Profiling of chemical constituents from AC
A total of 43 chemical constituents were identified by GC-MS (Figure 2), which were profiled compound name, PubChem ID, retention time (RT), area, and taxonomic classification (Table 1). The identified all 43 compounds were confirmed by Lipinski’s rule (Molecular Weight ≤ 500 g/mol; Moriguchi octanol-water partition coefficient≤ 4.15; Number of Nitrogen or Oxygen ≤ 10; Number of NH or OH ≤ 5), additionally, TPSA value (< 140 Å2) (Table 2). Thus, we considered all 43 chemical constituents as drug-like compounds (DLCs).
Identifying of overlapping targets between SEA and STP databases
The targets related to DLCs were retrieved by SEA and STP databases, suggesting that targets recognized from SEA (423), and STP (444), respectively (Supplementary Table S1). The Venn diagram exhibited the overlapping targets (77) as significant targets of DLCs (Supplementary Table S1) (Figure 3A).
Recognition of obesity-related targets and final overlapping targets of AC on obesity
A total of 3,028 targets responded to occurrence and development of obesity were identified by DisGeNET and OMIM databases. Then, the Venn diagram showed that a total of 40 targets were overlapped between obesity-associated targets (3,028) and overlapped 77 targets (Supplementary Table S2) (Figure 3B).
The protein-protein-interaction (PPI) networks and topological analysis
The PPI networks were conducted to identify the uppermost target via RPackage, indicating that IL6 was the highest degree value (DV) among 40 targets. The 9 (ADH1B, CA3, SLC22A2, HSD11B2, OXER1, GSTK1, ENPP2, PAM, and SLC22A6) out of final 40 targets were removed due to no correlations each other, consisting of 31 nodes and 74 edges (Figure 4). In topological analysis, we performed analysis of degree centrality (DC) and betweenness centrality (BC) to identify an important target. The topological-based analysis was revealed that the IL6 was the highest DC (10) and BC (1) among 31 targets (Table 3). Thus, we considered the IL6 as a core target.
A hub pathway of AC against obesity
The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis suggested that a total of 40 targets were associated with 2 pathways: (1) PPAR signaling pathway, (2) Insulin resistance against obesity. Based on the PPI network and topological analysis, we identified a hub pathway (Insulin resistance) correlated with IL6, in contrast, PPAR signaling pathway was no associated with IL6. The targets of the two pathways were exhibited in Table 4. In addition, a bubble graph indicated that insulin resistance is inactivated by DLCs of AC to alleviate obesity, due to lower rich factor than PPAR signaling pathway (Figure 5).
The analysis of pathways-targets-compounds (PTC) networks
A pathway-target-compound (PTC) network was constructed in Figure 6. The integrated network consisted of 30 nodes 61 edges. The nodes indicate a total number of each element: pathways, targets, and compounds. The edges stand for the associations of the three elements. The PTC network suggested that the 2 pathways, 7 targets, and 21 compounds are the significant factors to dampen obesity.
The molecular docking test (MDT) on a key target
The MDT was unveiled that IL6 (PDB ID: 4NI9) was related to 3 compounds out of DLCs from AC: (1) Andrographolide (PubChem ID: 5318517), (2) Deoxy-d-mannoic lactone (PubChem ID: 541561), and (3) Linoleic acid (PubChem ID: 5280450).
It was observed that Andrographolide (PubChem ID: 5318517) can dock as the most stable complex on IL6 (PDB ID: 4NI9), with -8.1 kcal/mol. Next, the binding energy of Deoxy-d-mannoic lactone (PubChem ID: 541561) was -6.3 kcal/mol with valid affinity (<-6.0 kcal/mol) [17]. Lastly, the affinity of Linoleic acid (PubChem ID: 5280450) was -5.0 kcal/mol with invalid binding energy. Noticeably, Andrographolide (PubChem ID: 5318517) can be considered as a potential IL6 inhibitor, compared with known 7 positive controls in aspects of binding energy (Figure 7), (Table 5).