CASTp quantitatively analyzes the topographical features of proteins and measures the area and volume of each pocket and cavity. It also identifies the amino acids crucial for docking studies. The target proteins were prepared for docking by removing all water molecules and non-protein residues, followed by structure optimization and energy minimization using Chimera21,24. Using AutoDock, all missing atoms in the target proteins were repaired. Subsequently, only polar hydrogens and Kollman charges were added, and the target proteins were converted into PDBQT format for docking. Docking was performed using AutoDock Vina, as described by Krishnaswamy et al.25. The grid box dimensions were set to 30 Å × 30 Å × 30 Å, which was found to be optimal for the default exhaustiveness value of 8. The ligand binding site was positioned at the center of the grid box. The spatial dimensions (XYZ axes) and grid box size were specified in a configuration file. Using AutoDock vina version 1.1.2's command line interface, docking was accomplished. The results were limited to nine binding modes. The generated log file included a list of binding modes in increasing order of binding energy. BIOVIA Discovery Studio Visualizer 2025 was used to visualize the binding modes and all non-bonded interactions.
Our primary goal was to assess the effects of biologically derived compounds on cell viability and differentiation. In Fig. 1A,B, light gray bars represent viable cell counts, while dark gray bars indicate non-viable cells for THP-1 and U-937 cell lines, respectively shown as the proportion of total cell population. The results show that the percentage of viable cells remained approximately 75% or higher across all experimental conditions for both cell lines, while non-viable cells made up less than or around 25% of the population. Furthermore, cells treated with the differentiation inducer and bioactive compounds showed no significant reduction in viability compared to control cells. These findings show that neither the differentiation inducer nor the bioactive compounds affected cell viability, and the cells stayed metabolically active during the experiment. This is consistent with our previous research, confirming that bioactive compounds derived from biological waste do not adversely impact cell viability, even after 24 h of exposure. The effects of PMA, a differentiation inducer, were also observed by Kuno et al. (2020), who found that treatment with PMA at concentrations up to 200 nM had no significant impact on cell proliferation or viability. Chanput et al. states that around 100 ng/mL concentration is able to induce differentiation. Browne et al. (2022) reported that concentrations up to 1000 ng/mL maintained cell viability above 80% after 72 h of incubation. However, increasing the concentration of bioactive compounds to between 50 and 150 µM was found to negatively affect cell proliferation and viability in cancer cell line models. Ultimately, these bioactive compounds have significant implications for research in two distinct ways: while low concentrations are safe for cellular viability, higher doses may be utilized for therapeutic purposes.
In our study, the effects of the differentiation inducer and bioactive compounds were assessed using confocal microscopy. Morphological changes in U-937 and THP-1 cells were captured after 96 h of treatment [72 h with PMA followed by 24 h with bioactive compounds]. Our results show that treatment with PMA altered cell surface morphology, changing from a spherical shape to a pseudopodia-like structure compared to untreated cells. Cells treated with chlorogenic acid and α-tomatine during the final 24 h exhibited a more pronounced amoeboid morphology in both cell lines, whereas pseudopodia-like protrusions were less prominent in cells treated with oleuropein and tyrosol (Figs. 2 and 3). The observed differentiation patterns and morphological changes are believed to result primarily from the action of the differentiation inducer rather than from the presence or absence of bioactive compounds. PMA, a known differentiation inducer, activates protein kinase C, which subsequently promotes NF-κB expression and stimulates the expression of proteins associated with macrophage maturation. It also inhibits cell proliferation and induces cellular differentiation. Additionally, PMA-treated cells exhibit an increased cytoplasm-to-nucleus (C/N) ratio, along with changes in cell surface area, cytoplasmic protrusions, and the formation of pseudopodia- and filopodia-like structures, which enhance cell adhesion, as confirmed by previous studies.
The objective of this experiment was to determine whether PMA treatment and the resulting alterations in cell morphology had any negative effects on cell membrane integrity. To assess this, FM4-64, a water-soluble dye that specifically binds to the cell membrane and Hoechst 33342, which stains the nucleus was used. Imaging results confirmed that U-937 and THP-1 cells treated with bioactive compounds showed no signs of cellular damage (Figs. 2 and 3). Notably, nuclear and membrane structural integrity was preserved under all experimental conditions. In our previous study, we also found that PMA and selected bioactive compounds had no adverse effects on cell membrane integrity. The observed cell integrity implies that these selective bioactive compounds are cyto-compatible without any cytolysis and apoptotic effects.
In protein expression studies, it is crucial to identify internal reference genes whose expression remains relatively stable under various experimental conditions. Housekeeping genes are essential for maintaining the structural and functional integrity of cells and are generally expected to exhibit consistent expression in both normal and altered cellular states. Commonly used reference proteins include tubulin, GAPDH, and β-actin.
Housekeeping genes are generally defined as those that are stably expressed across different tissue types, developmental stages, cell cycle phases, and in response to external stimuli. They are considered essential for basic cellular functions, regardless of their specific roles in particular tissues or organisms. However, some studies have shown that the expression of β-actin can increase during induced differentiation processes, making it an unreliable loading control under such conditions. Furthermore, β-actin expression has been observed to fluctuate in various conditions, including certain diseases, hormonal imbalances, infections caused by viruses and bacteria, and during neuronal differentiation. The ambiguous expression levels of β-actin are thought to reflect the cellular and morphological changes that occur under these diverse conditions. Due to its dynamic expression and plasticity, β-actin has been reported to be unsuitable as a reliable reference or internal loading control in experimental setups. In contrast, GAPDH expression has been found to remain stable under differentiation conditions, as reported by Murphy and Polak. Additionally, GAPDH has been shown to maintain consistent expression levels across various experimental conditions, further supporting its use as a more stable reference gene.
In our current study, we selected β-actin and GAPDH as reference proteins and examined their expression in both non-differentiated and differentiated cells across all experimental conditions. As part of the experimental design, U-937 and THP-1 cells were treated with PMA for 72 h to induce differentiation, followed by 24 h of incubation with selected bioactive compounds: chlorogenic acid, tomatine, oleuropein, or tyrosol. Across all conditions, β-actin expression varied between differentiated and non-differentiated cells. Specifically, non-differentiated cells showed lower levels of β-actin expression, whereas differentiated cells demonstrated higher expression levels (Figs. 4A and 5A). This variability in β-actin expression is likely due to cytoskeletal remodelling that occurs during cellular differentiation to support structural and functional demands. GAPDH expression remained largely stable in both monocytic cell lines (Figs. 4B and 5B). Based on these observations and corroborating previous reports, we conclude that GAPDH is a more appropriate reference gene for studies of cell differentiation.
Upon activation, macrophages are known to produce inflammatory cytokines such as IL-6, TNF-α, and IL-12. Our research focused on evaluating the impact of selected bioactive compounds on inflammatory and anti-inflammatory responses based on previously described methodologies. IL-4, an anti-inflammatory cytokine, showed upregulated expression in response to tomatine and chlorogenic acid treatment compared to PMA-treated controls predominatly in U-937 cells (Fig. 6). In U-937 cells, IL-4 expression was markedly elevated following treatment with chlorogenic acid and tomatine, while in THP-1 cells, a partial increase in IL-4 expression was observed under the same treatments (Figs. 6A,B). The elevation of IL-4 expression is known to suppress pro-inflammatory cytokines such as TNF-α and IL-12 through both STAT6-dependent and STAT6-independent mechanisms. Consistent with our findings, a study by Zhou et al. (1994) demonstrated that IL-4 expression inhibits the transcription of inflammatory cytokines, including IL-1α, IL-1β, IL-8, and TNF-α. Furthermore, exogenous IL-4 has been shown to resolve pro-inflammatory states in neutrophils, promoting a shift toward an anti-inflammatory phenotype in macrophages and thereby facilitating the clearance of apoptotic neutrophils. Additionally, the localized and controlled application of IL-4 within the periodontium has been shown to skew macrophage polarization toward the M2 phenotype, consequently reducing inflammation-mediated bone degradation and facilitating bone regeneration.
In our previous study, we found that the four selected bioactive compounds possess anti-MDA activity. The formation of malondialdehyde (MDA) adducts is indicative of ROS generation and lipid peroxidation processes closely associated with the activity of key enzymes such as LOX-5 and MPO. LOX-5 contributes to the oxidation of polyunsaturated fatty acids, while MPO facilitates the formation of HOCl, thereby promoting oxidative stress and inflammatory responses in monocytes.
Our experimental findings revealed that MPO expression was significantly downregulated in the presence of chlorogenic acid, oleuropein, and tyrosol, whereas upregulated in presence of tomatine (Fig. 7A). Concurrently, MPO enzyme activity elevated after the addition of PMA relative to untreated U-937 cells. Furthermore, among the four bioactive compounds, chlorogenic acid and tyrosol induced a decrease in the MPO activity. MPO activity enhanced in the presence of the tomatine correlating their expression at the protein level. The MPO activity declined by 0.82- and 0.09-fold upon incubation with chlorogenic acid and tyrosol respectively, in the comparison with PMA treated cells (Fig. 7B). Similarly, LOX-5 expression was suppressed by oleuropein and tyrosol, with no substantial changes observed under the influence of tomatine or chlorogenic acid (Fig. 8A). As seen in Fig. 8B, treatment with PMA led to a significant increase in LOX-5 enzyme activity compared to undifferentiated control THP-1 cells. However, treatment with the bioactive compounds resulted in a significant reduction in LOX-5 activity. Notably, tyrosol treatment showed the most pronounced effect among the compounds tested. Specifically, LOX-5 activity was reduced by approximately 0.42-, 0.68-, 0.79-, and 0.47-fold following treatment with chlorogenic acid, oleuropein, tyrosol, and tomatine, respectively, when compared to PMA-treated differentiated THP-1 cells. LOX-5 activity was also measured in a biological replicate, and the data are presented in Supplementary Data 2.
MPO not only directly contributes to oxidative stress and lipid peroxidation but also indirectly promotes the activation of LOX-5. The breakdown of arachidonic acid catalyzed by LOX-5 leads to the production of pro-inflammatory mediators such as leukotrienes. Moreover, upregulation of lipoxygenases has been linked to the progression of chronic diseases, including asthma, atherosclerosis, rheumatoid arthritis, and cancer. Elevated levels of MPO protein have also been implicated in the development and progression of various inflammatory chronic diseases. Altogether, targeting LOX-5 and MPO expression with bioactive compounds could represent a promising therapeutic strategy for treating chronic and inflammatory diseases. Expression of TNF-α was also measured, and a significant difference was observed in the bioactive-treated samples. Although the expected molecular weight is approximately 26 kDa, a band was observed at around 60 kDa (Supplementary Data 3). The complete uncropped images of the western blots are available in Supplementary Data 4 and 5.
Overall, our study highlights the beneficial potential of these bioactive compounds in managing inflammatory conditions and modulating ROS generation. Among the compounds tested, chlorogenic acid, oleuropein and tyrosol exhibited strong antioxidant and ROS scavenging properties, potentially by regulating the expression of key enzymes involved in oxidative stress and inflammatory pathways.
Molecular docking was done to estimate the free energy of binding between target human proteins (IL-4, LOX-5, MPO, and TNF-α) and bioactive compounds, as well as with their respective native ligands. Since the tested bioactive compounds in this study were of plant origin, the target proteins were also docked with curcumin, a well-known natural, plant-derived anti-inflammatory compound. The binding energies of the best-docked ligands compared to their native ligands, along with their binding interactions with the active site residues of the target proteins are presented in Table 1. Additionally, the 3D and 2D structures of the best-docked complexes of target proteins with bioactive compounds are presented in Figs. 9, 10, 11 and 12. TNF-α docking results revealed that Tomatine (TOM) was the only bioactive compound showing slightly lower binding energy (- 6.67 kcal/mol) than the native ligand (- 6.49 kcal/mol) but higher binding energy than the control ligand, curcumin (- 5.48 kcal/mol) (Table 1). Tomatine (TOM) formed one hydrogen bond with Tyr151 and four hydrophobic contacts with Tyr119 (× 3) and Leu57 residues of human TNF-α (hTNF-α) (Fig. 9). Interestingly, spotted residues were notably identical to the co-crystallised native ligand which formed eight hydrophobic contacts with amino acids Tyr59, Tyr119, Tyr59, Tyr59, Leu57, Ile155, Leu57, and Ile155 of hTNF-α, but did not form hydrogen bonding (Fig. 9). On the other hand, curcumin formed two hydrogen bonds with Tyr59 and Ser60, and one hydrophobic interaction with Tyr151 (Table 1). It has been reported that Tyr119 plays a key role in the formation of π-π stacking with the native ligand. This interaction helps anchor the inhibitor to hTNF-α. On the other hand, Ley57 is important for the formation of the hydrophobic pocket in the hTNF-α stabilization. Previous studies also showed that plant-derived natural compounds exhibit strong binding affinity, predominantly interacting with Leu57 and Tyr119, resulting in complete burial within the binding pocket without affecting the conformational structure of hTNF-α.
Figure 10 shows the 3D and 2D structures of the best-docked complexes of LOX-5 with bioactive compounds. As evident from Table 1, three bioactive compounds -- chlorogenic acid, oleuropein, and tomatine showed high binding affinity towards LOX-5. The binding affinity of the ligands was in the order: tomatine (- 9.45 kcal/mol) > oleuropein (- 8.01 kcal/mol) > curcumin (- 7.97 kcal/mol) > chlorogenic acid (- 7.62 kcal/mol) > native ligand (arachidonic acid; ACD) (- 5.69 kcal/mol). As presented, ACD formed three hydrogen bonds with amino acids Gln557, Tyr558, and Val604 and seven hydrophobic contacts with Val671, Phe555 (× 3), and Phe610 (× 3). Tomatine formed three hydrogen bonds with Asn180, Gln413, and Ala603, along with nine hydrophobic interactions with Phe555 (× 3), Leu368 (× 2), Ala410, Val671, and Phe610 (× 2) (Fig. 10). Oleuropein formed nine hydrogen bonds with Phe555, Gln557, Ser608 (× 2), Val604, Ser547, Ala546, Glu614, and Ala603. It also formed six hydrophobic contacts with Ala551, Val671, Phe610, and Phe555 (× 2). Chlorogenic acid formed four hydrogen bonds with Gln557, Ser608, Val604, and Gln549. However, it formed only two hydrophobic contacts with Phe555 and Phe610 (Fig. 10). Many previous studies reported that Asn554 and Tyr558 play key role in the inhibition of LOX-5 protein. Although the tested bioactive compounds in this study did not specifically interact with Tyr558, they formed multiple hydrophobic contacts with the neighbouring amino acids Phe555 and Gln557. These interactions suggest that the best-docked compounds may occupy a similar spatial active site pocket and potentially inhibit LOX-5 activity. Therefore, these top-docked compounds might serve as potential lead compounds to inhibit LOX-5 and its associated inflammation.
The best-docked complexes of human MPO (PDB ID: 1DNU) with bioactive compounds are depicted in Fig. 11. Docking results showed that only oleuropein (- 8.66 kcal/mol) and chlorogenic acid (- 8.69 kcal/mol) exhibited slightly lower binding energies than the native ligand, HEME (- 8.4 kcal/mol). HEME formed two hydrogen bonds with His95 and Arg239 and eight hydrophobic contacts with amino acids Arg424, Glu102 (× 3), Phe147, Ala104 (× 2), and Arg239 (Table 1). Chlorogenic acid formed five hydrogen bonds with Phe147, Arg424 (× 2), Glu102, and Phe146, but only one hydrophobic contact with Arg239. Oleuropein also formed one hydrophobic contact with Arg239 and three hydrogen bonds with His95, Arg239, and Pro220 (Fig. 11). Arg239 and Phe99 are key amino acids located in the binding cavity of MPO. The observed binding affinities of chlorogenic acid and oleuropein toward MPO can be attributed to their interaction with Arg239, suggesting their potential involvement in the inhibition of MPO activity. The best-docked complexes of IL-4 with bioactive compounds are shown in Fig. 12. Oleuropein (- 6.34 kcal/mol) and chlorogenic acid (- 6.43 kcal/mol) showed slightly lower binding energy values compared to the control ligand, curcumin (- 6.28 kcal/mol) indicating moderate binding affinity towards IL-4. Chlorogenic acid formed one hydrogen bond with Asn19 and two hydrophobic contacts with Phe86 and Lys16 (Fig. 12). Oleuropein formed seven hydrogen bonds with Arg89, Gln82, Glu23, Ser20, Glu13, Thr17, and Gln82, but only two hydrophobic contacts with Phe86 and Lys16. Curcumin formed five hydrogen bonds and five hydrophobic bonds with the active site residues of IL-4 (Table 1).