Hepatoprotective Activity of Leptadenia hastata (Asclepiadaceae) on Acetaminophen-Induced Toxicity in Mice: In Vivo Study and Characterization of Bioactive Compounds through Molecular Docking Approaches

Materials and Methods Various aqueous extracts were prepared from this plant and preadministered per os to albino mice 3 h before APAP administration, once daily for one week. Animals from the normal group were given only distilled water while those from negative control received only APAP 250 mg/kg. After treatment, mice were sacrificed, the liver was collected for histopathology analysis, and different biochemical markers (alanine aminotransferase (ALT), superoxide dismutase (SOD), catalase (CAT), glutathione (GSH), malondialdehyde (MDA), and tumor necrosis factor-alpha (TNFα)) were measured. The content of the active extract was analyzed by HPLC/UV. Molecular docking was conducted using iGEMDOCK software, and the drug-likeness and pharmacokinetic profiles were evaluated using Swiss ADME. Results APAP administration significantly increased (p < 0.001) ALT in liver homogenates when compared to normal controls whereas the stem decoction at 250 mg/kg significantly (p < 0.001) reduced this activity to a normal value comparable to silymarin 50 mg/kg which is better than leaf and root extracts. Moreover, the stem decoction also significantly reduced the MDA levels (p < 0.05) and increased those of GSH, SOD, and CAT (p < 0.001) at doses of 250 and 500 mg/kg compared to the negative control. A significant (p < 0.001) decrease of TNFα levels and leukocyte infiltration was observed following treatment with this extract. The HPLC/UV analysis of the decoction revealed the presence of dihydroxycoumarin, quinine, and scopoletin with the following retention times: 2.6, 5.1, and 7.01 min, respectively. In silico studies showed that quinine and dihydroxycoumarin had great potentials to be orally administered drugs and possessed strong binding affinities with TNFα, TNF receptor, cyclooxygenase-2, iNOS, cytochrome P450 2E1, and GSH reductase. Conclusion Based on these results, L. hastata could be considered a source of promising hepatoprotective compounds with antioxidant and anti-inflammatory properties.

Background and Objectives. Leptadenia hastata is a liana from the family of Asclepiadaceae used in tropical Africa to treat diabetes mellitus. In this study, we investigated its hepatoprotective mechanisms on acetaminophen-(APAP-) induced toxicity through in vivo and in silico approaches. Materials and Methods. Various aqueous extracts were prepared from this plant and preadministered per os to albino mice 3 h before APAP administration, once daily for one week. Animals from the normal group were given only distilled water while those from negative control received only APAP 250 mg/kg. After treatment, mice were sacrificed, the liver was collected for histopathology analysis, and different biochemical markers (alanine aminotransferase (ALT), superoxide dismutase (SOD), catalase (CAT), glutathione (GSH), malondialdehyde (MDA), and tumor necrosis factoralpha (TNFα)) were measured. The content of the active extract was analyzed by HPLC/UV. Molecular docking was conducted using iGEMDOCK software, and the drug-likeness and pharmacokinetic profiles were evaluated using Swiss ADME. Results. APAP administration significantly increased (p < 0:001) ALT in liver homogenates when compared to normal controls whereas the stem decoction at 250 mg/kg significantly (p < 0:001) reduced this activity to a normal value comparable to silymarin 50 mg/kg which is better than leaf and root extracts. Moreover, the stem decoction also significantly reduced the MDA levels (p < 0:05) and increased those of GSH, SOD, and CAT (p < 0:001) at doses of 250 and 500 mg/kg compared to the negative control. A significant (p < 0:001) decrease of TNFα levels and leukocyte infiltration was observed following treatment with this extract. The HPLC/UV analysis of the decoction revealed the presence of dihydroxycoumarin, quinine, and scopoletin with the following retention times: 2.6, 5.1, and 7.01 min, respectively. In silico studies showed that quinine and dihydroxycoumarin had great potentials to be orally administered drugs and possessed strong binding affinities with TNFα, TNF receptor, cyclooxygenase-2, iNOS, cytochrome P450 2E1, and GSH reductase. Conclusion. Based on these results, L. hastata could be considered a source of promising hepatoprotective compounds with antioxidant and anti-inflammatory properties.

Introduction
The liver is a central organ in energy metabolism and the biotransformation of xenobiotics. Therefore, frequent exposure to toxic xenobiotics is likely to provoke a liver injury, resulting in cirrhosis, liver cancer, and acute liver failure [1]. Acetaminophen (paracetamol or N-acetyl-para-aminophenol (APAP)) is one of the most widely used analgesics and antipyretic drugs worldwide. Although considered safe at therapeutic doses (up to 4000 mg/day), APAP, at higher doses, can induce centrilobular necrosis which generally leads to a fatal outcome [2]. APAP intoxication would be responsible for about one-half of all cases of acute liver failure in the United States and the United Kingdom [3]. In a recent study in Thailand where 184 patients with APAP overdose were included, 15.6% were reported with mild hepatotoxicity and 6.4% developed severe hepatotoxicity while 3 (1.6%) patients had acute liver failure [4].
APAP hepatotoxicity is initiated by the production of Nacetyl-p-quinone-imine (NAPQI), a reactive metabolite generated by cytochrome P450 enzymes that metabolize the drug when present at high doses. This compound then depletes glutathione stores and binds to several cellular proteins especially mitochondrial proteins, thereby leading to mitochondrial oxidant stress which causes cell death [4]. Moreover, it was shown that the immune system plays a role in the progression of APAP-induced hepatotoxicity. Indeed, activated Kupffer cells produce some proinflammatory and chemotactic cytokines which promote infiltration of neutrophils and macrophages in the liver tissue, causing therefore an exacerbation of liver damage [5]. To block this toxicity, coadministration of N-acetylcysteine (NAC), a cysteinederived antidote, is often useful, but some side effects, such as hypotension, limit its efficacy [6]. Consequently, the search for novel liver-protective agents is necessary to reinforce the existing therapeutic arsenal.
L. hastata is a plant of the Asclepiadaceae family found everywhere in tropical Africa. The plant is consumed as legumes and used in traditional herbal medicine to treat diabetes [7]. Previous studies reported trypanocidal [8], hypoglycaemic and antidiabetic [9,10], anti-inflammatory [11,12], and antiandrogenic properties [13]. Previous reports highlighted the presence of several phytochemicals such as saponins, proanthocyanidins, phenolic glycosides, and flavonoids in the leaf extract [14][15][16] but also lutein, lupeol, β-carotene, kidjolanin, cynanforidin, and gagaminin were previously identified [16,17]. In recent years, the hepatoprotective effects of the methanolic leaf extract have been demonstrated against alcohol-induced liver injury [18]. The plant extract inhibited transaminases, total cholesterol, triglycerides, and total bilirubin in comparison with the alcohol-treated rats. Although this activity was reported, little is still known about the hepatoprotective mechanisms of L. hastata especially against APAP and especially how the plant active compounds interact with cell proteins to afford liver protection. Therefore, the objective of this study was to evaluate the protective effects of different aqueous extracts of L. hastata on APAP-induced liver damage in mice. We also analyzed the phytochemical profile of the aqueous decoction by high-performance liquid chromatography (HPLC) and conducted in silico studies of some of the active ingredients on anti-inflammatory and antioxidant mediators.

Plant
Material. The stems, leaves, and roots of L. hastata used in this study were harvested during the rainy season (August 2018) in Touboro (North, Cameroon). After harvesting and washing with water, the samples were shadedried for two weeks, then pulverized in a mortar, and the obtained powder was sieved through a 0.5 mm mesh sieve. The botanical identification of the plant was made by Prof. Mapongmetsem Pierre-Marie, a botanist at the Faculty of Science of the University of Ngaoundere, Cameroon.

Preparation of Plant Extracts.
For the different extract preparations, three methods of preparation, namely, infusion, decoction, and maceration were used. Twenty-five (25) g of the L. hastata powder was weighed and mixed with 250 mL of distilled water. For the decoction, the mixture was boiled at 100°C for 20 minutes. For the infusate, the powder was mixed with boiling water for 20 min, while for macerate, the powder was mixed with distilled water for 20 min at room temperature. After cooling, the mixtures were filtrated and the filtrates were frozen and lyophilized at -55°C. The freeze-dried extracts were packed in aluminum foil and stored in boxes at 4°C until use. Finally, test samples were solubilized in distilled water before giving to animals at the desired doses.

High-Performance Liquid Chromatography Analysis.
To identify the components of the stem aqueous decoction of L. hastata, an HPLC analysis was carried out on an Agilent 1260 Infinity brand chromatographic chain using a reversed-phase C18 Inerstil-ODS-3 column (150 × 4:6 mm, 5 μm, GL Sciences Inc., Tokyo, Japan) and equipped with a binary pump, autosampler, column oven, and a diode array detector from Agilent 1100 series, as previously described [19]. Twenty microliters of extract dissolved in methanol was injected into the column, and the column injector was kept at 35°C during the whole chromatography. The mobile phase consisted of phosphoric acid (H 3 PO 4 ), pH 2.5 (A), and acetonitrile (B). The gradient elution system was programmed as follows: 77% A and 23% B, 0-10 min at a flow rate of 0.7 mL/min; 55% A and 45% B, 11-20 min at a flow rate of 0.7 mL/min; 35% A and 65% B, 21-27 min, 5% A and 95% B, 28-31 min; and 77% A and 23% B, 32-35 min at a flow rate of 0.9 mL/min. The detection wavelength of the constituents was 320 nm. Compounds were identified based on the retention time and absorbance spectra of pure standards.  [20]. Additionally, hepatoprotective effects were recently demonstrated with leaf extracts using 250 and 500 mg/kg doses [18]. Consequently, we investigated the effects of aqueous leaf extracts using these doses. In order to evaluate the effect of extracts from different plant parts on APAP toxicity, 30 mice were segregated into 6 groups of 5 mice each, presented as follows: group I-normal control was provided with 10 mL/kg distilled water; group II-negative control was treated with 250 mg/kg APAP; group III-positive control received 250 mg/kg APAP and 50 mg/kg silymarin; and groups IV, V, and VI-these groups received 250 mg/kg of APAP and 250 mg/kg of the decoction of leaves, stems, and roots, respectively. Hepatotoxicity was induced by administering 250 mg/kg of APAP to all mice, except the normal controls which only received distilled water. The plant extract and silymarin were given orally 3 h after APAP treatment and once daily for one week.
At the end of this first experiment, different aqueous extracts (decoction, macerate, and infusion) were prepared from the most active plant part, and the hepatoprotective effects were assessed. Mice were grouped and treated as mentioned above. However, in groups IV, V, and VI, the animals received 250 mg/kg of APAP and 250 mg/kg of the decoction or macerate or infusion of the most effective plant part. Treatments lasted 7 days, and on the 8 th day, animals were sacrificed and the livers were collected, crushed and then centrifuged at 4000 rpm for 10 minutes at 4°C. Afterward, alanine aminotransferase (ALT) levels were then measured in liver samples. At the end of this test, the method of extraction with the most effective hepatoprotective activity was retained for further investigations at different concentrations of the plant extract. Seven groups of 5 mice each were used in this test as follows: groups I, II, and III were treated as mentioned above while Group IV, V and VI received 250 mg/kg of APAP and 100, 250, or 500 mg/kg of the stem decoction, respectively. Group VII received only 500 mg/kg of stem decoction daily for one week. At the end of treatment, mice were fasted for 12 h and thereafter sacrificed by cervical dislocation. The liver was collected, rinsed with a physiological saline solution, and weighed.
2.6. Biochemical Analysis. Liver ALT, superoxide dismutase (SOD), catalase (CAT), malondialdehyde (MDA), and glutathione (GSH) levels were determined using commercially available kits following the manufacturer's instructions. Briefly, the livers were homogenized in nine volumes of icecold 0.1 M phosphate buffer (pH 7.4). The homogenate was centrifuged at 10,000× g for 15 min at -4°C, and aliquots of 3 BioMed Research International the supernatants were collected for the quantification of biochemical parameters. ALT, SOD, CAT, MDA, and GSH were measured by colorimetric assays while tumor necrosis factor-alpha (TNFα) levels were determined by sandwich ELISA using a commercial kit from Solarbio Life Sciences (Beijing, China).

Histopathological Examination.
The liver tissue was fixed in 4% paraformaldehyde, embedded in paraffin, sectioned into 5 μm thickness, and stained with hematoxylin (H&E) for the evaluation of histopathological changes. Microscopic observations were done under a bright-field microscope. Different pathological grades were identified to evaluate the severity of inflammation: grade 0 (no pathological change), grade 1 (minimal leukocyte infiltration), grade 2 (mild), grade 3 (moderate), and grade 4 (important leukocyte infiltration) [20,21].
2.8. In Silico Molecular Docking. The phytochemical analysis of L. hastata aqueous extract by HPLC detected the presence of some compounds such as quinine, scopoletin, and dihy-droxycoumarin. These compounds (structures shown in Figure 1) were selected to evaluate the possible interactions with target proteins involved in the inflammation and the oxidative stress during APAP toxicity in mice. This allowed us to better analyze and understand the molecular mechanisms underlying the hepatoprotective activity of this plant. Docking studies were performed using iGEMDOCK version 2.1, a graphical automatic drug design system for docking, screening, and analysis provided by BioXGEM Lab.  BioMed Research International crystal structure of Mus musculus cytochrome P450 2E1 was not present in the PDB repository, its FASTA sequence (ID: Q05421) was obtained from Uniprot and a homology modeling was performed using the SWISS-MODEL program. The ligands were docked with PDB file's binding sites using an exact docking feature (standard docking). The following parameters were applied: population size = 200, generation = 70, and number of solutions = 2. The postanalytic method permitted us to visualize and categorize the evaluated compounds by combining drug interactions and energy-based measurement function.
2.9. ADME and Drug-Likeness Evaluation. For a compound to be considered a good drug candidate, its physicochemical properties must determine great chances of bioavailability. A means to verify this is to assess the drug-likeness profile of the molecule using Lipinski, Ghose, Veber, Egan, and Muegge's rules before any in vivo study. These rules help to predict the pharmacokinetic profile of the compound in the human body, notably absorption, distribution, metabolism, and excretion (ADME) [22]. In our study, we evaluated the drug-likeness features of the 4 docked compounds using calculations based on the abovementioned rules. The ADME parameters such as gastrointestinal absorption (GI), Blood-Brain Barrier (BBB), P-glycoprotein-(Pgp-) mediated efflux, and the capacity to be used as substrates of cytochrome P450 (CYP) were also analyzed using SwissADME, a free web tool available at http://www.swissadme.ch/.
2.10. Statistical Analysis. All data were expressed as the mean ± standard error to mean (SEM). Statistical analysis was carried out using the GraphPad Prism 5.0 software for windows. The comparison of means was performed using a one-way analysis of variance (ANOVA) followed by Dunnett's post hoc tests. Differences were considered significant when p < 0:05.

Results
3.1. L. hastata Significantly Improved Liver ALT Activities in APAP-Treated Mice. The quantification of ALT in liver

BioMed Research International
homogenates has been previously reported to be a reliable marker of liver injury [20,22]. Figure 2 shows the effect of L. hastata on the ALT activity in different experimental groups.

The Stem Decoction of L. hastata Modulated the Levels of Oxidative Parameters in the Liver of APAP-Treated Mice.
Oxidative stress is an important phenomenon in APAP pathogenesis. To evaluate the capacity of our plant extracts to regulate this process, we measured the effect of different doses of the decoction of L. hastata stems on MDA and GSH levels, as well as on CAT and SOD activities in each batch of mice after APAP poisoning.

Stem Decoction of L. hastata Counteracted Liver Inflammation in APAP Toxicity by Downregulating TNFα
Expression and Decreasing Leukocyte Infiltration. Administration of APAP (250 mg/kg) to mice induced an increased expression of TNFα protein in liver homogenates by 1.95fold as compared to normal control ( Figure 4). However, treatment with the stem decoction showed a significant dose-dependent decrease of TNFα with the 500 mg/kg dose standing out as the one with the better activity. Stem decoction at this dose decreased the TNFα concentration from 1250 ± 16:67 pg/mL to 633:3 ± 34:69 pg/mL similarly as silymarin (566:7 ± 66:67 pg/mL) as compared with the APAPtreated group.
The effect of L. hastata decoction was also examined through hematoxylin/eosin staining to confirm the necroinflammatory activity found in the liver tissue. As shown in Figure 5(a), APAP administration led to significant neutrophil infiltration around the portal vein, with hepatic necrosis compared to the normal group ( Figure 5(b)) where no pathological changes were observed. The plant decoction at 100 and 500 mg/kg decreased considerably the pathological score as compared to the APAP group. This decrease was similar to that of silymarin at 50 mg/kg (Table 1 and Figures 5(c)-5(f)). Nonetheless, with the plant extract at 250 mg/kg, a partial restoration of the liver structure was observed.
3.4. HPLC Profile of the Stem Decoction of L. Hastata. Figure 6 shows the HPLC-UV chromatogram of the aqueous decoction of L. hastata at 320 nm in comparison with those of the standard compounds used.
It appears from this figure that the plant extract contains several compounds, some of which have retention times corresponding to quinine (2.60 min), dihydroxycoumarin  BioMed Research International (5.51 min), and scopoletin (7.01 min). These compounds were therefore used for in silico studies.
3.5. In Silico Screening Data. From the docking results, higher binding energies were particularly found with dihydroxycoumarin and quinine than scopoletin during the interaction with the different PDB files. However, these binding energies remained lower compared to silibinin, used here as a refer-ence compound. As shown by Table 2 (Figure 7(b)). As far as CYP 2E1 is concerned, strong hydrogen bindings were established with Thr 307, Leu 363, Val 364, and Ala 429 in the enzyme binding site (Figure 7(c)). Concerning quinine, hydrogen bonds were formed with MSR 601 in the interaction with iNOS (Figure 8(a)), Arg 32 and His 101 in the interaction with GSH reductase (Figure 8(b)), and Ala 275, Tyr 284, and NAG 404 in the interaction with TNFα (Figure 8(c)). Also, quinine presented a binding profile closer to silibinin as far as TNFα, iNOS, and GSH reductase are concerned. For dihydroxycoumarin, its binding mechanism on TNFR is

Drug-Likeness and Pharmacokinetic Profile of the Docked
Compounds. The predicted pharmacokinetic and druglikeness profile of the different docked compounds is shown in Table 3. Except for silibinin, quinine, dihydroxycoumarin, and scopoletin have a high GI absorption and complied with almost all the drug-likeness rules including the Lipinski filter which proves their potential to be used as oral drugs. Besides, all docked compounds showed a good bioavailability score of 55%. However, silibinin and dihydroxycoumarin were found unable to cross the BBB. Furthermore, none of these compounds stood out as a substrate of permeability glycoprotein (P-gp), a transporter that permits to estimate active efflux through biological membranes. Concerning the skin permeation, the more negative the log Kp, the less skin permeant is the molecule. According to our data, quinine has the greatest chances to cross skin with a log Kp of -6.23 cm/s followed by scopoletin, dihydroxycoumarin, and silibinin. CYP p450 enzymes play an important role in drug elimination, through metabolic biotransformation. It was suggested that the inhibition of these enzymes is one of the major causes of drug-drug interactions leading to toxic or undesirable effects [22]. Quinine showed selectivity for CYP 2D6 while scopoletin and silibinin showed their selectivities for CYP 1A2 and CYP 3A4, respectively, suggesting their capacity to cause drug interactions. However, dihydroxycoumarin was found unable to cause such inhibition with all the p450 isoforms tested.

Discussion
APAP-induced hepatotoxicity especially in mice is an attractive and clinically relevant model that is widely used to assess the efficacy of hepatoprotective natural products. Indeed, similarities were reported with mechanisms of APAP toxicity in human cells. In this model, liver injury usually appears after 3-5 h and peaks at 12 h with doses of 200-400 mg/kg APAP, similar to human overdoses, which varied from 150 to 500 mg/kg [23]. In this study, we tested the hepatoprotective potential of aqueous extracts of L. hastata on APAP-induced liver injury in mice. Our results showed that APAP administration (250 mg/kg) led to a significant (p < 0:01) increase in liver ALT levels compared to normal control. APAP also induced a significant increase in MDA (p < 0:05) and TNFα (p < 0:001) levels compared to the normal control. Conversely, it induced a very significant decrease in the activity of CAT, SOD, and GSH compared to the normal control group. These observations would be due to the accumulation of NAPQI which would lead to GSH depletion, thereby promoting the formation of reactive oxygen species [23,24]. The results obtained are similar to those reported by Hu et al. who showed that the administration of APAP at a dose of 250 mg/kg was sufficient to induce hepatotoxicity with hemorrhagic lesions and increased hepatic MDA levels and decreased GSH values [25]. The increase in inflammation is thought to be due to the formation of NAPQI-cysteine protein adducts [26] and MDA-lysine which are presented to cytotoxic T cells by anti-gens. This causes a cascade of reactions that result in the activation of Kupffer cells, allowing the secretion of inflammatory mediators, cytokines, and chemokines that promote hepatocyte apoptosis and activate the innate immune system [1,3]. These results are also in concordance with earlier reports which showed that the hepatoprotective effect of tempol is mediated by an overexpression of antioxidant enzymes in APAP-treated mice [27]. Therefore, activation of these enzymes might suggest induction of the nuclear factor erythroid 2-related factor (Nrf-2) pathway. Furthermore, we also found that stem decoction of L. hastata was able to reverse all these biochemical and histopathological changes observed in APAP toxicity. These effects may be attributed to the active metabolites contained in this extract which can interfere with the inflammatory process and the oxidative stress induced by APAP. These findings are in line with previous studies that demonstrated the protective effect of L. hastata on alcohol-induced damages [18] and the effect of L. reticulata against carbon tetrachloride-induced toxicity [28]. Although we found that the hepatoprotective effects of L. hastata were mediated by inhibition of the oxidative stress and inflammation in the liver, further biochemical markers should be explored for a better understanding of the mechanism of action. In particular, the effect on anti-inflammatory and other proinflammatory cytokines, P450 enzyme expression, Nrf2 activation, apoptosis, and autophagy processes should be investigated. Also, the effect of this plant on liver regeneration was not yet evaluated although it was considered a critical aspect of APAP's pathophysiology [29]. Such studies must be conducted using isolated phytochemicals from L. hastata.
To characterize the constituents of the active extract of L. hastata, an HPLC analysis was performed and results revealed the presence of many compounds including dihydroxycoumarin, scopoletin, and quinine, a natural alkaloid known for its antipyretic, analgesic, and hepatoprotective properties in malaria patients [30]. These results are consistent with previous phytochemical studies which reported the presence of alkaloids and coumarins in Leptadenia hastata [31].
In order to better examine the role of these compounds in the hepatoprotective activity against APAP, their binding affinities with some antioxidant and proinflammatory mediators were analyzed by molecular docking using the iGEMDOCK software. The binding energy, van der Waals interaction, electrostatic energy, and hydrogen bonding were calculated to determine how the ligand binds to the target protein and visu-alize the pharmacological interactions. As shown by Guimarães et al., negative ΔG binding values indicate favorable interactions of the ligand-receptor complex [32]. The docking interactions presented in Table 2 demonstrated that dihydroxycoumarin had the lowest binding energies with COX-2 (-101.3 kcal/mol), CYP 2E1 (-103.45 kcal/mol), and TNFR (-110.86 kcal/mol), while quinine presented the strongest affinity with GSH reductase (-98.18 kcal/mol), TNFα (-68.81 kcal/mol), and iNOS (-119.63 kcal/mol).
COX-2 are key enzymes involved in prostaglandin biosynthesis in inflammatory cells. It was previously showed that the COX-2 active site possesses three important regions including a hydrophobic pocket identified by the presence of Tyr 385, Trp 387, Phe 518, Ala 201, Tyr248, and Leu 352, a second key region associated with three hydrophilic amino acid residues (Arg 120, Glu 524, and Tyr355) located at the entrance of the active site and a side pocket characterized by the presence of His 90, Arg 513, and Val 523 [33]. On contrary to the reported data, our docking results seem to indicate that the COX-2-dihydroxycoumarin complex involves interactions at a position different from the active site. Indeed, as shown in Figure 9

10
BioMed Research International mediators. This metabolite is generated by a family of enzymes called inducible NO synthase or iNOS from the Larginine metabolism. The role of NO in APAP-induced hepatotoxicity has been demonstrated in a previous study. NO levels were found to increase in hepatocytes isolated from APAP-treated rats due to iNOS overexpression [34]. The molecular docking analysis of the iNOS enzyme (PDB ID: 2ORQ) with our ligands showed high binding affinities with quinine (-119.63 kcal/mol). The interaction between quinine and iNOS in the binding cavity involved a mechanism closer to that of silibinin ( Figure 9). These results are similar to those reported by Chen et al. (2005) who demonstrated the inducible effect of chloroquine on human iNOS [35]. TNFα is a cytokine released by activated macrophages (Kupffer cells) in the liver in response to toxic agents like APAP. TNFα has been linked to increased oxidative stress (increased formation of reactive oxygen species and reactive nitrogen species) and is known to recruit and activate other inflammatory cells [2]. According to our results, quinine exhibited the best binding affinity with TNFα. This interaction is favored by the formation of hydrogen bonds with Ala 275, Tyr 284, and NAG 404 residues and hydrophobic bonds with Asn 161, Trp 220, Arg 274, Ala 275, Tyr 276, Tyr 284, and NAG 404 residues. Previous studies showed that some quinoline derivatives with anti-inflammatory properties might form a stable complex with TNFα by docking in the middle pocket of dimers [36]. However, no hydrogen bond was observed in these studies which could be explained by the different nature of the docked molecule and the docking approach. Furthermore, we also found a strong affinity between quinine, dihydroxycoumarin, and the TNFR as shown in Table 2. But this affinity was lower than that observed with silibinin (-123.1 kcal/mol). These findings are consistent with previous reports who identified novel inhibitors for TNFα and TNFα receptors using pharmacophore-based approaches [37]. According to Chen et al. (2017), small molecules that inhibit TNFα or its receptor, are equally capable to block the interactions between both proteins and/or regulate related signaling pathways [38].

Conclusion
This study revealed that L. hastata stem decoction protected mice from APAP-induced hepatotoxicity through antioxidant and anti-inflammatory activities in the liver tissue. The HPLC profile of this extract revealed the presence of quinine (an alkaloid), dihydroxycoumarin, and scopoletin. The in silico analysis demonstrated that quinine had great binding affinities for iNOS, TNFα, and GSH reductase while dihydroxycoumarin fitted perfectly with Cox-2, TNFR, and CYP 2E1. Furthermore, both compounds showed good drug-likeness and pharmacokinetic scores, suggesting their potentials to be highly bioavailable oral drugs. The interactions of these compounds with the target proteins seem to be important factors in providing liver protection and might contribute to valorize the potentialities of this plant against liver diseases. Further investigations with the isolated com-pounds are therefore necessary to have complete information on the biological activity on the liver.

Data Availability
All data are available in the manuscript.

Disclosure
This research was also conducted as part of the academic activities of the University of Ngaoundere.

Conflicts of Interest
The authors declared that there is no conflict of interest.