Amphetamines promote mitochondrial dysfunction and DNA damage in pulmonary hypertension

Abstract

Amphetamine (AMPH) or methamphetamine (METH) abuse can cause oxidative damage and is a risk factor for diseases including pulmonary arterial hypertension (PAH). Pulmonary artery endothelial cells (PAECs) from AMPH-associated-PAH patients show DNA damage as judged by γH2AX foci and DNA comet tails. We therefore hypothesized that AMPH induces DNA damage and vascular pathology by interfering with normal adaptation to an environmental perturbation causing oxidative stress. Consistent with this, we found that AMPH alone does not cause DNA damage in normoxic PAECs, but greatly amplifies DNA damage in hypoxic PAECs. The mechanism involves AMPH activation of protein phosphatase 2A, which potentiates inhibition of Akt. This increases sirtuin 1, causing deacetylation and degradation of HIF1α, thereby impairing its transcriptional activity, resulting in a reduction in pyruvate dehydrogenase kinase 1 and impaired cytochrome c oxidase 4 isoform switch. Mitochondrial oxidative phosphorylation is inappropriately enhanced and, as a result of impaired electron transport and mitochondrial ROS increase, caspase-3 is activated and DNA damage is induced. In mice given binge doses of METH followed by hypoxia, HIF1α is suppressed and pulmonary artery DNA damage foci are associated with worse pulmonary vascular remodeling. Thus, chronic AMPH/METH can induce DNA damage associated with vascular disease by subverting the adaptive responses to oxidative stress.

Keywords: Cell Biology; Vascular Biology.

Figure 1. Unrepaired DNA damage in PAECs isolated from D&T-PAH patients.

Pulmonary artery endothelial cells (PAECs) were isolated from lung explants of unused control donor (Donor-Ctl, n = 4) and from patients with pulmonary arterial hypertension (PAH) associated with amphetamine (AMPH), methamphetamine (METH) or anorexigens (D&T-PAH, n = 4, including 3 AMPH-PAH and 1 anorexigen-PAH), as described under Methods. PAECs were treated in full medium with vehicle (Veh) or doxorubicin (Dox) for 6 hours. A separate set of Dox-treated cells was allowed to recover (Dox-recovery) for 12 hours in A, or for 6 hours in B. (A) Cells were analyzed by alkaline comet assay. Representative images show intact DNA (comet head), and comet tails, indicative of damaged DNA. The percentage of DNA in the tail of 100 to 150 cells per subject was analyzed using ImageJ. Scale bar: 60 μm. (B) Cells were immunostained for γH2AX and DAPI to assess DNA damage. Representative images show DNA damage foci in the nuclei. Scale bar: 30 μm. Number of γH2AX foci per cell was quantified using ImageJ. Ten to 15 confocal images were assessed per subject. Box-and-whisker plots represent values within the interquartile range (boxes) and the minimum to maximum (whiskers). The line within the box shows the median. n = 4, Donor-Ctl PAECs, and n = 4, D&T-PAH PAECs, each line of PAEC has 3–4 replicate images. ****P < 0.0001 vs. Donor-Ctl+Veh; ^##P < 0.005 and ^####P < 0.0001 vs. Donor-Ctl+Dox; ^&&&&P < 0.0001 vs. D&T-PAH+Dox; ^$$$$P < 0.0001 vs. Donor-Ctl+Dox-recovery, by 2-way ANOVA, Bonferroni’s post-test.

Figure 2. Amphetamine exacerbates genotoxic stress–induced DNA damage.

(A) Commercially available human pulmonary artery endothelial cells (PAECs) were treated in full medium with indicated doses of amphetamine (AMPH) daily for 3 days, and then stimulated with doxorubicin (Dox, 0.2 μg/ml) for 24 hours in the presence of vehicle (Veh) or AMPH. Cell lysates were immunoblotted for γH2AX and β-actin (for normalization). (B) PAECs were treated with 1 mM AMPH daily for 3 days, and then stimulated with Dox in the presence of Veh or AMPH for 6 hours. Cells were analyzed by comet assay as described for Figure 1. Scale bar: 60 μm. (C) Cells treated as in B were immunostained for γH2AX (red), pDNA-PK (green), and DAPI (blue). Ten to 15 confocal images with at least 40 cells per image per condition were analyzed using ImageJ. Scale bar: 15 μm. (D) PAECs were treated daily for 3 days with the indicated dose of AMPH, and then cultured with Veh or AMPH under 0.5% O₂ hypoxia (Hx) or normoxia (Nx) for 48 hours. Cell lysates were assayed as in A. (E) Comet assay of PAECs treated with 0.5 mM AMPH and Hx/Nx as in D. Scale bar: 60 μm. (A and D) Dot plots represent mean ± SEM, n = 3–4. (B, C, and E) Box-and-whisker plots represent values within the interquartile range (boxes) and the minimum to maximum (whiskers). The line within the box shows the median. n = 3 independent experiments with 3 to 5 replicates per experiment. (A–C) *P < 0.05, **P < 0.005 vs. Veh; ^##P < 0.005 vs. Dox+Veh. (D and E) *P < 0.05, **P < 0.005, ***P < 0.0005 vs. Nx+Veh; ^#P < 0.05, ^##P < 0.005, ^####P < 0.0001 vs. Hx+Veh or Vec+Hx+Veh; ^&P < 0.05, ^&&P < 0.005, ^&&&&P < 0.0001 vs. Hx+AMPH or Vec+Hx+AMPH, by 2-way ANOVA, Bonferroni’s post-test.

Figure 3. Amphetamine exaggerates hypoxia-induced DNA damage in a p-Akt– and caspase-3–dependent manner.

(A) Pulmonary artery endothelial cells (PAECs) were treated daily for 3 days with the indicated 0.5 mM amphetamine (AMPH), and then cultured with vehicle (Veh) or AMPH under 0.5% O₂ hypoxia (Hx) or normoxia (Nx) for 48 hours, immunoblotted for p-Akt, cleaved caspase-3 (cCasp3), and β-actin (loading control). (B) PAECs were transfected with vector (Vec) or HA-tagged myristoylated Akt1 (myr-Akt) construct. Twenty-four hours after transfection, cells were treated with AMPH and Hx/Nx as in A. Cell lysates were immunoblotted for γH2AX, cCasp3, β-actin (for normalization), and HA, to verify the expression of HA-myr-Akt. (C) PAECs were treated as in A together with the indicated concentrations of zVAD-FMK, and lysates were immunoblotted for γH2AX, cCasp3, β-actin (loading control), and p-Akt. Dot plots represent mean ± SEM, n = 3–4. (A–C) *P < 0.05, **P < 0.005, ****P < 0.0001 vs. Nx+Veh; ^#P < 0.05, ^##P < 0.005, ^####P < 0.0001 vs. Hx+Veh or Vec+Hx+Veh; ^&P < 0.05, ^&&P < 0.005, ^&&&&P < 0.0001 vs. Hx+AMPH or Vec+Hx+AMPH; by 2-way ANOVA, Bonferroni’s post-test.

Figure 4. Amphetamine-induced SIRT1 stabilization under hypoxia exaggerates caspase-3 activation and DNA damage (see also Supplemental Figure 3).

Pulmonary artery endothelial cells (PAECs) were treated daily with 0.5 mM amphetamine (AMPH) for 3 days and then cultured with vehicle (Veh) or AMPH under normoxia (Nx) or hypoxia (Hx) for (A) 48 hours or (B) 8 hours. Cells were (A) immunoblotted for acetylated histone 3 (H3K9Ac), histone 4 (H4K8Ac), and β-actin (for normalization) or (B) extracted for RNA and then screened for histone deacetylase (HDAC) expression via quantitative real-time PCR (qRT-PCR). qRT-PCR results are displayed as a heatmap with blue to yellow indicating low to high levels relative to Veh for each condition. (C) PAECs were treated as described in A, with or without sirtinol (10 μM) during the treatment. Cell lysates were immunoblotted for γH2AX, cleaved caspase-3 (cCasp3), and β-actin (loading control). (D) PAECs were transfected with 20 nM control (siCtl) or SIRT1 (siSIRT1) siRNA. Twenty-four hours after transfection, cells were treated as in A and analyzed for SIRT1, p-Akt, γH2AX, cCasp3, and β-actin (loading control). Dot plots represent mean ± SEM, n = 3–4. *P < 0.05, **P < 0.005, ***P < 0.0005, ****P < 0.0001 vs. (siCtl or Vec)+Nx+Veh; ^##P < 0.005, ^####P < 0.0001 vs. (siCtl or Vec)+Hx+Veh; ^$$P < 0.005 vs. siSIRT1+Hx+Veh; ^&P < 0.05, ^&&&&P < 0.0005 vs. (siCtl or Vec)+Hx+AMPH; by 2-way ANOVA, Bonferroni’s post-test.

Figure 5. Suppression of p-Akt by amphetamine stabilizes SIRT1 under hypoxia (See also Supplemental Figure 3).

(A) Pulmonary artery endothelial cells (PAECs) were treated with 0.5 mM amphetamine (AMPH) for the indicated times under normoxia (Nx) or hypoxia (Hx), and cell lysates were immunoblotted for phospho-SIRT1 (p-SIRT1), SIRT1, p-Akt, p-JNK, and β-actin (loading control). (B) PAECs were transfected with vector (Vec) or HA-tagged myristoylated Akt1 (myr-Akt) construct. Twenty-four hours after transfection, cells were treated daily with 0.5 mM AMPH for 3 days and then cultured with vehicle (Veh) or AMPH under Nx or Hx for 48 hours, and SIRT1, myr-Akt, and β-actin (loading control) protein levels were analyzed by immunoblotting. Dot and line graphs represent mean ± SEM, n = 3–4. **P < 0.005, ***P < 0.0005 vs. Nx+Veh; ^##P < 0.005 vs. Hx+Veh; ^&P < 0.05, ^&&&P < 0.0005 vs. Hx+AMPH; by 2-way ANOVA, Bonferroni’s post-test.

Figure 6. Stabilization of SIRT1 by amphetamine deacetylates and destabilizes HIF1α, impairing its transcriptional activity.

(A) Pulmonary artery endothelial cells (PAECs) were treated daily with vehicle (Veh) or the indicated dose of amphetamine (AMPH) for 3 days and then continued with vehicle or AMPH under normoxia (Nx) or hypoxia (Hx) for 48 hours. Cell lysates were immunoblotted for HIF1α and β-actin (for normalization). (B) PAECs were treated daily with 0.5 mM AMPH or AMPH with sirtinol (10 μM) for 3 days, and then cultured with the same stimuli under Nx or Hx for 24 hours. Cell lysates were immunoprecipitated with anti–acetyl-lysine antibody or IgG as negative control. Immunoprecipitates were probed for HIF1α to evaluate the levels of acetylated HIF1α. Whole-cell lysates were immunoblotted to determine the levels of total HIF1α and β-actin (loading control). A representative experiment is shown; 3 independent experiments were conducted with similar results. (C) PAECs were transfected with control or SIRT1 siRNAs, treated as in A and analyzed for HIF1α levels. (D) Cignal HIF firefly luciferase reporter and Renilla luciferase plasmids were cotransfected with control or SIRT1 siRNAs into PAECs. Twenty-four hours after transfection, cells were treated with 0.5 mM AMPH as in A and assessed for HIF activity, shown as the ratio of Firefly- to Renilla-luciferase-catalyzed light emission. (E and F) PAECs were treated as in B but with only 8 hours of Nx/Hx; RNA was extracted, and PDK1 and COX4I1 mRNA evaluated by qRT-PCR. (G) PAECs transfected and treated as in C were analyzed for PDK1 and COX4I1 protein levels. In C and G, β-actin was probed as a loading control, and SIRT1 to verify SIRT1 knockdown efficiency. (A, C, and G) Dot plots represent mean ± SEM, n = 3–4. (D, E, and F) Box-and-whisker plots represent values within the interquartile range (boxes) and the minimum to maximum (whiskers). The line within the box shows the median. n = 3 independent experiments with 2 to 3 replicates per experiment. *P < 0.05, **P < 0.005, ***P < 0.0005, ****P < 0.0001 vs. Nx+Veh or siCtl+Nx+Veh; ^#P < 0.05, ^##P < 0.005, ^###P < 0.0005 vs. Hx+Veh or siCtl+Hx+Veh; ^&P < 0.05 vs. Hx+AMPH or siCtl+Hx+AMPH; by 2-way ANOVA, Bonferroni’s post-test.

Figure 7. In PAECs under hypoxia, amphetamine impairs the transition from oxidative phosphorylation to glycolysis.

(A) Pulmonary artery endothelial cells (PAECs) were treated with vehicle (Veh), 0.5 mM amphetamine (AMPH), or AMPH with 10 μM sirtinol, and placed in hypoxia (Hx) for 24 hours with the stimuli. Extracellular acidification rates (ECARs) and oxygen consumption rates (OCRs) were measured using kits (Seahorse Bioscience). (B) Glycolysis, glycolytic reserve, and glycolytic capacity as well as baseline respiration, maximal respiration, and spare respiratory capacity. (C) Representative live-cell images of PAECs treated as in A, loaded with 100 nM MitoTracker Green FM to stain mitochondria, and 5 μM MitoSOX Red to detect mitochondrial superoxide. Scale bar: 10 μm. (D) Fluorescence intensities of MitoTracker and MitoSOX were quantified using ImageJ, and ROS production was expressed as the ratio of MitoSOX to MitoTracker fluorescence. (E) PAECs were treated as in A. Mitochondrial membrane potential was determined using the JC-1 dye assay. (F) PAECs were treated with 0.5 mM AMPH, or AMPH with 10 μM of antimycin A (AMA), and then cultured with the stimuli under normoxia (Nx) or Hx for 48 hours. Cell lysates were immunoblotted for γH2AX, cCasp3, β-actin, and H2AX (loading controls). (A and F) Line and dot plots represent mean ± SEM, n = 3 independent experiments. (B, D, and E) Box-and-whisker plots represent values within the interquartile range (boxes) and the minimum to maximum (whiskers). The line within the box shows the median. n = 3 independent experiments with 3 to 4 replicates per experiment. **P < 0.005, ***P < 0.0005 vs. Nx+Veh; ^#P < 0.05, ^##P < 0.005, ^###P < 0.0005, ^####P < 0.0001 vs. Hx+Veh; ^&P < 0.05, ^&&P < 0.005, ^&&&P < 0.0005, ^&&&&P < 0.0001 vs. Hx+AMPH; by 2-way ANOVA, Bonferroni’s post-test.

Figure 8. Proposed model.

Under hypoxia, stabilization of HIF1α is augmented by acetylation due to decreased sirtuin 1 (SIRT1) activity. HIF1α activates pyruvate dehydrogenase kinase 1 (PDK1) expression and cytochrome c oxidase 4 (COX4) isoform switch, thereby facilitating the transition from oxidative phosphorylation to glycolysis and optimizing electron transport during hypoxia. Elevation of SIRT1 by amphetamine (AMPH) under hypoxia, due to protein phosphatase (PP)–mediated lowering of p-Akt, suppresses HIF1α-mediated PDK1 expression and the COX4 isoform switch, thereby increasing mitochondrial ROS production, cytochrome c (CytoC) leakage, caspase-3 (Casp-3) activation, and ultimately DNA damage via caspase-activated DNases (CADs).

Figure 9. Methamphetamine and hypoxia increase γH2AX foci in mouse pulmonary arteries (PAs) and pulmonary vascular remodeling and impaired HIF1α gene regulation.

Mice were treated with 10 mg/kg methamphetamine (METH) twice daily for 3 days in room air, and then exposed to 10% O₂ hypoxia for 4 days, and the cycle repeated for a total of 4 weeks. (A and B) Mouse lung sections were immunostained for von Willebrand factor (vWF, green), DAPI (blue), and α-smooth muscle actin (α-SMA) (A) or γH2AX (B) (red), to indicate the PA endothelial cell layer, nuclei, muscularization, and DNA damage foci, respectively. (A) Representative images of muscularized distal PAs (DPAs), indicated by arrows. Scale bar: 80 μm. Scatter plot on the right shows the percentage of fully, partially, and nonmuscularized DPAs, scored based on 4 to 5 confocal images taken for each mouse. (B) Representative images of PAs immunostained for γH2AX. Scale bar: 20 μm. Insets show magnified areas of endothelial cells with γH2AX foci. Scale bar; 8 μm. Right, γH2AX foci were scored using ImageJ, in 10 to 15 confocal images of PAs for each mouse. (C) Lung homogenates from the mice were immunoblotted for SIRT1, HIF1α, p-Akt, PDK1, α-SMA, PDK1, cCasp3, COX4I1, and β-actin (loading control). Each lane represents lung lysate of one mouse. Dot plots in A–C represent mean ± SEM; n = 5, vehicle-treated (Veh) group and n = 5–6, METH group. *P < 0.05, **P < 0.005, ****P < 0.0001 vs. vehicle by unpaired t test.