Protective effects of a chemically characterized extract from solanum torvum leaves on acetaminophen-induced liver injury

Abstract Distinct parts of Solanum torvum Swartz. (Solanaceae) are popularly used for a variety of therapeutic purposes. This study determined the phytochemical composition of a phenolic fraction of S. torvum leaf aqueous extract and investigated its antioxidant and liver-protective properties. A phenolic compound-enriched fraction, or phenolic fraction (STLAE-PF) of an infusion (STLAE) of S. torvum leaves, was tested in vitro (antagonism of H2O2 in cytotoxicity and DCF assays with HepG2/C3A cells), and in vivo for antioxidant activity and protective effects against acetaminophen (APAP)-induced liver injury in mice. Thirty-eight compounds (flavonoids, esters of hydroxycinnamic acid, and chlorogenic acid isomers) were tentatively identified (high-performance liquid chromatography coupled to high-resolution electrospray mass spectrometry) in the STLAE-PF fraction. In vitro assays in HepG2/C3A cells showed that STLAE-PF and some flavonoids contained in this phenolic fraction, at noncytotoxic levels, antagonized in a concentration-dependent manner the effects of a powerful oxidant agent (H2O2). In C57BL/6 mice, oral administration of STLAE (600 and 1,200 mg/kg bw) or STLAE-PF (300 mg/kg bw) prevented the rise in serum transaminases (ALT and AST), depletion of reduced glutathione (GSH) and elevation of thiobarbituric acid reactive species (TBARs) levels in the liver caused by APAP (600 mg/kg bw, i.p.). The hepatoprotective effects of STLAE-PF (300 mg/kg bw) against APAP-caused liver injury were comparable to those of N-acetyl-cysteine (NAC 300 or 600 mg/kg bw i.p.). These findings indicate that a phenolic fraction of S. torvum leaf extract (STLAE-PF) is a new phytotherapeutic agent potentially useful for preventing/treating liver injury caused by APAP overdosing.


Introduction
The imbalance between pro-and antioxidants species in favor of oxidants, or oxidative stress, is in some way involved in the mechanisms through which most chemicals and inflammatory conditions cause liver injury (Cicho_ z-Lach and Michalak 2014, Du et al. 2016, Jadeja et al. 2017, Garc ıa-Ruiz and Fern andez-Checa 2018, Masarone et al. 2018). Not surprising, a number of plant extracts and phytochemicals with potent antioxidant activities were shown to protect the rodent liver against the injury caused by drugs, hepatotoxins and diseases (Li et al. 2015).
Flavonoids and hydroxycinnamates are phenolic secondary plant metabolites with remarkable free radical scavenging properties. Some plant phenolics are among the most powerful natural antioxidants (Szymanska et al. 2018). Nonetheless, pharmacological modes of action other than merely the antioxidant activity, such as the interaction with cell signaling pathways, drug-metabolizing enzymes and others might also account for, or contribute to the putative hepatoprotective properties of several phytochemicals and plant extracts (Cicho_ z-Lach and Michalak 2014). Moreover, plant extracts are complex mixtures of multiple constituents each one of which may have a distinct pharmaco-toxicological profile, and thus, the net effect of extracts shall depend ultimately on how their phytochemical constituents interact with each other.
In this framework, and with an eye to the development of novel phytotherapic medicines, we studied the antioxidant activity and liver protective properties of a phenolic-rich fraction of extracts of a plant widely used in Brazilian folk medicine for dyspepsia (indigestion, abdominal discomfort), that in ordinary people's mind are generally thought to result from some kind of hepatic dysfunction (Agra et al. 2007, Nurit-Silva et al. 2011.
Solanum torvum Swartz. (Solanaceae), known as turkey berry, devil's fig, wild eggplant, and by other common names (e.g., 'jurubeba' in Brazil), is a perennial shrub 2-3 m height, native to southern Mexico, Central America and the Caribbean and tropical South America. It has been widely naturalized in tropical Africa, Asia, United States (Florida and

Extract preparation and partionioning
Preparation of aqueous extract and its partitioning to obtain a phenolic-rich fraction was carried out essentially as previously described for a S. paniculatum leaf extract (de Souza et al. 2019). Briefly, 150 g of S. torvum dried leaf powder was soaked with boiling distilled water (150 g/1500 mL) and let rest at room temperature for 3 h. The crude extract was then vacuum-filtered and freeze-dried to obtain the aqueous extract (STLAE, 13.1% yield) that was further partitioned with ethyl acetate (1:1). This fraction of STLAE, hereafter designated as phenolic fraction (STLAE-PF), is rich in phenolic compounds. The yield of the ethyl acetate fraction (phenolic fraction) was 1.45% w/w of dried leaves.

Phytochemical characterization of S. torvum phenolic fraction (STLAE-PF)
A preliminary thin-layer chromatography (TLC) analysis of the phenolic fraction of S. torvum leaf extract (STLAE-PF) was performed. The TLC analysis of STLAE-PF was carried out basically as described in detail by de Souza et al. (2019) for another plant extract.
The phenolic fraction of S. torvum leaf extract (STLAE-PF) was analyzed by liquid chromatography (HPLC-DAD and HPLC-ESIMS). Equipment, columns, and chromatographic conditions used in both analyses were basically the same employed previously for the phytochemical characterization of S. paniculatum leaf aqueous extracts. Details of the HPLC-DAD and HPLC-ESIMS methods can be found in de Souza et al. (2019).
Shortly, a HPLC-DAD Agilent 1260 system (Agilent Technologies, CA, USA) consisting of a quaternary pump, an autosampler and a diode-array detector (DAD) was used. Separation was achieved by using ACE-3 C18 column (100 mm Â 2.1 mm i.d., 3.0 mm particle size) coupled to a 5 mm Â 2.1 mm (i.d.) Â 2.7 mm C18 guard column. The mobile phases and the gradient elution program were as described by de Souza et al. (2019).
The phenolic compounds found in STLAE-PF were quantified by external standardization with rutin and chlorogenic acid as standards in concentrations ranging from 2.5 to 1000 mg/mL. All injections were in triplicate and the STLAE-PF constituents were determined by peak-area measurement. Mono-and di-caffeoylquinic acid isomers, and the monocoumaroylquinic acid, were quantified based on their molar absorptivities as described elsewhere (Trugo and Macrae 1984).
The HPLC-ESIMS analysis used an Agilent Infinity 1260 equipment and a MicrOTOF Bruker mass spectrometer. Analytical column, gradient, ESI LC-MS parameters, and optimum values of ESI-MS parameters were those previously reported by de Souza et al. (2019). The UV spectra and the fragmentation profiles were used for compounds identification, taking the measured masses at a resolution of 20.000 and a standard error better than 5 ppm. The spectra were calibrated before the identification of compounds, and operations were controlled by a Data Analysis 3.4 software (Bruker Daltonics). The software provides a list of possible elemental formulas by using the Generate Molecular Formula Editor.
2.5. Evaluation of the effects of S. torvum extract (STLAE) or its phenolic fraction (STLAE-PF) on acetaminophen-induced liver injury and oxidative stress

Animals
Eight-to 10-week-old male mice (Mus musculus, C57BL/6 inbred strain, 20-25 g body weight) bred by the Oswaldo Cruz Foundation were used in the tests. Upon arrival at the laboratory animal facilities, all mice were housed in standard plastic cages (5 in each cage) with stainless steel coverlids and certified pinewood shavings as bedding. The animals were kept under controlled environmental conditions (12-h light/12-h dark cycle; lights on from 8:00 a.m. to 8:00 p.m.; room temperature, 23 ± 2 C; air relative humidity, approximately 70%) with free access to a commercial rodent pellet diet (Nuvital CR1, Nuvital, Curitiba, PR, Brazil) and filtered tap water, except on the day of treatment. Animals were handled and used in accordance with the Brazilian animal protection and welfare legislation and international guidelines on the ethics of animal experimentation. All efforts were made to minimize the experimental animals' suffering, stress, and discomfort. The study protocol was approved (LW-82/12) by the Ethics Committee on the Use of Animals of the research institution.

Treatment
The mice were treated with a single intraperitoneal (i.p.) injection of acetaminophen (APAP) suspended in pharmaceutical grade corn oil (Sigma-Aldrich) with DMSO (10% v/v) (APAP 600 mg/kg bw), or N-acetyl-cysteine (NAC) dissolved in phosphate-buffered saline (PBS) (NAC 300 or 600 mg/kg bw). The S. torvum leaf aqueous extract (STLAE, 600 and 1200 mg/ kg bw) and its phenolic fraction (STLAE-PF, 300 mg/kg bw) were dissolved in PBS with DMSO (10% v/v) and administered by an intragastric cannula (gavage, p.o.). Control mice received equal volumes À 8 mL/kg bw -of the vehicle alone by i.p. or p.o. routes. The vehicle used for each substance, and doses, routes of administration and volumes given to the animals are summarized in Table 1. All control and treated groups (N ¼ 7 per group) are listed in Table 2. NAC, STLAE and STLAE-PF were administered 60 min after APAP injection. Eighteen hours after the administration of NAC or plant extracts, or their respective vehicles, and 19 h after injection of APAP, or its vehicle, experimental animals were euthanized by cervical dislocation. The model mouse gender (males), dose of APAP and time intervals between APAP injection and blood sampling and subsequent euthanasia were chosen based on previous experiments and on published studies (Dai et al. 2006, Mossanen andTacke 2015).
2.5.3. Determination of alanine (ALT) and aspartate (AST) aminotransferase activities in the blood serum Serum alanine (ALT) and aspartate (AST) aminotransferase activities were determined by a colorimetric method (Reitman and Frankel 1957) using a commercially available kit (Laborclin V R , Pinhais, PR, Brazil) adapted to a multi-well plate spectrophotometer reader (Spectramax Plus V R , Molecular Devices, USA). The absorbance was registered at 505 nm and results were expressed as IU/L. Blood samples were taken (from retro-orbital sinuses) immediately before euthanasia and the serum was separated by centrifugation.

Determination of reduced glutathione (GSH) levels in the hepatic tissue
The livers were removed immediately after mouse death, promptly cleaned from extra tissue, and wrapped in aluminum foil for storage in liquid nitrogen until further use. Concentrations of glutathione (GSH) in the liver tissue were measured as previously described in the literature (Zhu et al. 1995, Cao et al. 2003. Briefly, frozen livers were thawed on ice and homogenized in 2 mL of 100 mM sodium phosphate buffer pH 8 with 5 mM EDTA. The supernatant of homogenate centrifugation (10,000 Â g, 15 min at 4 C) was aliquoted in cryotubes that were immersed in liquid nitrogen. The supernatant was thawed on ice, and the supernatant fraction (10 mL) was added to tubes with 37 mL 5 mM EDTA 0.1 M sodium phosphate buffer pH 8 and 12.5 mL 25% phosphoric acid. The tubes were let rest on ice for 10 min in the dark. The mixture containing supernatant fraction was centrifuged (13,000 Â g, 10 min, 4 C) and the supernatant from this second centrifugation was distributed into wells of 96-well microplates. A phosphate alkaline buffer and a 0.1% o-phthaldialdehyde methanol solution were added to microplate wells. Microplates were incubated at room temperature for 10 min in the dark. Concentrations of GSH were then measured using a fluorimeter reader for microplates (Gemini XPS V R Molecular Devices) with wavelength parameters set at 350 nm (excitation), 420 nm (emission) and a standard curve made with serial dilutions of a 20 mM GSH solution.
Concentrations were expressed as nmol of GSH per mg of protein. The total protein concentration in the supernatant fraction was determined by the Bradford method (Bradford 1976) using bovine serum albumin (BSA) as the standard.
2.5.5. Measurement of thiobarbituric acid reactive substances (TBARs) TBARs in the liver tissue was assayed basically as reported by Hermes-Lima (1995) with a few adaptations. Shortly, two pieces of 0.1 g of hepatic tissue (one for the sample and the other for the blank) were separately homogenized (1:5 w/v) in a Potter-Elvejhem homogenizer with 0.4 mL of 0.2% phosphoric acid on ice. After addition of 0.5 mL of 2% phosphoric acid, the tubes were homogenized again (9,500 Â g for 10 min) and the supernatant was placed into new glass tubes where it was incubated with 1% TBA in 50 mM NaOH and 7% phosphoric acid. The blank tubes received identical volumes of 4 mM HCl (instead of TBA solution) and 7% phosphoric acid. Both sample and blank tubes were heated in boiling water for 15-16 min and the appearance of a typical red pinkish color was noted. After cooling (room temperature), 1.5 mL butanol was added to the tubes that were vortexed vigorously for 40 sec and centrifuged (3,500 Â g) for 10 min. Two distinct phases appeared, an organic phase on the top (TBA-MDAcomplex) and an aqueous phase on the bottom. The organic phase was pipetted off, put into fresh disposable borosilicate glass tubes and the absorbance was read (Shimadzu UV1601 spectrophotometer) at 600, 547, 532, 520 and 450 nm. The production of TBARs was expressed as nmol of TBARs per g of liver tissue.
2.6. In vitro assays of extracts in human hepatoma cell line 2.6.1. Cell culture C3A cells [ATCC CRL-1074, a hepatocarcinoma cell line derived from HepG2 (ATCC HB-8065)], were seeded in DMEM complemented with 100 U/mL penicillin-100 mg/mL streptomycin (Cultilab LTDA, Brazil), 10% fetal bovine serum (FBS, Cultilab LTDA, Brazil) and 2 mM glutamine in a 37 C incubator with 5% CO 2 . Cells were subcultured in 25-or 75-cm 2 culture flasks once a week and culture medium was also changed once a week. The cells used in the experiments were from passages 10 through 29.

Cytotoxicity assays
Cytotoxicity assays to evaluate lactate dehydrogenase (LDH) release were performed in 96-well plates seeded with 5 Â 10 4 cells/well that were exposed to treatment solutions for 20 h at 37 C in an incubator with 5% CO 2 . CytoTox 96 V R Non-Radioactive Cytotoxicity Assay kit (Promega, USA) was used and bovine serum LDH was the positive control. Three independent experiments (each one with triple replicates) were performed according to the manufacturer's instructions.
Results were read at 490 nm in a Biochrom V R EZ Read 2000 microplate spectrophotometer reader and expressed as % of total LDH released in the culture medium after addition of a 0.9% Triton X-100 solution.
Initially, cells were treated with a range of concentrations of STLAE-PF and SPLAE-PF (6.25-800 mM), and isoquercitrin, rutin, quercetin and chlorogenic acid (12.5-800 mM) to find the No Observed Effect Concentrations (NOEC). Controls received only the culture medium without or with DMSO 0.1%. The H 2 O 2 solutions were added to microplate wells after the plant extracts (200 mg/mL STLAE-PF or SPLAE-PF) or isolated phenolic compounds (25 mM quercetin, 100 mM chlorogenic acid, 200 rutin or isoquercitrin).

Oxidative stress indices
Oxidative stress was evaluated using the hydrophilic chemical 2 0 ,7 0 -dichlorofluorescein diacetate (DCFH 2 -DA) as follows. C3A cells were seeded in 24-well microplates with a density of 2.5 Â 10 5 cells/well and incubated at 37 C with 5% CO 2 for 24 h. At the end of the incubation period, cells were washed witth a Krebs-Ringer Hepes (KRH) buffer, DCFH-DA 10 mM was added and the plates were incubated at 37 C with 5% CO 2 for 45 min. The cells were washed again with KRH buffer before addition of H 2 O 2 (positive control), plant extracts or phenolic compounds. In co-treatments (H 2 O 2 þ test substances), H 2 O 2 was added to the microplate wells after the test substances. The plates were incubated for 2 h and then fluorescence was measured using a spectrofluorimeter Shimadzu RF5301PC with the following parameters: 480 nm excitation, 530 nm emission wavelengths and band slit width at 5 nm (Severi et al. 2007, Grasberger et al. 2013).

Statistical methods
The data were evaluated by one-way ANOVA followed by Bonferroni post hoc test or Kruskal-Wallis and Mann-Whitney. Student's t-test was used wherever only two group means were compared. Differences were considered as statistically significant when p < 0.05. Statistical software Graph Pad Prism version 5 software was used for calculations.

Chemical characterization of phenolic fraction (STLAE-PF) of S. torvum leaf aqueous extract
The thin-layer chromatographic (TLC) analysis of the phenolic fraction (STLAE-PF) of S. torvum leaf aqueous extract (STLAE) confirmed that it is rich in phenolic compounds and devoid of saponins and glycoalkaloids including solanine, solanidine, solasodine and other toxic alkaloids found in Solanum spp. A similar TLC analysis of S. paniculatum leaf aqueous extracts also found that they are virtually free of saponins and glycoalkaloids (de Souza et al. 2019).
The chemical profile of STLAE-PF was further determined by liquid chromatography with mass sprectrometry analysis (HPLC-HRESIMS) and 38 major constituents were tentatively identified ( Table 3). The STLAE-PF base peak chromatogram in negative electrospray ionization mode is displayed in Figure 1 and compounds' retention times, UV absorption peaks, experimental and calculated m/z, molecular formula, error, sigma values (comparison of theoretical with measured isotope patterns), and the proposed identities of 38 major compounds are presented in Table 3. The major constituents of STLAE-PF are glycosylated flavonols, esters of hydroxycinnamic acid and isomers of chlorogenic acid such as monocaffeoylquinic acids, monocoumaroylquinic acids, quercetin robinobioside, rutin, hexoside of hydroxyluteolin, dicaffeoyl quinic acids, quercetin and hexoside of methylquercetin. The total content of flavonoids and hydroxycinnamates in STLAE-PF was quantified by HPLC-DAD analyses using rutin and chlorogenic acid standards to build up a calibration curve. Both curves showed good linearity between concentration and absorbance. The total amount of hydroxycinnamates, expressed as chlorogenic acid, was 110.41 ± 0.12 mg/g of dried sample. The total flavonoids content expressed as rutin was 8.58 ± 0.03 mg/g of dried sample (Table 4).

Effects of S. torvum extracts (STLAE and STLAE-PF) on the acetaminophen-induced liver injury and oxidative stress
As shown in Figure 2, an overdose of APAP (600 mg/kg bw i.p.) caused marked elevations in mouse serum transaminase activities (>3-fold and 2-fold increases in ALT and AST, respectively). This rise in both serum biomarkers of liver injury was effectively prevented by a subsequent injection of NAC (300 and 600 mg/kg bw i.p.), the antidote of choice for treatment of APAP-poisoning. Similarly, a single oral administration of STLAE (600 and 1200 mg/kg bw p.o.) after APAP injection prevented or drastically attenuated, in a dosedependent manner, the enhancement of ALT and AST activities. The hepatoprotective effect of the STLAE-PF fraction (300 mg/kg bw p.o.) was comparable to that of NAC (300 mg/ kg bw i.p.), and stronger than the effect of the crude aqueous extrat (STLAE).
APAP overdosing also caused a depletion of reduced gluthatione (GSH) in the mouse hepatic tissue (Figure 3). The APAP-produced reduction in the hepatic content of GSH was effectively counteracted by NAC (600 mg/kg bw i.p.), STLAE (600 or 1,200 mg/kg bw p.o.) and STLAE-PF (300 mg/kg bw p.o.), but not by the lowest dose of NAC (300 mg/kg bw i.p.) ( Figure 3). It is of note that GSH levels in the liver of mice treated with APAP plus STLAE-PF were greater than the GSH levels in the hepatic tissue of both APAP-treated (GSH depleted) and APAP-untreated control (i.e., baseline GSH levels) mice ( Figure 3).
Moreover, S. torvum leaf extract (STLAE) and its phenolic fraction (STLAE-PF) and NAC prevented APAP-induced rise in TBARs levels in the liver (Figure 4). TBARs are by-products of lipid peroxidation elicited by reactive oxygen species (ROS) and their hepatic levels are a tissue oxidative stress biomarker (Tsikas 2017).
The cells were exposed to 2 0 ,7 0 -dichlorodihydrofluorescein diacetate (DCFH 2 -DA) that, by passive diffusion, enters the cytosol where esterases hydrolyze the ester-bound thereby releasing the acetate groups. The 2 0 ,7 0 -dichlorodihydrofluorescein (DCFH 2 ) formed inside the cell is nonfluorescent but under the action of oxygen/nitrogen reactive species it undergoes a two-electron oxidation to generate the highly fluorescent compound DCF (dichlorofluorescein) (Wardman 2007).
SPLAE-PF, STLAE-PF, quercetin and rutin produced a concentration-dependent inhibition of H 2 O 2 elicited enhancement of fluorescence or cellular oxidative stress. At higher concentrations of these substances (supplementary file), however, both phenolic fractions (400 mg/mL of STLAE-PF or SPLAE-PF) and quercetin (50 mM) overtly enhanced the DCFassay oxidative stress response over the cell background (untreated) levels, while the glycosylated flavonoid rutin (400 mM) did it only modestly.
These results showed that quercetin inhibited the H 2 O 2 -caused cytotoxicity and oxidative stress at molar concentrations lower than those required by its glycosylated derivatives (rutin and isoquercitrin) to do it, a finding that is consistent with the notion that aglycones are as a rule more potent antioxidants than their corresponding glycosides (Kumar and Pandey 2013). At higher concentrations, however, quercetin proved to be a more powerful cytotoxic and prooxidant agent (NOECs ¼ 25 mM) than its glycosylated counterparts, rutin (NOECs ¼ 200 mM), isoquercitrin (NOEC ¼ 200 mM) and than chlorogenic acid (cytotoxicity NOEC ¼ 100 mM) as well.

Discussion
Overall, results from this study indicated that a phenolic-rich fraction (STLAE-PF) of S. torvum leaf aqueous extract has liver protective properties at doses comparable to those of N-acetyl-cysteine (NAC), the antidote of choice for APAP poisoning. The administration of a single overdose of APAP to mice is one of the most suitable experimental models to study the pathophysiology of acute liver injury in humans (Maes et al. 2016). APAP is rapidly absorbed in the small intestines, conjugated with glucuronic acid or sulfate by liver UGTs or SULTs, respectively, and excreted by the kidney. A small proportion of the absorbed APAP (<20%), however, undergoes oxidation by CYP2E1 and, to a lesser degree by CYP1A2, to form N-acetyl-p-benzoquinone imine (NAPQI), a highly reactive metabolite that, if doses of APAP are within the recommended therapeutic dose-range, is promptly conjugated with glutathione (GSH) by GSTs. Nonetheless, if large doses of APAP are consumed over a short time period, formation of NAPQI may exceed the liver capacity to conjugate it because the organ stores of GSH are depleted, and thus NAPQI remains largely unconjugated. The excess NAPQI is free to react with other cellular targets, such as sulfhydryl groups in cysteine residues on proteins, triggering a cascade of events that leads to cell death. The mode by which NAPQI ultimately damages the liver tissue is complex and so is the mode of action of its classical antidote N-acetyl-cysteine (NAC). Besides binding directly NAPQI to its cysteine thiol group, NAC also promotes GSH replenishment, scavenges reactive oxygen species in mitochondria and seems to support the mitochondrial energy metabolism .
The data presented here showed that the phenolic fraction (STLAE-PF) of S. torvum leaf extract mimicked some or most of NAC actions in the liver after an overdose of APAP, that is, it promotes the replenishment of GSH and prevents the rise in TBARS levels.
The quali-quantitative chemical analysis of the STLAE-PF fraction revealed that it is a complex multicomponent mixture, the major phenolic constituents of which are glycosylated flavonoids (e.g., glycosides of kaempferol and quercetin) and hydroxycinnamates (Figure 1, Table 3). The chemical composition of STLAE-PF is similar to that of a phenolic-rich fraction of S. paniculatum leaf extract that de Souza et al. (2019) described that also presents potent liverprotective, antioxidant and analgesic activities. Compared to the previously studied S. paniculatum phenolic fraction (de Souza et al. 2019), the S. torvum STLAE-PF has a nearly 20% higher total content of phenolics and contains a greater amount of hydroxycinnamates (110.41 ± 0.12 vs 76.29 ± 0.01 mg/g) and a smaller amount of flavonoids (8.58 ± 0.03 vs 18.01 ± 0.04 mg/g) (Table 4).  Table 3. 1.86 ± 0.013 Values are mean ± SD.
As aforementioned, bioavailable antioxidants in plant extracts are likely to confer a certain degree of protection against a variety of hepatotoxic agents that cause liver injury through mechanisms involving oxidative stress, or a disruption of cellular redox status. Biological activities of plantderived phenolics other than merely the antioxidant capacity, however, may also contribute to their liver protective properties.
Quercetin, rutin and other flavonoids, for instance, inhibit the NF-jB signaling pathway thereby exerting antiinflammatory actions in the hepatic tissue (Granado-Serrano et al. 2012, Tang et al. 2016. Inflammation plays an important role in a diversity of liver diseases, including hepatocarcinogenesis and fibrosis, and on drug-(e.g., APAP-) elicited liver injury (Del Campo et al. 2018, Woolbright and. Moreover, it was shown that the Nrf2 signaling pathway also plays a pivotal role in inflammatory responses. A variety of phytochemicals including quercetin, curcumin and others activate the Nrf2 transcription factor and, by doing so, they upregulate the expression of heme oxygenase 1 (HO-1) and other cytoprotective genes, what ultimately leads to inhibition of inflammation progression and tissue damage (Yao et al. 2007, Ahmed et al. 2017. It is plausible to think, therefore, that distinct actions of plant-derived phenolic compounds (i.e., direct ROS scavenging activity, inhibition of NF-jB pathways and pharmacological activation of Nrf2 transcription factor) are likely to act together enhancing not only their intracellular antioxidant activity, but also their anti-inflammatory effects.
Moreover, some studies in rodents suggested that quercetin and possibly also other plant phenolics could prevent hepatic fibrosis and/or reverse cirrhosis (induced by CCl 4 and other hepatotoxic agents) by inhibiting liver stellate cell activation, and attenuating autophagy via suppression of TGF-b1/Smads and activation of PI3K/Akt signaling pathways (Marcolin et al. 2012, Casas-Grajales et al. 2017, Wu et al. 2017, Li et al. 2018. These findings indicate that plantderived phenolic compounds besides preventing the acute hepatocellular injury caused by APAP, might also be useful agents to attenuate and/or reverse fibrosis and cirrhosis following liver diseases (hepatitis), chronic alcoholism and exposure to hepatotoxic drugs and toxins.
Clinical investigations have demonstrated that, although undergoing a marked pre-systemic clearance (liver first-pass effect), NAC, be it given by the oral or by the intravenous route, is equally effective in preventing drug-induced liver injury (Borgstr€ om et al. 1986, Kanter 2006, Green et al. 2013. Studies, however, pointed out that NAC is effective as a hepatoprotective agent only if patients receive it within the first 8 to 10 hours after APAP poisoning. There is no clinical evidence that it is of benefit if used at later stages of APAPcaused liver injury. Along this line, findings of experimental research are inconsistent, but at least one study by Yang et al. (2009) indicated that prolonged treatment with NAC could even impair hepatic tissue regeneration in acute liver injury caused by APAP in mice.
Since rodent studies consistently suggested that flavonoids ameliorate liver fibrosis/cirrhosis, we could expect a similar beneficial effect of repeated administration of phenolic-enriched fraction of plant extracts such as STLAE-PF. This hypothesis deserves to be further tested. If it holds true, then STLAE-PF and similar plant extractsthat generally cause no remarkable adverse events -might be of therapeutic benefit not only at the early stage, but also at later stages of development/progression of liver injury caused by APAP and other drugs and conditions.
Notwithstanding the fact that there is abundant experimental evidence (from in vitro and in vivo assays) indicating that flavonoids and related plant phenolics are potentially beneficial to health, their bioavailability in humans remains a critical issue for their possible use as phytotherapeutic agents  ,360.1 ± 25.58;NAC300 þ APAP,190.3 ± 53.09;NAC600 þ APAP,140.1 ± 8.22;STLAE600 þ APAP,160.5 ± 9.73;STLAE1,200 þ APAP,161.8 ± 44.36 and STLAE-PF300 þ APAP, 155.9 ± 11.04. Data were evaluated by ANOVA and Bonferroni's test applied to the original values (untransformed data). Differences (p < 0.05) from control and APAP only treated groups are indicated by letters above the boxes (a 6 ¼ controls and b 6 ¼ APAP). (Del Rio et al. 2013, Kumar and Pandey 2013, Williamson et al. 2018. Oral bioavailability of plant phenolics is relatively low owing to a pronounced pre-systemic clearance. Glycosylated flavonoids are mostly deglycosylated within the digestive tract. The released aglycones are then absorbed in the small intestines, undergo conjugation in the intestines and liver by SULTs, UGTs or COMTs, and appear in the blood as sulfate, glucuronide or methylated phase II metabolites. Colonic microbiota also plays a role in the metabolism of flavonoids. Enterohepatic recirculation may also occur. In the colon lumen, microorganisms cleave conjugating moieties and the resultant aglycones undergo ring fission giving rise to smaller molecules among which phenolic acids and hydroxycinnamates that are then absorbed. Since these products of flavonoid gut metabolism may also be active, the in vivo pharmaco-toxicological activity of ingested flavonoids arises from the interaction of absorbed parent compounds with their metabolites produced by the action of gut microbiota (Williamson et al. 2018). The gut microbiota thus seems to be a key factor in the absorption of plant phenolics and in their pharmaco-toxicological effects as well (Kawabata et al. 2019).
In this study, we demonstrated that a mixture of plant phenolics (STLAE-PF) antagonized the GSH depletion, the increase in lipid peroxidation and the rise in serum biomarkers of hepatocellular damage caused by an overdose of APAP in mice. These results indicated that, as far as mice are concerned, pharmacologically active compounds in STLAE-PF were bioavailable to exert a liver protective activity. It is of note that a delay of gastric emptying caused by plant extracts (STLAE-PF) that could have reduced APAP absorption did not take place because mice received APAP by the i.p. route. An interference of the plant phenolics with APAP kinetics, however, might have contributed to the hepatoprotection noted in the mouse model. Some flavonols and hydroxycinnamates are inhibitors of CYP2E1 and/or CYP1A2 activity in in vitro assays, and Bedada and Neerati (2018) demonstrated that quercetin prolonged the elimination of chlorzoxazone (a CYP2E1 substrate) in healthy volunteers. It is unclear, however, whether S. torvum leaf extract and its phenolic fraction STLAE-PF would inhibit CYP2E1 and CYP1A2-mediated conversion of APAP into its NAPQI reactive metabolite, or if it indeed does it, whether it would inhibit APAP metabolic activation to the degree needed for preventing the hepatocellular injury.
In conclusion, the phenolic compounds found in STLAE-PF were shown to possess a set of distinct pharmacological activities that may converge to explain its liver protective properties, particularly against APAP-induced acute liver injury.
As shown by its chemical profile, STLAE-PF is a rather complex mixture of plant phenolics the major constituents of which are glycosylated flavonols and hydroxycinnamates. Flavonoids and other plant phenolics go through extensive gut microbiota metabolism and undergo a marked first-pass metabolism in the intestines and liver, so that they appear in the blood as aglycone conjugates, smaller phenolic acids and hydroxycinnamate metabolites rather than as the ingested parent compounds. Therefore, the in vivo liver protective properties of STLAE-PF are likely to result from a combined action of a mixture of biologically active metabolites formed from the ingested plant phenolic compounds.
APAP is one of the most used analgesic and antipyretic drug worldwide and accounts for nearly 50% of acute liver failure cases in the United States (Hodgman and Garrard 2012). If not promptly and effectively treated, APAP poisoning may result in patient death or liver transplantation. So far, NAC is the only antidote widely available for treatment and prevention of APAP-induced acute liver injury. NAC, however, is an expensive antidote the use of which in all patients is not cost-effective and unaffordable in many developing countries (Senarathna et al. 2012). Plant phenolics and several S. torvum extracts were reported to present analgesic and antiinflammatory activities as well (Ndebia et al. 2006). A possible combination of analgesic/antiinflammatory, antioxidant and liver protective activities seems to make this plant phenolic mixture (STLAE-PF), either alone or combined to APAP in a pharmaceutical formulation, a promising novel and comparatively more costeffective phytotherapeutic agent.
Since gut microbiota plays a key role in the bioavailability and in vivo activity of plant phenolics (Cassidy andMinihane 2017, Kawabata et al. 2019), and the intestinal microbiota varies between rodent models and humans (Franklin andEricsson 2017, Turner 2018), differences may eventually exist between STLAE-PF hepatoprotective activity in the murine model and its clinical efficacy in humans. Evidence from further clinical trials of STLAE-PF or similar plant phenolic-rich fractions should clarify this issue.