Characterization of stenocephol from Seriphidium stenocephalum as potent HepG2 cell growth and glycogen phosphorylase inhibitor

Abstract Plant-derived compounds represent an important source for developing innovative drugs. One of the widely distributed plants, especially in Afghanistan and Pakistan, Seriphidium stenocephalum, was investigated in this study to identify bioactive compounds. The plant extract was subjected to silica gel column chromatography, four phenolic acid derivatives were isolated, while stenocephol was obtained by ethyl acetate fraction. Stenocephol was subjected to experimental screening for anti-diabetic and anti-cancer activities, measuring its inhibitory potency against glycogen phosphorylase, and its cytotoxicity against HepG2 cells. Further insights into the mechanism of action of stenocephol were obtained from a computational investigation. Stenocephol showed a dose-dependent manner of inhibition against glycogen phosphorylase and HepG2 cells in the low micromolar range. Notably, coupling in vitro and computational investigation, we identified the natural product stenocephol as a possible anti-diabetic and anti-cancer agent, representing a possible starting point for developing novel therapeutics, enriching the armamentarium against the mentioned diseases. Graphical Abstract


Introduction
Natural products represent a critical source for the drug discovery field. Plants produce a unique class of more than 100,000 low-molecular-weight secondary metabolites; many of these are under consideration for drug development (Siddiqui et al. 2018), having wide applications in the biological and pharmaceutical fields. Notably, medicines from plants have limited side-effects, and many of them are cost-effective. By looking around the ancient world, the crude forms of plant extracts, paste, tincture, and tea have been used as therapeutic agents to treat different diseases (Carabajal et al. 2017;Shedoeva et al. 2019). Anti-cancerous plants containing bioactive secondary metabolites can prevent, reverse, or delay the carcinogenic process, inducing apoptosis, DNA damage, causing G2/M arrest, inhibition of rapid cell division, migration, and invasion of cancer cells (Shin et al. 2018;Siddiqui et al. 2018). Furthermore, many traditional medicinal plants have anti-diabetic properties. More than 350 traditional plants have been used to treat diabetes mellitus (Aggarwal and Shishu 2011;Surya et al. 2014). Extracts of Morinda lucida (Rubiaceae) have shown a hypoglycemic activity in alloxan-induced diabetic rats (Chen et al. 2001;Salehi et al. 2019). Allium cepa (Lilliaceae) shows a significant hypoglycemic effect (Khan et al. 2014). Seriphidium is the largest genus belonging to the Compositae family with a wide distribution in Asia, Europe, North America, and North Africa regions (Shafiq et al. 2020). Plants belonging to the genus Seriphidium have been widely used by local communities as paste, tincture, tea, and nutraceuticals to treat different ailments due to their pharmacological properties. Due to the significant properties of the Seriphidium plants, it has been analyzed qualitatively and quantitatively. Secondary metabolites belonging to several chemical classes have been isolated from different species of the genus Seriphidium (Markham et al. 1978;Deng et al. 2004;Shafiq et al. 2013Shafiq et al. , 2014Shafiq et al. , 2015Ali et al. 2017) ( Figure S1). Several of these compounds have been intensively used in the pharmaceutical, cosmetic, and food industries. The plant Seriphidium stenocephalum (Seriphidium stenocephalum (Kraschen. ex Poljakov) Poljakov; synonym Artemisia stenocephala Krasch. ex Poljakov) (WFO 2022), belongs to this genus, is used to remove worms as purgative. Furthermore, the phytochemical survey of the Seriphidium genus displayed a broad spectrum of antioxidants and anticancer compounds (Sridevi et al. 2003).
Stenocephol (4(4 0 -hydroxy benzyl)-3-methoxybenzen-1,2-diol) was previously isolated, structurally characterized using spectroscopic techniques ( Figure S2). It is a hydroquinone derivative from Seriphidium stenocephalum, containing 4-hydroxy benzyl joined to 3-methoxycatechol by a methylene linkage (Shafiq et al. 2013) (Figure 1). A similar compound as bis-(4-hydroxyphenyl)methane has been reported from an alga Nitophyllum marginata, while seven 4-hydroxy benzyl-substituted derivatives have been identified in Gastrodiaelata genus. Being an attractive molecule and considering the lack of data about the biological potential of stenocephol from Seriphidium stenocephalum, we here report a comprehensive assessment, employing biological evaluation and computational studies, to find further activities of this compound, investigating its potential ( Figure S3).

Results and discussion
Due to the urgency to find novel sources for developing innovative drugs, in this study, we screened stenocephol to assess its potential as anti-cancer and anti-diabetic agents, combining in vitro and computational methods for preliminary characterization of the compound.

Biological investigation
2.1.1. Anticancer activity Stenocephol was evaluated for its potential capability to arrest the tumor cell growth. To this purpose, we used an MTT cytotoxicity assay employing HepG2 human liver cancer cells. Stenocephol stopped the tumor cell growth with an IC 50 value of 2.912 mM ( Figure 1A). One point to be considered that contributes to the anticancer activity is that the presence of hydroxyl moieties at C-4, C-5 and C-12 positions enhances the free radical scavenging power, reducing the cancerous cell growth and cell mutation ratio (Fiuza et al. 2004;Liew et al. 2020).

Anti-diabetic activity
Glycogen phosphorylase is a key enzyme, having a direct and great influence, in controlling the blood glucose level. It is considered as a promising drug target for treating type II diabetes. Glycogen phosphorylase catalyzes the transformation of glycogen to glucose-1-phosphate, thus acting as a key enzyme in the consumption of liver and muscle reserves of glycogen. The in vitro test revealed a significant inhibitory activity of stenocephol against glycogen phosphorylase enzyme (IC 50 8.794 mM) ( Figure 1B). Figure 2A shows the representative docked pose of stenocephol within the EGFR binding site (binding affinity À7.2 kcal/mol; binding affinity of the reference inhibitor (erlotinib) À7.8 kcal/mol). Table S1 summarizes the DG values, types of interaction and residues targeted by stenocephol and erlotinib. Considering the docking studies, stenocephol showed a binding affinity for the active site of EGFR comparable to that found for erlotinib. Stenocephol interacted with Leu764, Met769, and Asp831, while erlotinib interacted with Lys721, and Glu738 by H-bonding interactions. The hydrophobic interactions did not show any unfavorable bumps, confirming the good affinity of the ligand. Stenocephol showed hydrophobic interactions with Lys721, Val702, Ala719, Leu694, and Leu820 ( Figure S4). These findings together with the biological evaluation indicated that stenocephol can serve as an effective lead compound for developing potential anticancer agents.

Molecular docking studies
Molecular docking of stenocephol within the glycogen phosphorylase active site postulated a binding affinity of À8.7 kcal/mol, while that found for the reference compound flavopiridol was À5.8 kcal/mol (see Table S2 for further details). Stenocephol can target Gly612 and Glu382 by H-bonds ( Figure 2B). Flavopiridol showed H-bonding interactions with Glu572, Tyr613, Gly612, Glu382, and Asn284 ( Figure S5).Due to the absence of any unfavorable interactions or stearic bumps, stenocephol could fit more strongly within the protein binding pocket compared to the reference drug flavopiridol (Table S2). Stenocephol showed hydrophobic interactions with Glu382, Phe285, Tyr613, and Ala610, respectively. Docking studies have highlighted the inhibitory potential of stenocephol against glycogen phosphorylase, showing favorable interactions compared with flavopiridol.

Molecular dynamics simulations
To validate the molecular docking results, we investigated the behavior of complexes derived from molecular docking studies (glycogen phosphorylase/stenocephol; EGFR/ stenocephol) by MD simulations in explicit solvent for 100 ns. The resulting MD trajectories for both complexes were examined using RMSD analysis for protein atoms and ligands and RMSF of individual amino acid residue. Overall, the complexes displayed reasonable stability during the simulation, as indicated by the RMSD (Figure S6, panel A-B regarding the MD simulation of stenocephol within glycogen phosphorylase and within EGFR, respectively). The stability of the complexes was also corroborated by examining the RMSF. The systems did not show considerable fluctuation events, excluding some residues at N-and C-terminal regions of the proteins ( Figure S7).
To better comprehend the behavior of stenocephol into the binding sites of glycogen phosphorylase and EGFR, we conducted an inclusive analysis of MD simulations, exploring the established contacts. In general, stenocephol within the glycogen phosphorylase binding site ( Figure S8) maintained the contacts found by docking calculation. Interestingly, we observed a stronger network of hydrophobic interactions with His571, Phe285, Ala610, and Tyr613. The H-bond with Gly612 is still evident, but in addition we noted more favorable polar contacts with Asn282 and with Asp283, sometimes water-mediated.
The analysis conducted on the trajectory of MD simulation for stenocephol within the EGFR binding site is illustrated in Figure S9. Here again, the crucial contacts established by stenocephol within the binding site were maintained. H-bonds with Leu764 and Met769 were still evident. Additional H-bonds, sometimes water-mediated, were detected with Thr830, Asp831, and Phe382. The most relevant hydrophobic contacts were established with Leu694, Val702, Ala719, Lys721, and Leu820, resulting in a more tightly binding within the active site of the enzyme.
Furthermore, we performed an extensive quantum mechanical calculation on stenocephol and the outputs are reported in the Supplementary Material (Figures S10-S13  and Tables S3-S5).
Overall, the computational outcome undeniably validated the experimental approach, indicating that the compound can strongly target the selected proteins.

Conclusion
In this work, we described a combined approach based on in vitro and in silico methods for characterizing the plant-derived product stenocephol. To this end, we performed in vitro experiments employing HepG2 cell lines to determine its possible role as anticancer molecule, while we evaluated the suppression of the production of glycogen phosphorylase products, measuring its inhibition. Gratifyingly, stenocephol showed an IC 50 value of 2.912 mM against HepG2 and can also dose-dependently inhibit the glycogen phosphorylase enzyme with an IC 50 value of 8.794 mM. To gain further insight into the biological activities of stenocephol, we conducted computational studies to investigate the mechanism of action at the molecular level, highlighting the structural determinants that govern its bioactivity. Accordingly, considering the results of the work, stenocephol represents an interesting starting point for developing novel drug candidates against cancer and diabetes.