Bovine serum albumin, 5,5′-dithiobis (2-nitrobenzoic acid) (DTNB), reduced glutathione (GSH), xylenol orange, K2HPO4, KH2PO4, 1 M Tris, ethylenediaminetetraacetic acid, Tris HCl (all from Sigma, St. Louis, MO, United States), pyrogallol, absolute ethanol, absolute methanol, ferrous ammonium sulfate, trichloroacetic acid, formaldehyde (all from Vetec, Rio de Janeiro, Brazil), and ultra-pure water from a Milli-Q system were used for eluent preparation.

There are no local traditional medicinal plant books with species that are popularly used in the region of Grande Dourados, Mato Grosso do Sul, Brazil. Therefore, Croton urucurana Baill. was selected based on a broad ethnobotanical study that was conducted by our research group to identify botanical species that are used by healers in the region. Data were collected through semi-structured interviews and participant observations. We were able to obtain information on the preparation, use, and relative importance of this medicinal species. C. urucurana is popularly used as an infusion of its leaves or bark to treat cardiovascular and gastrointestinal disorders (Coelho et al., 2019).

Leaves of C. urucurana were collected in May 2020 in Dourados, Mato Grosso do Sul, Brazil (“22°20.9299’south, 54°83.7713 west). A voucher specimen (no. 5536) was deposited at the Herbarium of the Federal University of Grande Dourados. The plant was dried in an oven and sprayed. The extract was prepared by infusion using the methodology of Barbosa et al. (2020). The pulverized material (100 g) was subjected to the extraction process by infusion with 1 L of boiling water. The resulting infusion was kept in an amber flask for 5 h. After filtration, the infusion was treated with 95% ethanol in a proportion of 1:3 (v/v) to precipitate proteins and polysaccharides, giving rise to a heterogeneous phase that was removed by filtration. The ethanol-soluble fraction was concentrated on a rotary evaporator and then lyophilized. The plant name was checked at http://www.theplantlist.org and found to be approved.

The phytochemical composition of C. urucurana was evaluated by liquid chromatography-mass spectrometry (LC-MS) using a high-performance liquid chromatograph (Prominence LC 20A, Shimadzu) coupled to a Maxis 3G Q-Tof high-resolution mass spectrometer (Bruker Daltonics). Chromatography was performed on a C18 column (250 nm × 4.6 nm, 5 μM particle size; Phenomenex, Torrance, CA, United States), held at 40°C. The solvent was composed of ultra-pure water and acetonitrile (Lichrosolv-Merck) that contained 0.1% formic acid (for positive ion mode) or 0.1% ammonium formate (for negative ion mode). A linear gradient was developed at a flow rate of 1 ml/min, increasing acetonitrile from 5 to 80% in 25 min, then to 100% in 29 min, and then returning it to 5% in 30 min, followed by 5 min to rebalance the system before the next injection. The sample (2 mg/ml) was prepared in methanol-water. The injection volume was 20 μL. A photodiode array (200–400 nm) or high-resolution MS (m/z 100–1,500) was used for detection.

The high-resolution MS analyses were performed in positive ion mode and negative ion modes, with energies set at 500 V in the end-plate offset and 4.5 kV in the capillary. Nitrogen was used for sample desolvation. The dry gas flow was 8 L/min. The nebulizer pressure was 2 Bar. The source temperature was 250°C. Data-dependent analysis was performed to obtain fragmentation spectra by collision-induced dissociation-MS using argon as the collision gas, with a voltage ramp of 10–60 eV.

The research model was developed in spontaneously hypertensive and Wistar Kyoto male rats, weighing 200–250 g, that were obtained from the central vivarium of the Federal University of Grande Dourados. The animals were housed in the vivarium of the Laboratory for Pre-Clinical Research of Natural Products at Paranaense University, with free access to food and water. They were housed under controlled environmental conditions (20°C ± 2°C temperature, 50 ± 10% relative humidity, 12 h/12 h light/dark cycle) with environmental enrichment. The total number of animals in the study was 48 (n = 8/group). The animals were weighed weekly on an analytical balance. The experimental protocol was approved by the Ethics Committee on the Use of Animals of Paranaense University (protocol no. 1000/2020). All national and international guidelines were followed. The reporting of animal investigations was performed and interpreted according to Animal Research Reporting of in vivo Experiments (ARRIVE) guidelines (Sert et al., 2020).

For 10 weeks, the animals received standard commercial food that was enriched with cholesterol ad libitum and were exposed to smoke from nine commercial cigarettes (0.8 mg nicotine, 10 mg tar, and 10 mg carbon monoxide) for 1 h daily, 5 days weekly, for 10 weeks, as proposed by Souza et al. (2020). For the induction of dyslipidemia, the animals received a commercial standard diet (Purina®) that was enriched with 0.5% cholesterol (150 g of standard feed for rodents, one egg yolk, and 13.5 ml of corn oil). All the ingredients were mixed with water, baked in an oven at 50°C for 36 h, and packed in vacuum bags as proposed by Souza et al. (2020). This preparation contains 225 mg cholesterol, 1.8 g saturated fat, 2.16 g monounsaturated fatty acids, and 0.72 g polyunsaturated fatty acids for every 150 g of standard feed for rodents. During the last 5 weeks of the study, the animals were treated orally, by gavage, with vehicle (0.1 ml of filtered water/100 g body weight; negative control [C-] group), the ethanol soluble fraction of Croton urucurana (30, 100, and 300 mg/kg), or simvastatin (SIM; 2.5 mg/kg) + enalapril (ENAL; 15 mg/kg), once daily. Normotensive, non-dyslipidemic, and non-smoke-exposed Wistar Kyoto rats (basal group) were treated with vehicle (filtered water; n = 8 The doses of the C. urucurana extract were defined according to its traditional use in Brazil (Coelho et al., 2019). The most-reported preparation is the use of 200 ml of pre-boiling water, that is, directly poured into an amount of crushed plant, that is, equivalent to a “closed hand” (∼2.5 g). The 30 mg/kg dose was calculated by dividing 2.5 g by the average weight of a human adult (70 kg). Thus, we used the 30 mg/kg dose, a 10-times higher dose (i.e., 300 mg/kg), and an intermediate dose on a logarithmic scale (100 mg/kg; Coelho et al., 2019). Electrocardiograms, heart rate, and blood pressure in these normotensive and hypertensive rats were described in a previous study by our group (Zago et al., 2021).

Blood samples were collected from the left carotid artery using heparinized syringes. Plasma was separated by centrifugation at 1,500 g × g for 10 min and stored at −80°C for the biochemical analyses. The rats were then euthanized by puncture of the diaphragm while under anesthesia, and the liver was removed. Samples were rapidly separated and frozen in liquid nitrogen to evaluate oxidative stress and perform biochemical analyses. Other organ samples were stored in 10% formalin solution for further histological analysis. Feces (representative of 2 days of feces accumulation) were collected directly from the animal cages on the last day of the experiment and stored at −20°C until processing.

Plasma levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), cholesterol, and triglycerides were measured by the colorimetric enzymatic method in an automated analyzer (Quick Lab, São Paulo, Brazil).

Lyophilized liver and fecal samples underwent lipid extraction using the gravimetric method as described by Lívero et al. (2016). Liver and fecal samples were mixed with hexane as the solvent (1:10; feces:solvent) and heated at 80°C. After 12 h, the supernatant was transferred to a second flask and naturally evaporated. This procedure was repeated three times. The lipid content was then weighed and suspended in 1 ml of chloroform plus 2 ml of isopropanol to determine hepatic and fecal levels of triglycerides and cholesterol by the colorimetric enzymatic method in an automated analyzer. The percentage of lipids in the liver was calculated as the following: (lipids [%] = 100 × [final flask weight/initial flask weight]/0.1 g).

To investigate the hepatic antioxidant system, liver samples were homogenized in a 1:10 dilution of potassium phosphate buffer (0.1 M, pH 6.5). Afterward, 100 µL was separated, suspended in 80 µL of trichloroacetic acid (12.5%), vortexed, and centrifuged at 6,000 rotations per minute (rpm; 15 min at 4°C) to analyze GSH levels (Sedlak and Lindsay, 1968). The remaining homogenate was centrifuged at 9,700 rpm for 20 min at 4°C to determine superoxide dismutase (SOD) activity and lipoperoxidation (LPO) levels according to Gao et al. (1998) and Jiang et al. (1992), respectively.

A sample of the liver were fixed in buffered 10% formalin solution (distilled water, 35–40% formaldehyde, and monobasic and dibasic sodium phosphate), dehydrated with alcohol and xylene, embedded in paraffin, sectioned at 6 μM, and stained with hematoxylin/eosin. The other liver sample underwent saturation in sucrose (10, 20, and 30% sucrose solutions for 24 h at each concentration), stored in Tissue Tek (O.C.T. Sakura), rapidly frozen in liquid nitrogen, and sectioned (6 µM) for Nile Blue staining (Lívero et al., 2016). The slides were analyzed by optical microscopy (Leica DM 2500) to evaluate cellular alterations. Hepatic lesions were classified as the following: 0 (0%; absence of lesions), 0.5 (1–5%; minor lesions), 1 (6–33%; moderate lesions), 2 (34–66%; marked lesions), and 3 (67–100%; massive lesions) according to Souza et al. (2020).

The data were analyzed for homogeneity of variance and a normal distribution. Differences between means were determined by one-way analysis of variance (ANOVA) followed by the Newman-Keuls post hoc test or by the Kruskal-Wallis test followed by Newman–Keuls’s post hoc test. The level of significance was set at 95% (p < 0.05). The data are expressed as the mean ± standard error of the mean (SEM).

The main constituents of the C. urucurana extract were identified as flavonoids and glycosides, along with alkaloids. High-resolution MS analysis was performed in positive and negative ion modes, but the positive polarity showed the best results—alkaloids were barely detected in negative ion mode. Thus, except as indicated, the results are presented in positive ion analysis, with neutral components from C. urucurana obtained as protonated ions [M⁺H]⁺.

Alkaloids are metabolic products, usually from amino acids. Therefore, they have a characteristically even m/z value because of the presence of a nitrogen element. Different peaks with even m/z values were observed in the C. urucurana extract. Although abundant on the chromatogram, peak one at m/z 266.124 and peak 2 at m/z 146.082 did not produce good fragments to allow their identification. Nevertheless, peak one had an m/z that was similar to the alkaloid anonaine (C₁₇H₁₅NO₂), and peak 2 was similar to N-methyl-trans-4-hydroxy-L-proline that was extracted from Myracrodruon urundeuva (Aquino et al., 2019). Other low-abundance peaks (peaks 3–8) were observed with even m/z values, suggesting other alkaloid compounds (Figure 1A; Table 1). The most abundant peaks were also recognized as alkaloids, appearing at m/z 344.187 (peak 9), 342.172 (peak 11), 342.170 (peak 13), and 370.129 (peak 14). Cordeiro et al. (2016) described structures of some alkaloids that were obtained from the bark of C. urucurana, with the same m/z values as those observed herein. Peak 9 (m/z 344.187) gave fragments at m/z 299.129, 175.075 and, 143.049 and 137.059 (Figure 1B), with the fragment at m/z 299.129, with the neutral loss (NL) of 45.05 atomic mass units (a.m.u.) being consistent with the loss of a dimethylamine group. These fragments were the same as those found by Yan et al. (2013), identified as the aporphinic alkaloid tembetarine, which was consistent with our findings. Peak 11 at m/z 342.172 gave fragments at m/z 297.113 (NL of 45.05 a.m.u. from the loss of a dimethylamine group), 282.090, 265.086, and 237.090 (Figure 1C). Yan et al. (2013) described the structure that was found at the same m/z value, with similar fragments as those found herein, indicating that compound 11 is the alkaloid magnoflorine. A possible isomer also appeared at m/z 342.170 (peak 13). The fragmentation profiles of these compound were different from peak 11, with main fragments at m/z 311.129, 279.102, and 251.107 (Figure 1D). However, the NL of 31.04 a.m.u. that yielded the fragment at m/z 311.129 was consistent with the loss of a methylamine rather than dimethylamine as from compound 11. Singh et al. (2017) described the structure of an alkaloid, the aporphinic alkaloid isocorydine, in Mahonia leschenaultia with similar fragments. Peak 14 at m/z 370.129 was consistent with the structure of the alkaloid taspine, which was also described by Cordeiro et al. (2016).

Despite other studies that indicated the presence of various condensed tannins in C. urucurana, we observed a single peak at m/z 579.149 (peak 10) in the present extract, with a main fragment at m/z 289.070, which was characteristic of (epi)catechin-(epi)catechin, as previously described (Souza et al., 2008a; Casao et al., 2020). Peak 12 at m/z 291.0862 was identified as catechin, confirmed by comparison with an authentic standard (Zago et al., 2021).

Flavonol O-glycosides were also identified in the C. urucurana extract (peak 15 at m/z 611.163 and fragments at m/z 465.104 and 303.052). This compound was confirmed as rutin by comparison with an authentic standard. Flavone C-glycoside was also observed in the C. urucurana extract. This compound was observed at m/z 433.113 (peak 16), exhibiting the characteristic NL values of 90 and 120 a.m.u. that were observed mainly in negative ion mode, which were characteristic of vitexin and isovitexin (Prando et a., 2016). Peak 17 at m/z 465.105 was confirmed as isoquercitrin (quercetin-3-O-glucoside) by comparison with an authentic standard, as described by Lopes Alves et al., 2020, in C. urucurana. Peak 18 at m/z 595.165 with main fragments at m/z 449.107 and 287.055 was characteristic of kaempeferol diglycoside, such as kaempferol rutinoside or its isomers (Souza et al., 2008b). Compound 19 at m/z 449.107 with a main fragment at m/z 287.055 was identified as a kaempferol-O-hexoside, being glucosides and galactosides the most commonly found in plants (Souza et al., 2016).

Compound 21 at m/z 625.175 had main fragments at m/z 479.117 and 317.065. The sequential NL of 146.05 and 162.05 a.m.u. indicated the elimination of a deoxyhexosyl (e.g., rhamnose) residue, followed by the loss of an hexosyl (e.g., glucose or galactose) residue. The fragment at m/z 317.061 was consistent with methoxy-quercetin, in which we observed the loss of the radical CH₃• in the negative fragment ions (Tirloni et al., 2018). Thus, compound 21 was tentatively identified as a methoxy-quercetin rutinoside (or isomer), similar to isorhamnetin-O-deoxyhexosyl-hexoside that was described by Lopes Alves et al., 2020.

Compound 22 at m/z 595.145 had a mass that was slightly less than the kaempferol rutinoside (m/z 595.165). This compound had two abundant fragments at m/z 287.055 (consistent with kaempferol) and m/z 147.044 (consistent with p-coumaroyl; Figure 1E). A compound known as trans-tiliroside has a p-coumaric acid, that is, attached to the glucose unit, that is, linked to the kaempferol (kaempferol 3-O-[6”-O-p-coumaroyl]-glucoside). In our analysis, a low-abundance fragment-ion appeared at m/z 309.096, indicating a p-coumaric ester that was linked to an hexosyl residue, as described in Croton cajucara (Nascimento et al., 2017). This strongly suggested that peak 22 was the trans-tiliroside compound. Compound 23 at m/z 300.050 was identified as quercetin, and compound 24 at m/z 287.055 was identified as kaempferol, based on comparisons with authentic standards.

Hypertension, dyslipidemia, and smoking increased ALT and AST levels by ∼180% compared with the basal group (29.50 ± 1.11 U/L and 36.53 ± 1.58 U/L, respectively). Treatment with 300 mg/kg C. urucurana extract completely reversed the increase in ALT and AST levels. Treatment with 30 and 100 mg/kg C. urucurana extract and SIM + ENAL partially reversed these changes (Figure 2).

Hypertension, dyslipidemia, and smoking increased plasma triglyceride and cholesterol levels by 203.90 and 429.76%, respectively, compared with the basal group. The risk factors also increased hepatic triglyceride and cholesterol levels compared with basal values. Treatment with 300 mg/kg C. urucurana extract and SIM + ENAL completely reversed these changes, whereas 100 and 300 mg/kg C. urucurana extract only partially reversed hepatic triglyceride and cholesterol levels. Finally, the C- group exhibited an increase in fecal triglyceride and cholesterol levels that were not reversed by any of the treatments (Table 2).

The combination of hypertension, dyslipidemia, and smoking induced significant hepatic oxidative stress (Figure 3). Decreases in hepatic GSH levels (48.83 ± 2.86 µg/g tissue) were observed compared with the basal group (165.9 ± 4.48 µg/g tissue). Increases in LPO and SOD levels were observed in the C- group compared with the basal group (69.95 ± 4.22 mmol/min/g tissue and 1,030.00 ± 27.18 U/g of tissue, respectively). Treatment with 300 mg/kg C. urucurana extract completely reversed these changes, whereas 30 and 100 mg/kg C. urucurana extract and SIM + ENAL only partially reversed these changes.

Liver histopathological alterations are shown in Figure 4. No alterations were observed in the basal group. Lesions in the C- group were classified as 3 (massive lesions). Treatment with 300 mg/kg C. urucurana extract exerted significant hepatoprotective effects (score = 0.5, minor lesions). Treatment with 30 and 100 mg/kg C. urucurana extract and SIM + ENAL exerted moderate hepatoprotective effects (score = 2, marked lesions).