C. adamantium O. Berg leaves and roots were collected following the identification of the plant and authorization of the SISBIO (Sistema de Autorização e Informação em Biodiversidade, permit number 54470).

Leaves and roots of C. adamantium O. Berg were collected in Dourados, in the state of Mato Grosso do Sul, Brazil (coordinates: 22° 02′ 47.9 S′ and 055° 08′ 14.3′ W). The samples were cleaned, dried in a convection oven at 45°C, and ground using a Croton knife mill. An exsiccated sample was deposited in the Herbarium of the Federal University of Grande Dourados, Mato Grosso do Sul, Brazil, with registration number 4108.

Extracts were prepared from the powdered material using an accelerated solvent extractor (ASE®-150, Dionex), as described by Espindola et al. (2016). The samples were placed in a cell of 100 mL and extracted with distilled water at temperature of 125°C through two 5 min static cycles, with an 80% flush volume and a 60 s purge. The extracts were lyophilized to obtain the dry extract. Thus, the aqueous extracts of Campomanesia leaves (AECL) and roots (AECR) were obtained with a yield of 13 and 6%, respectively.

The analyses were performed at a Shimadzu Prominence UFLC coupled to a diode array detector (DAD) and a mass spectrometer (MicrOTOF-Q III, Bruker Daltonics) with an electrospray ion source. Kinetex C18 (2.6 μm, 150 × 2.1 mm, Phenomenex) chromatographic column was used. The AECL and AECR extracts were prepared at concentration 1 mg/mL and the volume of 1 μL was injected on the chromatography. The flow rate and oven temperature were 0.3 mL/min and 50°C, respectively. Formic acid 0.1% (v/v) in both water (solvent A) and acetonitrile (solvent B) were used as mobile phase, applying the following gradient elution profile: 0–8 min at 3% B, 8–30 min at 3–25% B, 30–60 min at 25–80% B, and 60–63 min at 80% B. The MS analyses were carried out both ion modes (positive and negative), applying 2.5 kV of capillary voltage. Nitrogen was used as nebulizer gas (4 Bar), dry (9 L/min) and collisional gas.

Human peripheral blood mononuclear cells from healthy donors were collected after informed patient consent. Separation of mononuclear cells was performed by gradient centrifugation methods using Ficoll Histopaque-1077 (1.077 g/cm³) (Sigma–Aldrich, Germany) follow the manufacturer's instructions at 400 g for 30 min. The use of human samples was approved by the local Ethical Committee of the University Center of the Grande Dourados under protocol number 123/12.

The Jurkat (human acute T lymphocytic leukemia) cell line was cultured in suspension in RPMI 1640 medium (Cultilab, Brazil) supplemented with 10% fetal calf serum (FBS, Cultilab, Brazil), 100 U/mL penicillin, and 100 μg/mL streptomycin in a humidified atmosphere at 37°C under 5% CO₂.

Peripheral blood mononuclear cells and Jurkat cells were seeded (10⁵ cells/mL) in 96-well microplates containing medium with vinblastine (positive control), EACL or EACR at different concentrations. After incubation for 24 h, the treatments were removed and 100 μL of 10% Alamar Blue solution was added. After 5 h, the fluorescence was read at 530 nm (Ex) and 590 nm (Em) in a microplate reader FlexStation 3 (Molecular Devices, USA). Each experiment was performed in triplicate.

The cell death profile was evaluated according to the methods described by Paredes-Gamero et al. (2012), with minor modifications. Jurkat cells were seeded in 96-well plates (10⁵ cells/mL) and cultured in medium containing 10% FBS for 24 h with IC₅₀ of AECL (40 μg/mL) or AECR (80 μg/mL) and the double of these concentrations. After this period, the cells were washed with PBS and resuspended in annexin labeling buffer (0.01 M Hepes, pH 7.4, 0.14 M NaCl, and 2.5 mM CaCl₂). The suspensions were labeled with annexin-FITC and propidium iodide (PI) (Becton Dickinson, USA) according to the manufacturer's instructions, and the cells were incubated at room temperature for 15 min. Finally, 10,000 events were collected per sample through analysis in a Accuri C6 flow cytometer (Becton Dickinson, USA) using FlowJo v10.2 LCC software (Oregon, USA).

Mitochondrial depolarization was evaluated through the incorporation of JC-1 (5,5′,6,6′–tetrachloro—1,1′,3,3′—tetraethylbenzimidazolylcarbocyanine iodide; Molecular Probes, Eugene, OR, USA) according to the method described by Moraes et al. (2013). JC-1 is cationic dye that shows potential-dependent accumulation in mitochondria via fluorescence emission. Cells that stained red were classified as exhibiting a high mitochondrial membrane potential, while those that stained green were classified as exhibiting a low membrane potential. For these assays, cells were seeded in 24-well plates (10⁵ cells/mL), where they were cultured in medium containing 10% FBS and treated with IC₅₀ of the AECL or AECR, for 24 h. After treatment, the cells were centrifuged and incubated with JC-1 (1 μg/mL) at room temperature for 15 min. Fluorescence was analyzed in a FACScalibur Flow Cytometer (Becton Dickinson, USA) using CellQuest software (10,000 events were collected per sample).

Antibodies against actin, Bcl-2 and Bax were obtained from Santa Cruz Biotecnology (Santa Cruz, CA, USA), whereas the monoclonals anti-cleaved caspase-9 and anti-cleaved caspase-8 antibodies were from Cell Signaling Technologies (Beverly, MA, USA). Horseradish peroxidase-conjugated secondary antibodies against rabbit (goat anti-rabbit), mouse (goat anti-mouse), and goat (donkey anti-goat) were from Millipore (Temecula, CA, USA).

After treatment with IC₅₀ of the AECL or AECR, cells were harvested in a lysis buffer (50 mM Tris-HCl, pH 7.4), 1% Tween-20, 0.25% sodium deoxycholate, 150 mM NaCl, 1mM EGTA, 1mM o-Vanadate, 1mM NaF, and 1X protease inhibitor cocktail (Thermo Fisher Scientific, Rockford, IL, USA) on ice for 2 h, and after clarification, protein concentration was determined using the Pierce BCA Protein Assay kit (Thermo Fisher Scientific). Equal volume of 2X SDS gel loading buffer was added and samples were boiled for 5 min. Cellular extracts were subjected to SDS-PAGE (10%) and transferred to nitrocellulose membranes. Blots were probed overnight at 4°C with appropriate primary antibodies at 1:1,000 dilutions. After washing in TBS/0.05% Tween-20, membranes were incubated with anti-goat, anti-rabbit, or anti-mouse horseradish peroxidase-conjugated secondary antibody at 1:3,000 dilutions for 1 h. Blots were imaged using Pierce SuperSignal West Pico Chemiluminescent Substrate (Thermo Fisher Scientific) on a LAS-3000 imaging system.

The activation of caspase-3 was evaluated via flow cytometry according to the method described by Moraes et al. (2013), with minor modifications. Cells were seeded in 24-well plates (10⁵ cells/mL), where they were cultured in medium containing 10% FBS and treated with IC₅₀ of the AECL or AECR, for 24 h. After treatment, the cells were centrifuged, washed, and fixed in 2% paraformaldehyde in PBS for 30 min. The cells were then washed with glicin (0.1 M) in PBS, permeabilized with 0.01% saponin for 30 min, and blocked in PBS containing 1% BSA for 30 min at room temperature. Subsequently, the cells were incubated with an anti-active-caspase-3 monoclonal antibody conjugated with FITC (BD-Pharmingen, San Diego, CA, USA) in the dark at room temperature for 40 min. After incubation, fluorescence was analyzed in a FACScalibur Flow Cytometer (Becton Dickinson, USA) using CellQuest software (10,000 events were collected per sample).

The importance of caspases and calcium for AECL or AECR cytotoxicity was evaluated according to the method described by Bechara et al. (2014), with minor modifications. Briefly, Jurkat cells that had been pretreated for 1 h with 10 μM of the Z-VAD-FMK (general caspases inhibitor) or 10 μM of the BAPTA-AM (a calcium chelator) before AECL or AECR incubation for 24 h, both with IC₅₀ values, were subjected to trypan blue exclusion assays. To assess cell viability, control and treated cells were resuspended in equal volumes of medium and trypan blue (0.05% solution) and were counted using a hemocytometer chamber.

Cytosolic Ca²⁺ concentration [Ca²⁺]_cyt was determined by incubating 10⁵ Jurkat cells/well in black 96-well microplates with the Fluo-4 Direct Calcium Assay reagent according to the manufacturer's instructions (Life, USA). After 1 h of indicator incorporation at 37°C, cells were stimulated with IC₅₀ of the AECL or AECR and the fluorescence was quantified in a FlexStation 3 microplate reader (Molecular Devices, USA). Ionomycin 0.1 μM, was used as positive control to obtain the maximal fluorescence. The Fluo-4 was excited at 490 nm, and the emission was detected at 525 nm (Leon et al., 2011). The records of fluorescence intensity as a function of time correspond to the mean of experiment performed in quadruplicate.

To determine the cell cycle distribution, Jurkat cells were seeded (10⁵ cells/mL) and cultured in medium containing 10% FBS and were treated with IC₅₀ of the AECL or AECR, for 24 h. After this period, the cells were washed, fixed and permeabilized as described previously. Then, the cells were treated with 4 μg/mL RNase type I for 1 h at 37°C and resuspended in PBS. The cells were subsequently stained with 5 μg/mL PI and analyzed (12,000 events collected per sample) in an Accuri C6 flow cytometer (Becton Dickinson, USA). The DNA content was evaluated using an FL2A detector on a linear scale. The dead cells were excluded to cell cycle analysis. The analyses of the cell percentage in the G1/S/G2 phases were performed using FlowJo v10.2 LCC software.

To investigate the mechanisms of extracts-induced cell cycle arrest, Jurkat cells were seeded in 24-well plates (10⁵ cells/mL), where they were cultured in medium containing 10% FBS and treated with IC₅₀ of the AECL or AECR, for 24 h. After this period, were fixed with 2% paraformaldehyde for 10 min, washed with BD Perm/Wash buffer and permeabilized with BD Perm Buffer III for 30 min. Then, the cells were labeled with 5 μL Ki-67-FITC antibody (Becton Dickinson, USA). To label phosphorylated retinoblastoma protein (p-Rb) anti-p-Rb (Ser780) (Cell Signaling, USA) antibody was used. After permeabilization intracellular protein was incubated for 1 h. Then, anti-rabbit IgG secondary antibody conjugated with Alexa Fluor 488 (Life, USA) were used for 40 min. Protein analyses were performed by quantification of the fluorescence geometric mean (Gm). The cells were analyzed in an Accuri C6 flow cytometer (Becton Dickinson, USA) using FlowJo v10.2 LCC software (Torquato et al., 2017).

The data are shown as the mean ± standard error of the mean (SEM) and were analyzed for statistically significant differences between groups. Student's t-test was employed for comparisons between two groups, and one-way analysis of variance (ANOVA) followed by Dunnett's test was employed for comparisons of more than two groups using Prism 5 GraphPad Software. The results were considered significant when P < 0.05.

The constituents from AECR and AECL were identified based on the UV, high resolution MS and MS/MS data (Figure 1, Table 1) and the comparison to the published data (Fabre et al., 2001; March et al., 2006; Ferreres et al., 2013; Guaratini et al., 2014; Zhang et al., 2016). The main compounds identified from AECR were di-hexoside/quinic acid (1), ellagic acid O-pentoside (27), ellagic acid (30), O-methyl ellagic acid O-hexoside (31), ellagic acid O-deoxyhexoside (33), and O-methyl ellagic acid sulfate (39) (Table 1). In addition, other constituents were identified such as gallic acid (3), ellagic acid O-hexoside (20), O-methyl ellagic acid O-deoxyhexoside (43), and O-dimethyl ellagic acid sulfate (44) (Figure 2).

While from AECL, forty-two compounds have been identified. The main constituents identified were di-hexoside/quinic acid (1), myricetin O-pentoside (23, 26, 29), myricetin O-deoxyhexoside (32), quercetin O-pentoside (38), and myricetin O-(O-galloyl)-pentoside (40).

The peaks 5–9, 11–12, 14, 18–19, 21–22, and 28 (Table 1) presented a similar UV (λ_max ≈ 280 nm), which are compatible to flavan-3-ols (Markham, 1982). The typical fragmentation for condensed tannins was observed for these chromatographic peaks, including the heterocyclic ring fission, reto-Diels-Alder (RDA) and quinone methide reactions (Gu et al., 2003; Guaratini et al., 2014; Nocchi et al., 2017). The condensed tannins esterified with gallic acid were confirmed by fragment ion at m/z 169 [M-H-flavan-3-ol units]⁻, such as for 14, 18, 22, and 28. The monomers were confirmed by co-injection of authentic standard as gallocatechin (5), epigallocatechin (6), catechin (7), and epicatechin (12)(Figure 2).

The compounds 19, 23–26, 29, 32, 34–38, 40–42, and 45 exhibited UV spectra relative to flavonols (Markham, 1982). The losses of 162, 146, 132, and 152 u confirmed the presence of hexoside, deoxyhexoside, pentoside, and gallic acid, respectively. The identification of aglycones was based on product ions yielded from C-ring fissions, as widely described in the literature (Fabre et al., 2001; March et al., 2006), and they corroborated to the aglycones quercetin, kaempferol, and myricetin. In addition, the compounds 47 and 48 showed intense ions at m/z 269 [M-H]⁻ and 271 [M+H]⁺, compatible to molecular formula C₁₆H₁₄O₄ and UV spectra characteristic of flavanone. The determination of the methyl position in A-ring was established from elution time described for 5,7-dihydroxy-6-methylflavanone and 5,7-dihydroxy-8-methylflavanone, which were isolated from C. adamantium (Coutinho et al., 2010).

The compounds 20, 27, 30, 31, 33, 39, 43, and 44 showed UV spectra compatible to ellagic acid (Ferreres et al., 2013). From m/z 303.0128 [M+H]⁺ (C₁₄H₇O₈⁺, error 4.5 ppm) of compound 30 were yielded the product ions at m/z 275, 257, and 247, which are relative to subsequent eliminations of CO (28 u) and water molecules and they were compatible to ellagic acid (confirmed by co-injection of the standard) (Ferreres et al., 2013; Zhang et al., 2016). The losses of 15 and 80 u confirmed the presence of methoxyl (31, 39, 44) and sulfate substituents (39, 44) and they were identified as O-methyl ellagic acid sulfate and O-dimethyl ellagic acid sulfate, respectively. Similar sulfate ellagic acid derivatives have been described from Lagerstroemia speciosa (Lythraceae) and Euphorbia sororia (Euphorbiaceae) (Bai et al., 2008; Zhang et al., 2008). The derivatives of ellagic acid were only detected from C. adamantium roots.

Peripheral blood mononuclear cells and Jurkat cells were treated with vinblastine, AECL or AECR to assess cell viability. Vinblastine promoted cell death in Peripheral blood mononuclear cells (IC₅₀ = 25.9 μg/mL) and Jurkat cells (IC₅₀ = 9.5 μg/mL) (Figure 3A). The AECL did not change the viability of Peripheral blood mononuclear cells at the evaluated concentrations, however, promoted death in Jurkat cell (IC₅₀ = 40 μg/mL) (Figure 3B). AECR showed low cytotoxicity against Peripheral blood mononuclear cells only at the highest concentration evaluated (160 μg/mL), but demonstrated concentration-dependent cytotoxicity against Jurkat cells (IC₅₀ = 80 μg/mL) (Figure 3C).

Both extracts resulted in double staining (annexin V⁺/PI⁺), indicating late apoptosis. This type of death was observed in 76.1 ± 3.2% of the cells treated with 80 μg/mL AECL (Figure 4A) and in 65.1 ± 1.4% of the cells treated with 160 μg/mL AECR (Figure 4B).

ΔΨmit was decreased in cells treated with AECL and AECR compared with untreated cells, as demonstrated by decreased red fluorescence and increased green fluorescence after a 24-h incubation. The rates of mitochondrial depolarization in cells treated with AECL and AECR were 85.4 ± 8.2 and 84.5 ± 2.0%, respectively (Figures 5A,B).

In order to evaluate the role of the extrinsic and intrinsic pathways of apoptosis in cell death, protein expression of cleaved-caspases-8 and -9, Bcl-2, and Bax were investigated by western blot. As demonstrated in Figure 6A, the extrinsic pathway did not influence the mechanism of action of these extracts as evidenced by the absence of the characteristic fragments of cleaved caspase-8 (Asp391). In contrast, there was an increase in the cleaved form of caspase-9 (Asp330) in cellular extracts treated with both AECL and AECR, suggesting a role for the intrinsic pathway in mediating apoptosis. Moreover, this was followed by a slight increase in the expression of Bax, whereas no alteration was seen in Bcl-2 expression (Figure 6B).

A monoclonal anti-cleaved caspase-3 antibody was used to evaluate caspase-3 activation in cells incubated with AECL and AECR, and the cells were analyzed via flow cytometry. Both extracts resulted in the cleavage of pro-caspase 3, as indicated by a shift in fluorescence to the right (Figures 7A,B). Caspase-3 activation was 5.0-fold higher in cells treated with AECL (Figure 7A) and 2.5-fold higher in cells treated with AECR compared with the untreated control (Figure 7B).

Jurkat cells were pre-incubated with Z-VAD-FMK, a general caspase inhibitor, to assess whether caspases were involved in the cytotoxicity induced by the tested extracts. We observed that AECL-induced cytotoxicity decreased after treatment with the inhibitor (Figure 8A), indicating that caspases are actively involved in the cytotoxicity of this extract. However, pre-incubation with Z-VAD-FMK resulted in no difference in the viability of cells treated with AECR (Figure 8B).

Additionally, Jurkat cells were pre-incubated with BAPTA-AM, an intracellular Ca²⁺ chelator, to investigate the role of Ca²⁺ in the cytotoxicity induced by the extracts. Our results showed that AECL- and AECR-induced cytotoxicity was decreased after treatment with BAPTA-AM (Figures 8A,B), demonstrating the importance of calcium in the cytotoxic activity of the extracts in this cell line.

To corroborate the involvement of cytoplasmatic Ca²⁺ in cell death AECL and AECR- induced we performed a temporal evaluation of Ca²⁺ concentration. Both compounds were able to promoted increase of Ca²⁺ concentration in Jurkat cells (Figures 9A,B).

The effect of the extracts on the cell cycle distribution is shown in Figure 10A. The numbers of Jurkat cells in G1 phase were decreased by 16.6 ± 1.3 and 8.0 ± 0.5% after 24 h of treatment with AECL and AECR, respectively, compared with untreated cells. The number of cells in S phase after treatment with AECL and AECR increased to 23.86 and 25.23%, respectively, compared with untreated cells 17.66% (Figures 10A,B).

The arrest of cell cycle was confirmed by the quantification of Ki-67 protein, a protein only expressed during the cell division cycle, and by quantification of p-Rb, a protein that prevent the progression of cell cycle when is not phosphorylated. Ki-67 was not altered by AECL, whereas AECR promoted a reduction of 27.3 ± 0.1% (Figure 11A). Furthermore, the cell cycle arrest was also confirmed by the reduction of p-RB levels (AECL 27.9 ± 3.7%; AECR 58.5 ± 2.5%) (Figure 11B).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.