPSC-ECs treated with e-cig.

E-cigs Inhibit Autophagic Flux in ECs

As emerging evidence suggests that autophagy plays an important role in cardiovascular diseases as well as pulmonary diseases,²⁸ we investigated whether e-cig affects endothelial autophagy. Following exposure to e-cig for 24 h, we found that e-cig induced autophagy impairment in both healthy and PAH iPSC-ECs. Significantly higher p62 levels were observed in healthy iPSC-ECs treated with e-cig, indicating impairment of autophagic degradation (Fig. 3A). In addition, impaired autophagic flux in PAH iPSC-ECs was evidenced by significantly enhanced expression of both LC3-II and p62, which was more pronounced when cells were treated with the autophagosome-lysosome fusion inhibitor bafilomycin A1 (BafA1) (Fig. 3A). This was further confirmed by immunofluorescence staining (Fig. 3B), and this enhanced impairment of autophagic flux in PAH iPSC-ECs was accompanied by increased EC apoptosis, which was determined with Annexin V/PI staining (Fig. 3C).

Figure 3.

E-cigs inhibit autophagic flux in healthy and PAH iPSC-ECs. (A) The effects of e-cig on the expression levels of LC3-II and p62 with or without bafilomycin A1 (BafA1) analyzed by Western blot (n = 5). Data are expressed as the mean ± SEM. ^*P < .05 vs. control within each group; ^#P < .05 vs. healthy iPSC-ECs; and ^!P < .05 vs. healthy iPSC-ECs or PAH iPSC-ECs treated with e-cig. (B) Representative images and quantitative analysis of the immunofluorescence staining of p62 after treating cells with either vehicle or e-cig for 24 h (n = 5). Cell nuclei were stained with 4ʹ,6-diamidino-2-phenylindole (DAPI). Data are expressed as the mean ± SEM. ^*P < .05. (C) The effects of e-cig treatment on the apoptosis or necrosis of healthy and PAH iPSC-ECs with or without BafA1 were analyzed by flow cytometry (n = 3). Data are expressed as the mean ± SEM. ^*P < .05 vs. control within each group; ^#P < .05 vs. healthy iPSC-ECs; and ^!P < .05 vs. PAH iPSC-ECs treated with e-cig.]

E-cig-Induced Akt3 Deficiency is Accompanied by Autophagic Flux Impairment and Contributes to Endothelial Dysfunction

As our RNA-seq analysis and Western blot analysis indicated that Akt3 was unaffected in healthy cells but decreased in PAH iPSC-ECs upon e-cig exposure, we assessed whether attenuated Akt3 expression caused by e-cig is responsible for the increased susceptibility of PAH iPSC-ECs. To examine whether Akt3 is an essential requirement as an important regulator of autophagy, which when perturbed leads to endothelial dysfunction, we first knocked down Akt3 in healthy iPSC-ECs using siRNA prior to e-cig exposure (Supplementary Fig. S4). As shown in Fig. 4A; Supplementary Fig. S5, knockdown of Akt3 markedly increased the expression of both LC3-II and p62, indicating impaired autophagy. Akt3 inhibition also led to significantly decreased EC viability (Fig. 4B) and increased ROS production (Fig. 4C) and apoptosis in ECs (Fig. 4D) compared with vehicle-treated controls. In addition, depletion of Akt3 combined with e-cig treatment further increased autophagic flux impairment, reduced cell viability, and increased ROS levels and the percentage of apoptotic cells significantly when compared to siAkt3-treated cells (Fig. 4A–4D). We also noted that tube formation (Fig. 4E) and cell migration (Fig. 4F) were significantly altered in healthy iPSC-ECs with either Akt3 knockdown or e-cig exposure, and the altered tube formation and migration remained further impaired by the combination of both Akt3 knockdown and e-cig.

Figure 4.

Akt3 deficiency coupled with e-cigs induces more endothelial dysfunction. Healthy iPSC-ECs were transfected with either negative control or siRNA targeting Akt3 for 24 h followed by exposure to e-cig for 24 h. (A) Representative immunoblots of Akt3, p62, and LC3 by Western blot (n = 3). The effects of siAkt3 on (B) cell viability and (C) intracellular ROS levels after treating cells with either vehicle or e-cig for 24 h (n = 5). (D) Cell apoptosis and necrosis with representative images and quantification data. Flow cytometry was used to detect apoptotic cells stained by Annexin V-FITC/PI (n = 3). Representative images and quantitative analyses from (E) tube formation (n = 5) and (F) migration assays (n = 3). Data are expressed as the mean ± SEM. ^*P < .05.

Akt3-Depleted iPSC-ECs Increase Susceptibility to Autophagic Flux Impairment, Which Might be Mediated Through the mTOR Pathway

Previous studies have shown the importance of the Akt/mTOR pathway in tobacco-carcinogen-induced lung tumorigenesis.²⁹ To study the possible dysregulation of mTOR in PAH iPSC-ECs after e-cig exposure, we examined mTOR activity. Consistent with previous reports,^30,31 we observed significantly increased levels of p-mTOR and p-S6 in PAH iPSC-ECs compared with healthy iPSC-ECs, and these levels were further increased with e-cig treatment (Supplementary Fig. S6). As we observed that rapamycin but not BafA1 or MG132 (a proteasome inhibitor) promoted Akt3 expression and activation of autophagy in PAH iPSC-ECs after e-cig exposure (Supplementary Fig. S7), we examined whether Akt3-dependent impaired autophagic flux in iPSC-ECs caused by e-cig is associated with increased mTOR activity by transfecting both healthy and PAH iPSC-ECs with siAkt3 prior to treatment with rapamycin (50 nM) for 24 h. We found that rapamycin significantly increased Akt3 expression as well as autophagic flux in both healthy and PAH iPSC-ECs (Fig. 5A). Importantly, the presence of rapamycin also partially restored the levels of Akt3 and activated autophagy in both healthy and PAH cells treated with siAkt3. Moreover, rapamycin treatment significantly blunted the siAkt3-mediated increase in cell apoptosis and necrosis (Supplementary Fig. S8A) and impaired cell migration in healthy iPSC-ECs (Supplementary Fig. S8B).

Figure 5.

E-cigs affect mTOR signaling in healthy and PAH iPSC-ECs. Healthy and PAH iPSC-ECs were transfected with siRNA targeting Akt3 for 24 h and then treated with 50 nM rapamycin for 24 h. (A and B) Representative immunoblots and quantitative analysis of Akt3, p62, LC3, Rictor, Raptor, p-mTOR, mTOR, p-S6, p-Akt, or Akt in healthy and PAH iPSC-ECs treated with/without siAkt3 or/and rapamycin analyzed by Western blot (n = 3). (C) The effects of rapamycin on the expression of Akt3, p62, LC3, p-mTOR, mTOR, p-S6, p-Akt, or Akt in healthy and PAH iPSC-ECs treated with/without siRictor (n = 3). Data are expressed as the mean ± SEM. *P < .05.

To further examine how the enhanced Akt3 expression induced by rapamycin regulates autophagy, we first treated healthy and PAH iPSC-ECs with siAkt3 and evaluated the activation of the 2 distinct protein components of mTOR complexes. In healthy and PAH iPSC-ECs, the phosphorylation of mTOR (p-mTOR) as well as downstream S6 (p-S6) served to assess mTORC1 activity was significantly increased by knockdown of Akt3, whereas the levels of Rictor (a key component of mTORC2) and p-Akt at Ser473 reflecting mTORC2 activity were significantly suppressed by siAkt3 (Fig. 5B). Interestingly, rapamycin treatment significantly prevented the increase in p-mTOR levels and mTORC1-dependent p-S6 seen in both healthy and PAH iPSC-ECs treated with siAkt3, while significantly restoring the levels of Rictor and mTORC2-dependent p-Akt at Ser473 in the siAkt3-treated group. Compared with the control group, the expression level of Raptor was not affected by either the loss of Akt3 or rapamycin treatment in either healthy or PAH iPSC-ECs.

To further determine the role of Rictor in e-cig-induced attenuated Akt3 expression and impaired autophagic flux in PAH iPSC-ECs, we knocked down Rictor in PAH iPSC-ECs, which would result in disruption of mTORC2, and immunoblotted for LC3-II, p62, p-mTOR, mTOR, p-S6, p-AKT at Ser473, and Akt3. Upon the loss of Rictor, the level of Akt3 was significantly reduced in PAH iPSC-ECs, while the levels of p-mTOR and p-S6 as well as the impairment of autophagic flux were significantly increased (Fig. 5C). Treatment with rapamycin also significantly rescued the expression levels of Akt3, autophagic flux, and p-Akt while significantly inhibiting mTOR phosphorylation and p-S6 in PAH iPSC-ECs treated with siRictor. In healthy iPSC-ECs, Rictor knockdown did not affect Akt levels, and rapamycin substantially increased Akt3 expression while inhibiting p-mTOR and p-S6 significantly compared with the siRictor-treated group (Fig. 5C). Taken together, these results suggest that rapamycin-mediated increased Akt3 expression and enhanced autophagy in healthy and PAH iPSC-ECs are associated with inhibition of mTORC1 with simultaneous mTORC2 activation, as reflected by increased p-Akt. In addition, our finding that Rictor remained reduced with even increased p-Akt at Ser473 in PAH iPSC-ECs treated with rapamycin suggests that feedback loop activation of Akt upon rapamycin treatment may utilize another Akt kinase or that Rictor is not essential for mTORC2-mediated p-Akt at Ser473.

Rapamycin Reverses E-cig-Induced Endothelial Dysfunction and Autophagy Impairment in PAH iPSC-ECs

To further determine the role of mTOR inhibition in e-cig-mediated endothelial dysfunction and autophagy impairment in PAH iPSC-ECs, PAH iPSC-ECs were pretreated with 50 nM rapamycin for an hour prior to e-cig exposure, followed by the measurement of Akt3 expression, autophagy, apoptosis, and migration. As shown in Fig. 6A e-cig-induced significantly reduced Akt3 expression and enhanced autophagic flux impairment in PAH iPSC-ECs were reversed upon treatment with rapamycin. Furthermore, rapamycin treatment significantly decreased the endothelial apoptosis rate (Fig. 6B) and improved migration (Fig. 6C) in PAH iPSC-ECs exposed to e-cig compared with the e-cig-treated group, suggesting that mTOR inhibition protects PAH iPSC-ECs against e-cig-induced endothelial dysfunction by activating Akt3-mediated endothelial autophagic flux.

Figure 6.

E-cig-induced endothelial dysfunction and autophagy impairment in PAH iPSC-ECs are ameliorated by rapamycin. PAH iPSC-ECs were pretreated with 50 nM rapamycin (Rap) for 1 h followed by e-cig exposure for 24 h. (A) The effects of rapamycin treatment on the expression of Akt3, p62, and LC3 and (B) apoptosis and necrosis in e-cig-treated PAH iPSC-ECs analyzed by Western blot and flow cytometry, respectively (n = 3). Data are expressed as the mean ± SEM. ^*P < .05 vs. control and ^#P < .05 vs. e-cig. (C) Representative images and quantitative analysis from the migration assay (n = 3). Data are expressed as the mean ± SEM. ^*P < .05 vs. 0 h; ^#P < .05 vs. control at 24 h; and ^!P < .05 vs. e-cig at 24 h.

Discussion

The use of human iPSCs from healthy and PAH subjects represents a cellular human-based model¹⁸ that complements current animal models in understanding how e-cig is linked to endothelial dysfunction in the setting of PAH and its underlying mechanisms. Consistent with our previous report,³² e-cig exposure resulted in endothelial dysfunction in both healthy and PAH iPSC-ECs. Notably, we found that PAH iPSC-ECs were more susceptible to e-cig exposure than healthy iPSC-ECs, supporting the possibility that e-cig might have a particularly maladaptive effect on PAH iPSC-ECs. Transcriptomic analyses between healthy and PAH iPSC-ECs upon exposure to e-cig revealed differential expression of Akt3, demonstrating that this might be responsible for triggering enhanced impaired autophagic flux leading to increased susceptibility of ECs in PAH. Rapamycin treatment prevented e-cig-induced increased apoptosis and impaired cell migration in PAH iPSC-ECs by inhibition of mTORC1 and corresponding activation of mTORC2 via Akt and enhanced autophagic flux.

Autophagy is a highly coordinated intracellular lysosomal-mediated catabolic process that degrades damaged or dysfunctional proteins and intracellular organelles.¹⁴ Although autophagy has been thought of as an essential intracellular response in maintaining EC homeostasis, the role of autophagy in PAH remains inconclusive but appears to play an important role in vascular modeling.³³ EC dysfunction in chronic thromboembolic pulmonary hypertension is mediated by inactivation of autophagy.³⁴ We also found that autophagy impairment caused by e-cig was more pronounced in PAH iPSC-ECs. Importantly, when PAH iPSC-ECs were pretreated with an autophagy inducer, rapamycin, before being exposed to e-cig, the e-cig-induced increased numbers of apoptotic cells and impairment of endothelial function were ameliorated, suggesting a beneficial role of autophagy in PAH iPSC-ECs exposed to e-cig.

Akt has 3 isoforms, Akt1, Akt2, and Akt3, which share more than 80% sequence homology but exhibit distinct isoform-specific functions.³⁵ Although Akt-dependent signaling, exclusively driven by Akt1 activation, is known to be significant for angiogenesis, the role of the individual isoform is less well understood. A functional role of Akt3 in promoting angiogenesis and inhibiting autophagy has been demonstrated in limited studies.^25,36 One study compared the function of Akt1 and Akt3 and reported a direct link between Akt3 and mitochondrial biogenesis that was independent of Akt1.³⁶ Akt3 blockade also caused increased autophagy in human umbilical vein endothelial cells, and Akt3-null mice failed to launch angiogenic responses to growth factors when challenged.²⁵ In the present study, we observed attenuated Akt3 levels in PAH iPSC-ECs but unchanged Akt3 levels in healthy cells upon e-cig exposure. Knockdown of Akt3 expression in healthy iPSC-ECs significantly induced autophagic flux impairment and increased endothelial dysfunction, and these vascular toxicities were further increased when exposed to e-cig, thus mimicking the PAH cell phenotype after e-cig exposure. Interestingly, our results showed that the expression levels of neither Akt1 nor Akt2 in both healthy and PAH iPSC-ECs were affected by e-cig, demonstrating the involvement of Akt3 in regulating endothelial autophagy following e-cig exposure. The exact mechanism underlying how Akt3 is involved in endothelial autophagy remains to be determined.

Activation of the mTOR pathway constitutes a common, targetable cellular process underlying both pulmonary vascular remodeling and right ventricular failure in patients with PAH.³¹ Consistent with a previous report,³⁰ we found that the basal level of mTOR phosphorylation in PAH iPSC-ECs was significantly higher than that in healthy iPSC-ECs, and following exposure to e-cig, a greater increase in mTOR phosphorylation was observed in PAH iPSC-ECs. mTOR serves as a core component of 2 multiprotein complexes, mTORC1 and mTORC2, which promote protein translation while suppressing autophagy.³⁷ While mTORC1 is known to be sensitive to rapamycin, mTORC2 is considered to be insensitive, as only chronic administration can inhibit mTORC2.³⁸ Rapamycin-mediated inhibition of mTORC1 with reciprocal activation of mTORC2 has been shown to provide protection in cardiovascular disease.^39,40 For example, rapamycin protects hearts from maladaptive hypertrophy by increasing mTORC2 signaling as a consequence of mTORC1 inhibition.³⁹ Of note, evidence for a direct role of mTORC1 and mTORC2 in regulating Akt3 protein expression and autophagic flux in PAH ECs after e-cig exposure has, to the best of our knowledge, never been reported. Our results also supported that rapamycin attenuated Akt3-mediated endothelial dysfunction and autophagic impairments in PAH iPSC-ECs upon e-cig exposure, as we demonstrated that rapamycin treatment controls mTOR signaling in favor of mTORC2 while suppressing mTORC1 in iPSC-ECs in vitro.

Despite its many strengths, there are several limitations to this study. First, a relatively small number of cell lines were used, and in future studies, additional cell lines would be beneficial to extrapolate the findings of our study to the general population of patients with PAH. Second, our study used e-cig AqE from menthol-flavored e-liquid, the most common flavor for adults,⁴¹ and the analysis was conducted after acute exposure. As there is enormous variability in e-cig products, including different hardware and e-liquid components (eg, PG/VG, flavorings, cooling agents, and nicotine), leading to the delivery of highly variable amounts of nicotine and potentially toxic substances, our study may not recapitulate what a typical e-cig user might experience during vaping. The current scope of the study serves as an initial proof-of-concept to define whether there is a difference in the susceptibility of healthy and PAH iPSC-ECs to e-cig in general. Further research is needed to explore the effects of diverse e-cig devices and individual components on endothelial autophagy to better understand the potential risks of toxic compounds produced by e-cig products.

In addition, even though rapamycin treatment did not affect the protein levels of the 2 key subunits of mTORC1 (Raptor) and mTORC2 (Rictor), the formation of these complexes might have changed. For example, rapamycin increased the recruitment of Rictor to mTOR, resulting in increased formation of p-Akt at serine 473 despite unchanged protein levels. Studies of protein–protein interactions are beyond the scope of the present study.

Conclusion

Considering the rapid growth of global e-cig consumption, it is vastly important to study the potential adverse effects of e-cigs on lung and vascular health. Collectively, our study provides the first evidence defining why patients with PAH are at increased risk when using e-cigs compared to healthy users and the link between e-cigs and endothelial dysfunction regarding differential regulation of autophagy between healthy and PAH iPSC-ECs. Our data also support the notion that e-cig-induced endothelial dysfunction is due to elevated autophagic flux impairment. Moreover, we show that activating mTORC2 signaling via rapamycin can improve impaired autophagic flux and inhibit the degradation of Akt3 in PAH iPSC-ECs upon exposure to e-cig. Based on these findings, understanding the signaling pathways implicated in e-cig-mediated autophagy provides new insights: (1) the effects of e-cig on endothelial autophagy and (2) how perturbations in endothelial autophagy via Akt3 increase susceptibility to e-cigs in patients with PAH, which can potentially be targeted for future PAH drug development.

Funding

This work was supported by the American Heart Association (AHA) Scientist Development Grant 16SDG27560003 (to Dr. Lee), the AHA postdoctoral fellowship 20POST35200257 (to Dr. Liu), and the National Institutes of Health R00HL130416 and R01HL148756 (to Dr. Ong).

Conflict of Interest

The authors declared no potential conflicts of interest.

Author Contributions

C.L.: conception and design, collection and/or assembly of data, data analysis and interpretation, financial support, and manuscript writing; H.H.T.L., P.D. III, and N.Y.S.: collection and/or assembly of data; Z.D.: manuscript revision; S.G.O.: conception and design, data analysis and interpretation, financial support, and manuscript writing; W.H.L.: conception and design, data analysis and interpretation, financial support, manuscript writing, and final approval of manuscript.

Data Availability

The raw sequence data were deposited at the NCBI Sequence Read Archive under accession number GSE199764.