Lepidium sativum Secondary Metabolites (Essential Oils): In Vitro and In Silico Studies on Human Hepatocellular Carcinoma Cell Lines

Hepatocellular carcinoma (HCC) is the most common primary liver cancer and the greatest cause of cancer-related death in the world. Garden cress (Lepidium sativum) seeds have been proven to possess extraordinary antioxidant, anti-inflammatory, hypothermic, and analgesic properties. In this study, in vitro cytotoxic efficiency evaluation of L. sativum fractions was performed against two hepatocellular carcinoma cell lines (HuH-7 and HEPG-2), and the expression of some apoptotic genes was explored. In addition, the chemical composition of a potent extract of L. sativum was analyzed using gas chromatography coupled with mass spectrometry. Then, molecular docking analysis was implemented to identify the potential targets of the L. sativum components’ most potent extract. Overall, the n-hexane extract was the most potent against the two HCC cell lines. Moreover, these cytotoxicity levels were supported by the significant downregulation of EGFR and BCL2 gene expression levels and the upregulation of SMAD3, BAX, and P53 expression levels in both HuH-7 and HEPG2 cell lines. Regarding L. sativum’s chemical composition, GC–MS analysis of the n-hexane extract led to the identification of thirty compounds, including, mainly, hydrocarbons and terpenoids, as well as other volatile compounds. Furthermore, the binding affinities and interactions of the n-hexane fraction’s major metabolites were predicted against EGFR and BCL2 molecular targets using the molecular docking technique. These findings reveal the potential use of L. Sativum in the management of HCC.


Introduction
Liver cancer is the second highest cause of cancer-related mortality worldwide. Meanwhile, primary liver cancer comprises intrahepatic cholangiocarcinoma (iCCA) and hepatocellular carcinoma (HCC) forms, with HCC being the most predominant [1]. HCC constitutes a global crisis, and its epidemiological data fluctuate from place to place. HCC is estimated to be the sixth most prevalent type of cancer worldwide; in Egypt, it represents the fourth most widespread cancer [2]. The prevalence of HCC is rising dramatically, especially in Africa and Asia. For example, in Egypt, health officials consider HCC the most critical health problem. Remarkably, the incidence of HCC has increased approximately two-fold over the last decade [3]. Chronic HBV (Hepatitis B Virus) infection, autoimmune chloride (IC 50 = 47 µg/mL), and ethyl acetate (IC 50 = 63.8 µg/mL) on HEPG 2 cells ( Figure 1A), while on HuH-7 cells, n-hexane (IC 50 = 59.3 µg/mL), methylene chloride (IC 50 = 59 µg/mL), and ethyl acetate (IC 50 = 63.5µg/mL) were compared to a non-treated control ( Figure 1B) using the normal cell line, THLE 2 ( Figure 1C). Thus, the n-hexane extract was observed to be the most effective. 95 The effects of different concentrations of methylene chloride, butanol, methanol, 96 ethyl acetate, and n-hexane extracts on the cellular proliferation of HEPG2 and HuH-7 cell 97 lines following 48 h of treatment were studied. The cell cytotoxicity results revealed a 98 concentration-dependent cytotoxic effect of n-hexane (IC50 = 45µg/mL), methylene chlo-99 ride (IC50 = 47 µg/mL), and ethyl acetate (IC50 = 63.8µg/mL) on HEPG 2 cells ( Figure 1A), 100 while on HuH-7 cells, n-hexane (IC50 = 59.3µg/mL), methylene chloride (IC50 = 59µg/mL), 101 and ethyl acetate (IC50 = 63.5µg/mL) were compared to a non-treated control ( Figure 1B) 102 using the normal cell line, THLE 2 ( Figure 1C). Thus, the n-hexane extract was observed 103 to be the most effective. Values are expressed as the mean ± SD of 3 separate experiments performed in quadruplicate. Statistical analysis was carried out utilizing analysis of variance (ANOVA) followed by post-hoc test, and Dunnett t-test was used for comparison with the control group. All tests were two-tailed. A p-value < 0.05 was considered significant (* p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001).
Meanwhile, terpenoids, with a concentration of (18.80%), were considered the second most identified class of compounds, including mono and sesqui-terpenoids, with the complete absence of diterpenoids. Five monoterpene compounds were identified from among the characterized terpenoids and represented (13.38%) of the overall mass. Methyleugenol (4.77%) and anisole, p-propenyl-, cis-(4.13%) were identified as the main monoterpenes, while ç-asarone (0.68%) was identified as a minor compound. Furthermore, sesquiterpenes were observed at remarkable concentrations (5.42%), involving only four compounds. Caryophyllene was identified as a major compound, with a concentration of 3.82% among the identified sesquiterpenes, but sesquisabinene isomer was identified as a minor one (0.74%). In addition to hydrocarbons and terpenes, other volatile compounds were identified at minor concentrations (9.71%). These identified compounds include three aromatic compounds, namely 2-methoxy-4-[(1e)-1-propenyl]phenol (1.15%), 9,12,15-octadecatrienoic acid, 2-phenyl-1,3-dioxan-5-yl ester (1.00%), and propanoic acid, 2-(3-acetoxy-4,4,14-trimethylandrost-8-en-17-yl)-(1.19%), and one tocopherol, ç-tocopherol (3.55%). Additionally, two nitrogenated compounds, 1H-purin-6-amine, [(2-fluorophenyl) methyl]-(1.79%) and 10-nethoxy-Nb-alpha-methylcorynantheol (1.03%) were identified (Table 1). To investigate the effects of L. sativum n-hexane extract on HCC inhibition or development, we presumed that L. sativum extract might re-modulate the expression levels of some apoptotic gene markers. Apoptosis is a biological pathway controlled by various signal transduction actions, including the regulation of the P53, BAX, and BCL2 genes [23]. Generally, the expression patterns obtained during this study using qPCR-based profiling of the proapoptotic gene markers (P53 and BAX) showed significant increases in the mRNA transcripts of these genes in both the HuH-7 and HEPG2 cell lines compared to the untreated control (Figure 2A analysis was performed using analysis of variance (ANOVA) followed by post-hoc test, and Dunnett t-test was 197 r comparison with the control group. All tests were two-tailed. A p-value < 0.05 was considered significant. 198 199 The binding mechanisms and affinities of the main metabolites in the n-hexane ex-200 tract with EGFR and BCL2 were predicted using molecular docking. Prior to in silico pre-201 diction, the performance of the AutoDock4.2.6 software with the employed parameters to 202 predict the binding modes of EGFR and BCL2 inhibitors was assessed based on the avail-203 able experimental data. The co-crystalized EGFR and BCL2 ligands-namely AQ4 and 204 DRO, respectively-were re-docked, and the predicted binding modes were compared to 205 A B Figure 2. Gene expression levels after treating hepatocellular carcinoma cells with L. sativum n-hexane extract. (A): Expression of BCL2, BAX, P53, TGF-β, SMAD3, and EGFR in HEPG2 cells. Blue: untreated control, red: hexane extract. (B): Expression of BCL2, BAX, P53, TGF-β, SMAD3, and EGFR in HuH-7 cells. Blue: untreated control, red: hexane extract. Statistical analysis was performed using analysis of variance (ANOVA) followed by post-hoc test, and Dunnett t-test was used for comparison with the control group. All tests were two-tailed. A p-value < 0.05 was considered significant (** p ≤ 0.01, *** p ≤ 0.001, **** p ≤ 0.0001).

Molecular Docking Calculations and Pathway Enrichment Analysis (PEA)
In contrast, the BCL2 gene is considered the most vital antiapoptotic protein, and many reports have confirmed its overexpression level in several human tumors [24]. For the L. sativum n-hexane extract, BCL2 showed a consistent downregulation pattern in the HEPG2 cell line, with a fold change of 0.41 ± 0.01, and downregulation in the HuH-7 cell line by 0.6 ± 0.1 fold (Figure 2A,B).
Moreover, the TGF-β signal transduction pathway is considered an important regulator of hepatocellular carcinoma and also accelerates liver fibrosis [25]. Under the TGF-β signal transduction pathway, the SMAD2 and SMAD3 genes were found to be critical intracellular mediators, even reflecting tumor-inhibitory or tumor-inducing effects [25]. The current results revealed that the expression level of TGF-β was upregulated upon treatment with the n-hexane extract of L. sativum in the HEPG2 cell line, with a fold change of 2.1 ± 0.1, and it was upregulated in the HuH-7 cell line by 1.7 ± 0.1 fold. Meanwhile, SMAD3 gene expression was significantly upregulated after treatment with n-hexane extract in the HEPG2 cell line, with a fold change of 3.4±0.1, and it was also raised in the HuH-7 cell line by 4.2 ± 0.1 fold.
The EGFR pathway is essential in liver regeneration, cirrhosis, and hepatocellular carcinoma. Its overexpression was confirmed to play a pivotal tumor-promoting role in HCC development [26]. The current study found a significant downregulation in EGFR gene expression upon exposure to the n-hexane extract in the HEPG2 cell line, with a fold change of 0.2 ± 0.1, and downregulation in the HuH-7 cell line by 0.4 ± 0.1 folds.

Molecular Docking Calculations and Pathway Enrichment Analysis (PEA)
The binding mechanisms and affinities of the main metabolites in the n-hexane extract with EGFR and BCL2 were predicted using molecular docking. Prior to in silico prediction, the performance of the AutoDock4.2.6 software with the employed parameters to predict the binding modes of EGFR and BCL2 inhibitors was assessed based on the available experimental data. The co-crystalized EGFR and BCL2 ligands-namely AQ4 and DRO, respectively-were re-docked, and the predicted binding modes were compared to the corresponding experimental poses ( Figure 3). Comparing the predicted and experimental binding modes revealed that the AutoDock4.2.6 software accurately reproduced the crystal structure and the observed binding modes of AQ4 and DRO ( Figure 3). The excellent performance of AutoDock4.2.6 in predicting the crystal binding mode of erlotinib (AQ4) with EGFR is in line with the results of our previous study [27].
Using the AutoDock4.2.6 software, n-hexane extract major metabolites were docked against EGFR and BCL2, and their binding affinities and modes were predicted. Predicted docking scores of the investigated metabolites are listed in Table 2, and 2D representations of their binding modes inside the active sites of the EGFR and BCL2 proteins are depicted in Figure 4. Table 2. Predicted docking scores (in kcal/mol) for n-hexane extract major metabolites with active EGFR, inactive EGFR, and BCL2.

No. Compound Name
Docking Score (kcal/mol) 27 10-Nethoxy-Nb-alpha-methylcorynantheol −7.9 −7.3 −5. Using the AutoDock4.2.6 software, n-hexane extract major metabolites were docked 215 against EGFR and BCL2, and their binding affinities and modes were predicted. Predicted 216 docking scores of the investigated metabolites are listed in Table 2, and 2D representations 217 of their binding modes inside the active sites of the EGFR and BCL2 proteins are depicted 218 in Figure 4.  The investigated metabolites showed relatively weaker docking scores towards ac-233 tive EGFR (PDB code: 1M17), with values in the range of −2.4 to −8.5 kcal/mol, compared 234 to the inactive EGFR. The same trend of a lower binding affinity towards active EGFR was 235 also observed for erlotinib (AQ4), with docking scores of −7.5 and −10.6 with active and 236 inactive conformations of EGFR, respectively. Interestingly, among the investigated me-237 tabolites, compounds 10 and 29 exhibited higher binding affinities with active EGFR than 238 erlotinib (docking scores of −8.5, −8.5, and −7.5 kcal/mol for compounds 10, 29, and erlo-239 tinib, respectively). The pi-based interactions and alkyl interactions dominated the bind-240 ing affinity of compounds 10 and 29 with active EGFR ( Figure 5). 241 Considering the predicted binding affinity of the co-crystalized ligand DRO with 242 BCL2 (docking score of −9.9 kcal/mol), the major metabolites from the n-hexane extract 243 showed much lower docking scores, with values in the range of −3.4 to −7.7 kcal/mol.

244
For pathway enrichment analysis (PEA) and Reactome mining, the results were in 245 line with the results obtained regarding the expression levels of some apoptotic and signal 246 transduction gene markers. Interestingly, Reactome representation based on Boolean net-247 work modeling confirmed that the EGFR pathway was among the most enriched path-248 ways by compounds 10 and 29; this pathway is vital in liver regeneration and hepatocel-249 lular carcinoma development. Notably, the programmed cell death pathway was not stim-250 ulated/influenced by compounds 10 and 29, revealing that these compounds had no or a 251 Compared to the binding affinity of erlotinib (AQ4) with inactive EGFR (−10.6 kcal/mol), compounds 10, 29, and 30 demonstrated higher docking scores, with values of −12.3, −12.3, and −11.7 kcal/mole, respectively. In contrast, the rest of the compounds exhibited relatively weak binding affinities, with values in the range of −5.2 to −10.3 kcal/mol. Inspecting the predicted binding modes revealed that the high potency of inactive EGFR inhibitors can be attributed to their ability to participate in pi-based and alkyl interactions, including pi-alkyl, pi-anion, and pi-sulfur interactions (Figure 4).
The investigated metabolites showed relatively weaker docking scores towards active EGFR (PDB code: 1M17), with values in the range of −2.4 to −8.5 kcal/mol, compared to the inactive EGFR. The same trend of a lower binding affinity towards active EGFR was also observed for erlotinib (AQ4), with docking scores of −7.5 and −10.6 with active and inactive conformations of EGFR, respectively. Interestingly, among the investigated metabolites, compounds 10 and 29 exhibited higher binding affinities with active EGFR than erlotinib (docking scores of −8.5, −8.5, and −7.5 kcal/mol for compounds 10, 29, and erlotinib, respectively). The pi-based interactions and alkyl interactions dominated the binding affinity of compounds 10 and 29 with active EGFR ( Figure 5).

259
In Egypt, liver cancer is one of the most prevalent types of cancer among both men 260 and women, according to the National Cancer Registry Program (NCRP) of the National 261 Cancer Institute (NCRP, 2018) [4]. In addition, hepatitis C is a major public health prob-262 lem, with the highest global prevalence of hepatitis C virus (HCV); the most common 263 genotype in Egypt is genotype 4 [28]. 264 Chemotherapy-associated side effects due to its non-selectivity and the resistance of 265 cancer cells remain a major obstacle in cancer treatment. Natural therapies, such as plant-266 derived compounds, are a promising addition to cancer therapy. Solutions to controlling 267 the initiation and progression of cancer are of great value [29]. The current research aimed 268 to evaluate the potential use of garden cress in inhibiting liver cancer at the in vitro level 269 and to explore its possible mechanism of action. 270 Previous research proved that L. Sativum extracts display a cytotoxic effect against 271 different types of cancer, including leukemia [19] and breast cancer [20]. In agreement 272 with this, the current study showed a concentration-dependent cytotoxic effect of meth-273 ylene chloride (47 ug/mL), ethyl acetate (63.8 ug/mL), and n-hexane (45 ug/mL) extracts 274 on HEPG2 cells, while on HuH-7 cells, methylene chloride (59 ug/mL), ethyl acetate (63.5 275 Considering the predicted binding affinity of the co-crystalized ligand DRO with BCL2 (docking score of −9.9 kcal/mol), the major metabolites from the n-hexane extract showed much lower docking scores, with values in the range of −3.4 to −7.7 kcal/mol.
For pathway enrichment analysis (PEA) and Reactome mining, the results were in line with the results obtained regarding the expression levels of some apoptotic and signal transduction gene markers. Interestingly, Reactome representation based on Boolean network modeling confirmed that the EGFR pathway was among the most enriched pathways by compounds 10 and 29; this pathway is vital in liver regeneration and hepatocellular carcinoma development. Notably, the programmed cell death pathway was not stimulated/influenced by compounds 10 and 29, revealing that these compounds had no or a negligible direct effect on inducing programmed cell death in cancerous cells ( Figures S1  and S2).

Discussion
In Egypt, liver cancer is one of the most prevalent types of cancer among both men and women, according to the National Cancer Registry Program (NCRP) of the National Cancer Institute (NCRP, 2018) [4]. In addition, hepatitis C is a major public health problem, with the highest global prevalence of hepatitis C virus (HCV); the most common genotype in Egypt is genotype 4 [28].
Chemotherapy-associated side effects due to its non-selectivity and the resistance of cancer cells remain a major obstacle in cancer treatment. Natural therapies, such as plant-derived compounds, are a promising addition to cancer therapy. Solutions to controlling the initiation and progression of cancer are of great value [29]. The current research aimed to evaluate the potential use of garden cress in inhibiting liver cancer at the in vitro level and to explore its possible mechanism of action.
Previous research proved that L. sativum extracts display a cytotoxic effect against different types of cancer, including leukemia [19] and breast cancer [20]. In agreement with this, the current study showed a concentration-dependent cytotoxic effect of methylene chloride (47 ug/mL), ethyl acetate (63.8 ug/mL), and n-hexane (45 ug/mL) extracts on HEPG2 cells, while on HuH-7 cells, methylene chloride (59 ug/mL), ethyl acetate (63.5 ug/mL), and n-hexane (59.3 ug/mL) from L. sativum extracts were compared to a non-treated control.
Additionally  (2015) found that the aerial parts of the L. sativum plant exert anticancer activity against the K562 leukemia cell line [19]. Furthermore, the aqueous extract of L. sativum seeds induced apoptosis and necrosis in breast cancer cells (MCF-7), causing a significant time-and dose-dependent reduction in cell viability [20]. Garden cress leaf extract showed a strong antiproliferative effect against CAL-27 cells, mediated through the apoptosis process [30].
L. sativum n-hexane extract contains tocopherol, linoleic acid, oleic acid, linolenic acid, methyleugenol, and anisole, according to GC-MS analysis. Garden cress is high in antioxidants such as vitamins A, B, C, E, isotiosinat, omega-3 fatty acids such as alpha-linolenic acid, and glucosinolates, as well as glucosinolates, which might reinforce its anticancer effects due to their antioxidant activity [19]. From another point of view, the anticancer effect might be attributed to the isothiocyanates, specifically benzyl isothiocyanate. Apoptosis was induced in MCF-7 upon treatment with 25% and 50% extract, while necrosis occurred after exposure to elevated extract concentrations (75%) [20]. In agreement with previous research, the present result indicates the presence of different antioxidant compounds in L. sativum extracts, which contribute to its antitumor activity.
On the other hand, the EGF receptor pathway plays a crucial role in different cancer types. In particular, the EGFR pathway is critical in liver regeneration, cirrhosis, and hepatocellular carcinoma (HCC) [33]. In HCC, EGFR is overexpressed in liver macrophages, where it has a tumor-promoting role [34]. Komposch and Sibilia (2015) found that the EGFR pathway in liver cells prevents the overproduction of interleukin 1 beta (IL-1β) and compensatory proliferation, suggesting its antitumorigenic role [35]. EGFR upregulation occurs in 68% of HCC cases, correlating with tumor aggressiveness, metastasis, and poor prognosis. Lanaya et al. (2014) discovered that deleting EGFR in liver cells increased the frequency of HCC, but deleting it in Kupffer cells and/or liver macrophages dramatically decreased the development of HCC in mice. They also discovered that animals without EGFR in macrophages had a reduced risk of hepatocarcinogenesis, but mice lacking EGFR in hepatocytes had a higher risk of HCC due to greater liver cell damage. EGFR is necessary for liver macrophages to produce interleukin-6 in response to interleukin-1 stimulation, which causes liver cell proliferation and HCC. The loss of EGFR in hepatocytes causes HCC, and EGFR-positive liver macrophages are linked to poor overall survival in HCC patients [34]. Since EGFR is a primary driver of tumorigenesis and is recognized as a resistance biomarker, and EGFR signaling is involved in the regulation of different metabolic pathways that are crucial for cancer cell proliferation [36], the current results regarding its downregulation by Lepidium sativum extracts might be one of the mechanisms involved in its cytotoxic activity.
Remarkably, the TGF-β signal transduction pathway is regarded as a central regulator of liver cancer [25] due to its tumor-suppressive properties and/or its pro-metastatic effects [37]. In clinical practice, increased TGF-β in the early stages is associated with a promising prognosis, but in progressed tumors, it is associated with increased tumor invasiveness and dedifferentiation, suggesting that TGF-β initially restrains hepatocellular carcinoma via its tumor-suppressive effects but may later aggravate the malignancy due to its pro-oncogenic functions [38]. Therefore, in contrast to EGFR, our obtained results revealed the significant upregulation of the TGFβ and SMAD3 genes in both the HUH7 and HEPG2 cell lines compared to the untreated control. These results might explain the dual role of TGF-β, acting either as a suppressor during early stages or contributing to tumor progression during late stages, when cells evade its cytostatic effects. In other words, TGF-β's protumorigenic effects include epithelial-mesenchymal transition (EMT) and alterations in the plasticity of tumor cells [39,40].
Furthermore, TGF-β is a pleiotropic cytokine that communicates with intracellular SMAD proteins and membrane receptors. Therefore, in early tumorigenesis, TGF-β has a tumor suppressor effect. On the other hand, it has a protumorigenic role in the late stages, promoting invasiveness and metastasis [41,42]. Massagué (1998) found that the expression of the central TGF-β signaling transducer, SMAD3, reduced the susceptibility to HCC in a chemically induced model murine. SMAD3's protective effect involves apoptosis induction by repressing BCL2 transcription. The proapoptotic effect of SMAD3 involves TGF-β signaling and activation, which selectively take place in liver cancer cells. Thus, SMAD3 aids the tumor suppression effect of TGF-β by serving as a mediator of TGF-β-induced apoptosis [43].
Moreover, SMAD3 has excellent potential as a gene therapeutic agent for liver cancer treatment. TGF-β/SMAD3 signaling plays a direct role in controlling the cellular level of BCL2, which is crucial for TGF-β-mediated apoptosis in the liver. SMAD3 downregulates the expression of BCL2, inducing apoptosis through TGF-β [44]. In agreement with this, the current work explains the observed cytotoxic effect of L. sativum extracts via the upregulation observed in the TGF-β and SMAD3 genes.
On the other hand, P53 is a core gene that regulates the cell cycle and suppresses cancer progression [45]. It has been reported that the P53 tumor suppressor gene exhibits an upregulation pattern, which promotes the cell apoptosis process through interaction with the BCL2 gene product [46]. The BCL2/BAX ratio proves the ability of L. sativum in apoptosis induction. In agreement with previous reports, our results demonstrated that the hexane extract of L. sativum significantly downregulated the mRNA expression patterns of the EGFR and BCL2 genes, which explains its proliferation effect. On the other hand, L. sativum extract significantly upregulated TGFβ, SMAD3, BAX, and P53 expression levels in both the HuH-7 and HEPG2 cell lines compared to the untreated control. Consequently, L. sativum inhibited the proliferation and growth of HCC cell lines and increased the apoptosis of these cells.

Method of Extraction
Seeds of L. sativum were purchased from a local herbal supplier (Ragab El-Attar Herbal Store, Cairo, Egypt). A copy of the purchased seeds used in the current study was deposited in the Botany Department's herbarium at the Faculty of Science, Cairo University, under voucher specimen number (10001). The L. sativum species were taxonomically authenticated by Prof. Dr. Reem Samir (Professor of Botany, Faculty of Science, Cairo University, Cairo, Egypt). The seeds were manually screened, and then uniformed ones were ground using an electrical grinder. At room temperature, the fine powder (1 kg) of L. sativum seeds was extracted with aqueous methanol (80%). Next, the seeds of L. sativum were extracted, and extracts were evaporated in vacuo at 45 • C, followed by the successive liquid-liquid extraction of the crude extract with n-hexane (33 g), CH 2 Cl 2 (43 g), EtOAc (15 g), and BuOH (13 g).

Cell Line Culture
HEPG2 and HuH-7 hepatocellular carcinoma cells were obtained from the American Type Culture Collection (ATCC), USA. The cells were maintained by serial subculturing into 75 cm 3 flasks in Dulbecco's modified eagle medium (DMEM; Invitrogen, Carisbad, CA, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 1mM sodium pyruvate, 2 mM L-glutamine (Invitrogen), and 1% penicillin G/streptomycin (Invitrogen). In a water-jacketed incubator, the cells were maintained in monolayer culture at 37 • C under a humidified atmosphere of 5% CO 2 (Revco, RCO 3000 TVBB, USA). The cell lines were repeatedly subcultured in order to keep them in the exponential growth phase. Sterile conditions were achieved under an equipped laminar flow (Microflow Laminar Flow Cabinet, Hampshire SP105AA, UK).

MTT Assay
The vitality of the cells was evaluated by performing the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium) bromide assay according to our protocol [47]. Media were aspirated, and fresh medium (without serum) was added to the cells with various concentrations (0, 3.125, 6.25, 12.5, 25, 50, and 100 µg/mL) of plant extracts and incubated for 48 hours. Cell-free controls (as a blank) were used in each plate that contained MTT plus compounds in culture media to correct any reduction of MTT in the absence of cellular mitochondria [48].

Gas Chromatography
Analysis of L. sativum n-hexane extract's chemical composition was achieved using a mass spectrometer (Trace GC-TSQ Thermo Scientific, Austin, TX, USA) connected to a direct capillary column, the TG-5MS (30 m × 0.25 mm × 0.25 µm film thickness). The temperature of the column was initially 50 • C, followed by an increase by 5 • C/min to 250 • C, held for 2 min; followed by an increase to reach 300 • C by 30 • C/min, held for 2 min. The instrument conditions, sample preparation, and sample injection procedures were derived from Abd El-Kareem et al., 2016 [49]. The components' retention time (RT) and mass spectra (MS) were compared to those in the WILEY 09 and NIST 14 mass spectral databases [49]. The ∆∆ Ct technique [50] was used to perform quantitative data analysis. The expression levels were adjusted to -Actin and presented as a percentage.

Molecular Docking and Pathway Enrichment Analysis (PEA)
Prediction of the binding affinities and modes of the major metabolites with EGFR and BCL2 was performed using AutoDock4.2.6 software [50]. Both active and inactive conformations of EGFR were considered in the current study. The resolved crystal structures of active EGFR, inactive EGFR, and BLC2 (PDB codes: 1M17 [51], 4HJO [52], and 2W3L [53], respectively) were downloaded and prepared for docking calculations. Protein preparation included: (i) cleaning of water molecules, ions, and ligands, and (ii) investigation of the protonation state and addition of missing hydrogen atoms using the H++ server [54]. The docking parameters were kept as the default, with an energy evaluation maximum number (eval) of 25,000,000 and a genetic algorithm (GA) run of 250. A docking grid size of 60 Å × 60 Å × 60 Å with a spacing value of 0.375 Å was employed and centered at the active site of the studied proteins. For the investigated metabolites, SZYBKI software was used to minimize the geometrical structures of the investigated metabolites with the MMFF94s force field [55,56]. The Gasteiger method was utilized to calculate the partial atomic charges of the investigated metabolites [57]. Furthermore, to explore all potential target-function interrelations based on biological network mining for the 20 most stimulated genes by compounds 10 and 29, pathway enrichment analysis (PEA) was performed using Cytoscape 3.8.2 [58]; finally, the ReactomeFIViz online tool was used for the modeling and illustration of all target interactions [59,60].

Statistical Analysis
Statistical analysis was carried out using IBM SPSS ® Statistical version 23 (IBM ® Corp., Armonk, NY, USA). Data were expressed as mean and standard deviation, and comparison between different groups was performed using analysis of variance (ANOVA) followed by post-hoc test. Dunnett t-test was used for comparison with the control group. All tests were two-tailed. A p-value < 0.05 was considered significant.

Conclusions
L. sativum is a promising medicinal plant that requires extensive research in order to reveal its beneficial therapeutic uses and explore its underlying mechanism(s) of action.
The results of the current study reveal that the n-hexane extract of L. sativum showed clear in vitro cytotoxic efficiency against two hepatocellular carcinoma cell lines. This cytotoxic efficiency was further investigated by identifying the expression levels of some apoptotic genes; we observed the significant downregulation of the expression of the EGFR and BCL2 genes and upregulation of the SMAD3, BAX, and P53 genes. This might explain its mechanism of action. Additionally, n-hexane extract fractionation demonstrated that hydrocarbons, terpenoids, and volatile compounds were the most predominant. Finally, our molecular docking results against EGFR and BCL2 molecular targets support the potency of L. sativum's major compounds against hepatocellular carcinoma. Ultimately, our findings highlight the potential use of L. sativum against HCC.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/plants10091863/s1, Figure S1: A Reactome map of the top enriched pathways affected by the top 20 gene targets in response to compound 10. The yel-low color code indicates the over-representation of that pathway in the input dataset. Light grey signifies pathways that are not significantly over-represented. Figure S2: A Reactome map of the top enriched pathways affected by the top 20 gene targets in response to compound 29. The yel-low color code indicates the overrepresentation of that pathway in the input dataset. Light grey signifies pathways that are not significantly over-represented.