P. strombulifera and T. absinthioides were collected in Lavalle County, Mendoza, Argentina (33°44′10′′S, 68°21′30.5′′W), and the voucher specimens were deposited in the Mendoza Ruiz Leal Herbarium as MERL-61824 (Ps) and MERL-61823 (Ta). Decoctions of P. strombulifera and T. absinthioides were prepared at 10% weight/volume, from 100 g of dried and milled plant (leaves), in 1 L of purified water. After 30 min of boiling, the decoctions were filtered and cooled for 24 h in a freezer at −20°C, and then three representative samples of 100 ml of each decoction were subsequently lyophilized in LA-B3 RIFICOR equipment (Buenos Aires, Argentina). The samples were stored in a freezer at −20°C until its use in the assays included in this work, in phenolic and flavonoid quantification, and in UHPLC–PDA–OT–MS analysis. Detailed phytochemical analyses were published by Persia et al. (2020) and Gomez et al. (2019).

VSMCs were isolated from aortas of 12-week-old Wistar rats according to a previously described technique (Castro et al., 2010). More details in Supplementary Material.

Plated in 24-well plates were 1 × 10⁴ VSMCs and serum-starved for 24 h. Then, cells were treated with 2.5, 5, 10, 20, or 40 (μg/ml) of AEPS or AETA in either 0.1 or 10% fetal calf serum (FCS) DMEM/F12 during 24–48 h. Cell behavior was measured using a CellTiter Non-radioactive cell proliferation Assay Kit (G4001; Promega, Madison, WI, United States). More details in Supplementary Material.

Serum-starved VSMCs were stimulated with growth media (10% FCS) alone or co-incubated with AEPs or AETa (5 and 20 μg/ml) during 24 h. Then, cells were harvested and fixed with ice-cold 70% ethanol overnight. The fixed cells were centrifuged and treated with 0.1 mg/ml of RNase A and 50 μg/ml of propidium iodide for 30 min at room temperature. To determine the cell population distribution of fluorescence intensity, a FACS Aria II flow cytometer (Becton Dickinson, Franklin Lakes, New Jersey, United States) was used. For each sample, at least 10,000 nuclei were analyzed. DNA histograms were analyzed by FCS express 7.06.0015 DeNovo software. Two separate experiments were performed with similar results.

Intracellular ROS levels were measured using the FluoroProbe CM-H₂DCFDA (C400; Invitrogen, Carlsbad, CA, United States). VSMCs were cultured in a 24-well plate until confluence and then serum deprived for 24 h. AEPs and AETa (2.5, 5, 10, 20, or 40 μg/ml) were applied during 1 h. Basal fluorescence was measured, and then 100 nM of angiotensin II (AngII) (A9525; Sigma-Aldrich Corp., St. Louis, MO, United States) was added. Fluorescence of stimulated cells was measured continuously during 40 min on a microplate fluorometer (Fluoroskan Ascent, Labsystems, Helsinki, Finland), λ_ex = 485 nm and λ_em = 538 nm.

To determine intracellular O₂^– production in VSMCs, dihydroethidium (DHE; Invitrogen) staining was performed. VSMCs were pre-incubated for 1 h with AEPs or AETa (5 or 20 μg/ml), stimulated with AngII (100 nM, 2 h), and incubated with DHE (5 μM) for 30 min at room temperature. Cells were imaged by fluorescence microscopy (Eclipse TE300 inverted microscopy, Nikon Instruments Inc., Melville, NY, United States), and the signal was quantized with ImageJ 1.50i software. To confirm that cell incubation with extracts affects ROS generation, control experiments with antioxidant agents including apocynin (a ROS scavenger; 100 μmol/L) and diphenylene iodonium chloride (DPI; a flavoprotein inhibitor) 10 μmol/L were carried out.

Total RNA was isolated with TRIzol (Invitrogen, 15596026) from VSMCs stimulated with AngII during 4 h and previously treated with or without AEPs or AETa (20 μg/ml). Real-time qPCR was performed with cDNA samples and EVA Green (31000; Biotium, Fremont, CA, United States) using a Rotor-Gene 6000 Series Software version 1.7 (Corbett Life Science, Sydney, NSW, Australia). The gene expression levels of NADPH oxidase (NOX) subunits NOX2 and NOX4 were normalized to 18S. The mRNA levels were expressed as a ratio, using the delta-delta method for comparing relative expression results between treatments. Relative expression was calculated as 2^–ΔΔCt. More details in Supplementary Material.

VSMCs were pretreated with AEPs or AETa (20 μg/ml) and then stimulated during 20 h with AngII (100 nm). Total or fractionated proteins were obtained with a Subcellular Protein Fractionation Kit (#78840; Thermo Fisher Scientific, Waltham, MA, United States). Protein content was determined with Bradford protein assay reagent, and lysates (25 μg) were separated by 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE), transferred to polyvinylidene difluoride (PVDF) membrane (Amersham, GE Healthcare, Buckinghamshire, United Kingdom), and probed with antibodies against NOX2 (1:500; #611414; BD Biosciences, San Jose, CA, United States) and NOX4 (1:500; Santa Cruz Biotechnology, Dallas, TX, United States). Detection was accomplished using horseradish peroxidase-coupled anti-rabbit or anti-mouse (1:2,000; The Jackson Laboratory, Bar Harbor, ME, United States) IgG antibody for 1 h at room temperature. Signal was detected using BioLumina (Kalium Tech, Buenos Aires, Argentina) detection system. Immunoblot signals were quantified by NIH ImageJ using β-actin expression as loading control.

The experiments were carried out according to institutional guidelines for animal experiments, approved by the Institutional Committee for the Care and Use of Laboratory Animals (CICUAL) of Facultad de Ciencias Médicas, Universidad Nacional de Cuyo (Aval approval # 77/2016). The study design that was followed is shown in Scheme 1A. C57BL/6J ApoE-KO mice aged 2 months (The Jackson Laboratory, Bar Harbor, ME, United States) were grouped-housed, had unrestricted access to water and standard chow (GEPSA, Cordoba, Argentina), and maintained on a 12-h light/dark cycle. Mice were fed a normal diet and treated 8 weeks with AEPs or AETa (150 or 300 mg/kg/day, respectively) in the drinking water. During this time, body weight, diet, and water intake control were monitored. Blood samples were collected after 4-h fasting period by cardiac puncture under anesthesia (ketamine 80 mg/kg–midazolam 5 mg/kg). The glucose concentrations were measured immediately after blood collection with an enzymatic kit (1400101; Wiener Lab^®, Rosario, Argentina). Plasma total cholesterol (TC) and triglycerides (TGs) were determined using colorimetric reactions with commercial kits (750220 and 791020, respectively; GT Lab, Buenos Aires, Argentina). As an indicator of lipid peroxidation, plasma malondialdehyde (MDA) was determined using the thiobarbituric acid reactive substances (TBARS) assay with own laboratory protocol. Total antioxidant status (TAS) in plasma was performed using a commercial kit (Randox Laboratories Ltd., Crumlin, United Kingdom) according to the manufacturer’s instructions. Trolox [(±)-6-hydroxy-2,5,7,8-tetramethylchromane-2-carboxylic acid] was used for calibration curve. Results are expressed as mmol equivalents of Trolox/L.

For the antioxidant enzyme determination, liver was homogenized in a proportion of 200 mg/ml of homogenization buffer (phosphate buffer plus cOmplete Protease Inhibitor 1,000×, Roche Diagnostics, Basel, Switzerland) and centrifuged at 10,000 × g for 15 min at 4°C to eliminate cellular organelle debris. Supernatant was used for the determination of the antioxidant enzyme glutathione peroxidase (GPx) activity with a Randox Laboratories commercial kit. The protein concentration was determined using the Bio-Rad Protein Assay (Bio-Rad Laboratories, Hercules, CA, United States).

Tissue superoxide (O₂^–) production was calculated from the initial linear rate of superoxide dismutase (SOD)-inhibitable cytochrome c reduction assay quantified at 550 nm using the extinction coefficient of 21.1 mM^–1 cm^–1 as previously described (Molshanski-Mor et al., 2007). Superoxide production is expressed as nM O₂^–min/mg protein using the extinction coefficient. More details in Supplementary Material. As any electron donors can also reduce cytochrome c, the assay was performed in the presence of SOD, which removes superoxide. Only the SOD-inhibitable signal truly reflects superoxide production. Data are presented as nmoles O₂^–min/mg protein generated in the presence of SOD or DPI, a general flavoprotein inhibitor.

Sections (5 μM) of 4% paraformaldehyde-fixed, paraffin-embedded aortic tissues were subjected to immunofluorescence staining. The sections were incubated with PBS plus 0.3% Triton X-100 (TPBS) for 5 min at room temperature, washed three times with PBS, blocked with 1% calf serum/TPBS for 1 h at room temperature, and incubated primary antibodies against 8-hydroxy-2′-deoxyguanosine (8-OHdG) conjugated to fluorescein isothiocyanate (FITC) (ab183393, 1:500; Abcam, Cambridge, United Kingdom) overnight at 4°C. The stained specimens were mounted and observed under fluorescence microscope (Olympus, Tokyo, Japan). Image analysis was performed by ImageJ 1.50i software.

The study design that was followed is shown in Scheme 1B. C57BL/6J ApoE-KO mice, 4 months old, were randomly divided into four groups (n = 6 each group), as follows: control diet (Control), HFD (standard chow supplemented with 30% bovine fat), HFD plus AEPs (150 mg/kg/day), and HFD plus AETa (300 mg/kg/day). HFD was administered during 8 weeks, and both extracts were administered in the drinking water during the last 4 weeks. All mice were sacrificed, and liver and whole aortas were excised. Each artery was fixed with 4% paraformaldehyde for 2 h and then placed in a 30% sucrose solution for 24 h. The anthropometric measurement of atherosclerotic lesions was determined according to Quesada et al. (2018).

Statistical analysis was performed using the GraphPad Prism 6.0 software. A p-value < 0.05 was considered statistically significant. Parametric or non-parametric tests were used according to variance homogeneity evaluated by F or Barlett’s tests. One-way ANOVA followed by Tukey’s multiple-comparison post-test was performed. Data represent the mean ± standard error of the mean (SEM) of three independent experiments for VSMC assays and three independent animal protocols.

Our first goal was to establish whether AEPs or AETa induced any harmful effect on VSMCs in culture. We tested a wide range of different concentrations of extracts (5–40 μg/ml) dissolved in 0.1% FCS. None of the concentrations tested affected cell viability when VSMCs were grown in medium containing a low concentration of FCS (0.1%, Figure 1A). Then, we stimulated VSMCs with 10% FCS during 48 h to promote cell proliferation, and we found that concentrations of 5–40 μg/ml of AEPs or AETa were able to affect VSMC proliferation (Figure 1B) assessed by MTT assay. In order to elucidate the molecular mechanisms of the antimitogenic effect of the extracts, we performed a cell cycle progression analysis by flow cytometry. VSMCs treated with 10% FCS showed an increase in the S phase population and a decrease in the G1 phase population as compared with 0.1% FCS control group (Figure 2A). A low concentration of either extract had no effect on 10% FCS-induced VSMC growth. After treatment with AEPs or AETa (20 μg/ml), the percentage of S phase cells was significantly reduced (15% for AEPs and 0.1% for AETa vs. 31.3% for 10% FCS), and the G2 phase percentage was increased (22.1% for AEPs and 22.4% for AETa vs. 8.3% for 10% FCS). These results indicate that AEPs or AETa (20 μg/ml) arrests VSMCs in the G2/M phase of the cell cycle. We also evaluated if the extracts were capable of inducing apoptosis, and we compared the percentages of cell on sub-G0 phase by flow cytometry. We found that AEPs produce a similar quantity of apoptotic cell as 10% FCS and less apoptosis that 0.1% FCS (Figure 2B). Remarkably, VSMCs growing in 10% FCS medium and treated 24 h with AETa (5 or 20 μg/ml) exhibited a greater number of cells in the sub-G0 phase, which would indicate that this extract could act as an inducer of apoptosis.

To investigate the antioxidant action of AEPs and AETa, we measured ROS generation in AngII-stimulated VSMCs pretreated with the extracts using FluoroProbe CM-H₂DCFDA (Figure 3A) or DHE (Figure 3B). The results showed that the addition of 2.5 up to 20 μg/ml of either AEPs or AETa significantly decreased intracellular ROS level induced by AngII. As seen in Figure 3B, DHE-derived fluorescent signals in VSMCs were increased by AngII compared with control (p < 0.0001). AEPs (5 and 20 μg/ml) and AETa 20 μg/ml significantly inhibited AngII-induced superoxide production (p < 0.001) in VSMCs, similar to apocynin and DPI effects. Our results suggest a strong antioxidant capacity for both aqueous extracts.

It is well established that the major and more specific enzymatic source of ROS in the vascular wall is NOX; and the isoforms NOX1, NOX2, and NOX4 are predominantly expressed in VSMCs (Garcia-Redondo et al., 2016). For this reason, we next evaluated the expression of NOX isoforms in VSMCs exposed to AngII, pretreated with AEPs or AETa. As is shown in Figure 4 (top panel), AngII-induced NOX4 expression is significantly downregulated by AEPs (both mRNA and protein expression), and AETa promoted decrease in NOX4 mRNA, which did not translate to protein expression. In addition, the expression of NOX2 enhanced by AngII was downregulated by AEPS (protein level) and significantly diminished by both AETa mRNA and protein levels (Figure 4, bottom panel).

Using an experimental hypercholesterolemia animal model such as ApoE-KO mice, the effects of chronic intake of AEPs or AETa were investigated. The administration of AETa (300 mg/kg/day) during 8 weeks significantly reduced blood glucose levels as compared with the control chow group (190 ± 23 mg/dl (control group) vs. 132 ± 7 mg/dl (AETa group); p = 0.022 Figure 5A). No change was found with AEPs intake, in plasmatic glucose level. On the other hand, both extracts significantly reduced plasma triglyceride levels as compared with control chow mice [119 ± 7 mg/dl (control) vs. 47 ± 13^∗ mg/dl (AEPs) and 50 ± 10^∗ mg/dl AETa; ^∗p < 0.0001, Figure 5B], while TC levels were not affected (Figure 5C).

To address the capacity of the extracts on preventing systemic oxidative stress, we evaluated the level of lipid peroxidation and the TAS in plasma. We found a significant decrease in plasmatic TBARS in animals treated with AEPs or AETa [3.13 ± 0.13 mmol/L (control) vs. 2.21 ± 0.29 mmol/L (AEPs) and 2.26 ± 0.11 mmol/L (AETa); ^∗∗p < 0.01] (Figure 5D). Simultaneously with reduced MDA levels, significantly higher levels of TAS were also found [22.3 ± 0.9 (control) vs. 31.7 ± 2.6 (AEPs) p < 0.01 and 30.1 ± 1.8 (AETa) p < 0.05 mmol of Trolox equiv./L] (Figure 5E). Interestingly, GPx activity, a measure to evaluate the level of antioxidant capacity in the liver, showed a significant increase in animals treated with the extracts [10.5 ± 0.9 U/mg (control) vs. 14.0 ± 0.9^∗ U/mg (AEPs) and 13.5 ± 0.8^∗ U/mg AETa; ^∗p = 0.022; Figure 5F].

An additional group of ApoE-KO mice were fed a HFD in order to exacerbate the pro-atherogenic state. Eight weeks of feeding HFD increased TC and triglycerides in ApoE-KO mice, compared with animals in control chow, and 4 weeks’ treatment with either extract was not enough to avoid the rise of these parameters (Table 1). However, 4 weeks’ treatment was enough to significantly reduce NOX-derived superoxide anion generation induced by HFD in the liver of ApoE-KO mice and almost completely removed superoxide anion generation in aortas (Figure 6A), demonstrating that both extracts have a potential ROS scavenging activity. As any electron donors can also reduce cytochrome c, the assay was performed in the presence of SOD, which removes superoxide. Only the SOD-inhibitable signal truly reflects superoxide production. Figure 6B shows nmoles O₂^–min/mg of protein generated by aorta tissues in the presence of SOD or DPI. As a marker of DNA oxidative damage induced by ROS, we detected immunofluorescence expression for 8-OHdG. Aortas from HFD-fed mice showed a stronger 8-OH-dG stain compared with aortas from mice fed control chow, and in the aortic wall of mice treated with HFD plus AEPs or AETa, 8-OH-dG stain was markedly weaker (Figure 7A). Next, we evaluated the atheroma development in our model. HFD showed an increase in atherosclerotic plaque size in aortas compared with control chow animals [6.37 ± 1.06% total area (control) vs. 20.33% ± 1.94% total area (HFD); ^∗∗p < 0.001 Figure 7B], which was remarkably decreased in mice fed HFD plus AEPs (10.2 ± 1.33%^∗∗ total area; ^∗∗p < 0.01) or AETa (13.72 ± 1.02%^∗ total area, ^∗p < 0.05).