Selection of proteins for the simulation stability studiesAmong the large set of proteins, the highest number of edge counts were obtained by SIRT3/FOXO1 and PPARGC1A. The “system builder” was utilized in the solvation of the protein-ligand complex, where the “TIP4P” was used as a solvent, and the buffer was added within a distance of 10/10/10 Å. The model was minimized using the neutral Na + ions via the OPLS4 forcefield. The protein-ligand complex was further minimized at 100 picoseconds, and the residues of the protein-TIP4P compound were simulated at 100 nanoseconds under a pressure of 1.01 at 311 K temperature20.Rat acute toxicityThe “GUSAR” (General Unrestricted Structure-Activity Relationships) (https://www.way2drug.com/gusar/acutoxpredict/) module was used to predict acute toxicity. The GUSAR module predicts acute toxicity by running over 10 K chemical entities, whereas the QSAR approach was used to estimate acute toxicity21.Results and discussionsIMPHY000797 and its physicochemical propertiesIMPHY000797 was determined to have a molecular weight of 521.08 Daltons and a log-p value of -3.2, with two hydrophobic rings; the NHBA and NHBD were found to be 15 and 8, respectively. The DLS score was found to be 0.81. Thirty-one genes, HDAC1, SOD2, FOXO1, PPARGC1A, UBB, AMPK, IDH2, NTRK2, CREBBP, SIRT3, CDC34, PRKAA2, ATP5F1B, MAP4K4, PRKAA1, CDK1, PDE2A, ALOX5, IGF1R, HSD11B1, PTGS1, RELA, ESRRB, EPHX2, HMGCR, MAP2K1, ADORA1, NTRK1, NTRK3, PRKDC, HIF1A, were found to be modulated by the compound IMPHY000797, and 3583 of genes were found to be involved in the modulation of PD which was obtained using DisGeNET database. The common genes were identified from both sets (compound genes and DisGeNET genes) which predicted 25 common genes (HDAC1, SOD2, FOXO1, PPARGC1A, UBB, NTRK2, CREBBP, SIRT3, CDC34, PRKAA2, ATP5F1B, MAP4K4, PRKAA1, CDK1, PDE2A, ALOX5, IGF1R, RELA, EPHX2, MAP2K1, ADORA1, NTRK1, NTRK3, PRKDC, HIF1A) as shown in Fig. 2.Fig. 2Venn illustration representation. (A) Targets involved in the modulation of PD (CUI: C0949855) and targets modulated by the IMPHY000797 compound. (B) Multiple sets of 3583 genes were collected from the different keywords involved in the pathogenesis of PD (CUI: C3825201, CUI: C0030567, CUI: C0751651, and CUI: C0949855). (C) The GO terms Biological Process (BP), Molecular function (MF), and Cellular component (CC) via KEGG-mediated pathways.Gene enrichment analysis for IMPHY000797 and network constructionThe genes modulated in the PD were obtained using the DisGeNET database with the access codes “CUI: C3825201, CUI: C0030567, CUI: C0751651, and CUI: C0949855”. Whereas targets modulated by the IMPHY000797 compound (IMPHY000797) were obtained using Swiss target prediction (http://www.swisstargetprediction.ch/) and Binding database (https://www.bindingdb.org/rwd/bind/index.jsp), which were having a probable score of 0.5-1. Overall, of 3583 genes were obtained using DisGeNET database, whereas the compound IMPHY000797 was found to modulate 31 genes (HDAC1, SOD2, FOXO1, PPARGC1A, UBB, AMPK, IDH2, NTRK2, CREBBP, SIRT3, CDC34, PRKAA2, ATP5F1B, MAP4K4, PRKAA1, CDK1, PDE2A, ALOX5, IGF1R, HSD11B1, PTGS1, RELA, ESRRB, EPHX2, HMGCR, MAP2K1, ADORA1, NTRK1, NTRK3, PRKDC, HIF1A). Furthermore, the commonly predicted 25 genes were queried in the STRING database (https://string-db.org/) to analyze the probable protein-protein interactions. The threshold used to choose high-confidence protein-protein interactions from the “STRING” database was set to “0.700,” which can be found in the “settings” section of the STRING database. Protein-protein interactions can be classified as highly confident if the high confidence factor exceeds 0.700. There were 25 nodes and 63 edges. The average node degree was determined to be 2.61, with a clustering value of 0.62. The protein-protein enrichment p-value was 4.33e-1, indicating that the proteins had many interactions. This enrichment suggested that the targets were physiologically related to each other. Eighty-one enriched KEGG pathways were obtained (Supplementary File 1)22,23. Figure 3 illustrates how the network between the gene/pathways and compound targets was built using “Cytoscape 3.10.0”. SIRT3/FOXO1 and PPARGC1A displayed the highest number of “edge counts” among all the genes with a neighborhood connectivity of 6.4, the “FOXO signaling pathway” was found to have an edge count of 7, followed by an in-degree of 3, and an outdegree of 6.Fig. 3Network between pathway/gene-mediated targets by the IMPHY000797 compound (IMPHY000797). (A) Protein-protein interaction with the highest edge counts among the common genes collected. (B) The most modulated genes/pathways via IMPHY000797.Gene Ontology (GO) and Pearson correlation analysisThe GO term analysis predicted 56 KEGG pathways, whereas the “Longevity regulating pathway” was found to have a false discovery rate (FDR) of 3.57E-08. The KEGG pathway also predicted the “FOXO signaling pathway” (hsa04068) with an FDR of 1.41E-07, indicating probable significance, which could have resulted in more false results. A total of 175 BP was determined, among which “Cellular response to oxidative stress” (GO:0034599) and “Response to oxidative stress” (GO:0006979) were found to have the lowest FDR of 1.44E-06 and 8.98E-09 and were found to modulate eight proteins (PPARGC1A, SIRT3, PRKAA2, FOXO1, CDK1, RELA, HIF1A, SOD2). The 29 MF were predicted, among which “Small molecule binding” (GO:0036094) was predicted with an FDR of 5.11E-07, whereas it was found to modulate 19 proteins (CDC34, ATP5F1B, NTRK2, HMGCR, MAP2K1, PRKDC, IDH2, PDE2A, MAP4K4, FOXO1, NTRK3, HSD11B1, ADORA1, PRKAA2, SIRT3, CDK1, NTRK1, SOD2, IGF1R). Overall, of 7 CF were predicted, among which “Cytoplasm” (GO:0005737) was found to modulate 31 genes (CDC34, ATP5F1B, CREBBP, PPARGC1A, NTRK2, HMGCR, MAP2K1, PRKDC, IDH2, PDE2A, MAP4K4, PRKAA1, NTRK3, PTGS1, HSD11B1, ADORA1, PRKAA2, HDAC1, ALOX5, FOXO1, ESRRB, SIRT3, CDK1, RELA, EPHX2, CA1, NTRK1, HIF1A, SOD2, UBB, IGF1R). All the enrichment analyses predicted various common genes, among which SIRT3, FOXO1, and PPARGC1A were identified as the most common genes involved in the modulation of oxidative stress. The diagram has been presented as a violin plot, which indicates the top-most modulated genes throughout the enrichment analysis, as shown in Fig. 4.Fig. 4Violin plot representation with high and low probable regions. (A) Biological Process indicating (red color) the highest modulated genes. (B) Molecular function indicating (red color) the highest modulated genes. (C) Cellular component indicating (red color) the highest modulated genes.The Pearson correlation matrix statistical analysis was performed for the GO terms BP, MF, and CC. The predicted Pearson correlation for BP at 95% between the “observed gene count vs strength” was found to be -0.834 and − 0.717 with a two-tailed p-value < 0.0001 (****). In contrast, the correlation between “observed gene count vs false discovery rate” was found to be -0.604 and − 0.380 with a two-tailed p-value < 0.0001 (****). The predicted Pearson correlation for MF at 95% between the observed gene count vs. strength was found to be -0.919 and − 0.663 with a two-tailed p-value < 0.0001 (****). In contrast, the correlation between “observed gene count vs false discovery rate” was found to be -0.639 and 0.030 with a two-tailed p-value of 0.071, indicating no significance; it means that there was a lack of statistical significance between the “observed gene count vs false discovery rate” which suggests that the null hypothesis was not rejected. The predicted Pearson correlation for CC at 95% between the “observed gene count vs strength” was found to be -0.986 to -0.230 with a two-tailed p-value of 0.0215, indicating a mild significance (*). In contrast, the correlation between “observed gene count vs false discovery rate” was found to be -0.908 and 0.632 with a two-tailed p-value of 0.047, indicating no significance, as shown in Supplementary File 1 and Fig. 5.Fig. 5Correlation matrix for the collected GO terms. (A) The correlation analysis between strength, False discovery rate, and gene count for biological process (BP). (B) The correlation analysis between strength, False discovery rate, and gene count for molecular function (MF). (C) The correlation analysis between strength, False discovery rate, and gene count for Cellular.component (CC).Clustering of genesThe clustering of gene analysis for the GO terms of BP, MF, and CC was performed in “ClueGo” via Cytoscape 3.10.0, where the ClueGo predicted the FOXO signaling pathway “KEGG:04068” of group 4 was found to modulate 6.34% of associated genes with a total gene count of 7 (CREBBP, FOXO1, IGF1R, MAP2K1, SIRT3, PRKAA2, SOD2). Similarly, group 4 Thyroid hormone signaling pathway “KEGG:04919” was found to modulate 3.53% of the associate genes with a total gene count of 5 (CREBBP, FOXO1, HDAC1, HIF1A, MAP2K1). The group 6 Longevity regulating pathway “KEGG:04211” was found to modulate 5.49% of the associated genes with a total gene count of 6 (FOXO1, IGF1R, PPARGC1A, PRKAA1, PRKAA2, SOD2). All the genes were evaluated using the Statistical Enrichment/Depletion (Two-sided hypergeometric) test with a p-value threshold of 0.05, and the correction technique was assessed by utilizing the Bonferroni step-down method. The ultimate Kappa Score for the groups was determined to be 7, displayed in Supplementary File 2 and Fig. 6.Fig. 6Clusters for the GO term biological process, molecular, and cellular function.Molecular docking studiesAutoDock 4.2The one-to-one (protein/ligand) docking was performed via AutoDock 4.2, among which the protein SIRT3 (PDB ID: 4FVT)/IMPHY000797 complex was found to have the binding energy of -12.42 kcal/mol, the amino acid interactions were found to be SER 149, ALA 146, ASP 156, HIS 248, SER 321 and various unfavourable bonds were also observed with the amino acids THR 320, ASN 229, ILE 254, ASP 231, and GLN 228. The FOXO1 (PDB ID: 5DUI)/ IMPHY000797 complex was found to have 2 hydrogen bond interactions with the amino acid SER 205 and SER 234, whereas there were various unfavourable bond interactions with the amino acids TRP 237, LEU 183, LEU 217, and THR 182 with a docking score of -9.55 kcal/mol. The PPARGC1A (PDB ID: 1XB7)/ IMPHY000797 complex was found to have 2 hydrogen bond interactions with the amino acid interactions ALA 516 and GLY 489 with the docking score − 8.55 kcal/mol, as shown in Supplementary File 3 and Fig. 7.Fig. 72D and 3D representation of the interaction between protein/compound via AutoDock 4.2. (A) 2D and 3D interactions showing multiple hydrogens and π − π interaction with the SIRT3 (PDB ID:4FVT) and IMPHY000797 compound. (B) 2D and 3D interactions showing multiple hydrogens and π − π interaction with the FOXO1 (PDB ID:5DUI) and IMPHY000797 compound. (C) 2D and 3D interactions showing multiple hydrogens and π − π interaction with the PPARGC1A (PDB ID:1XB7).Glide (maestro)The maestro Schrodinger suite 2022-24 (glide) module was used in the prediction of docking scores and interacting amino acids between one protein and multiple ligands: SIRT3 (PDB ID: 4FVT)/ IMPHY000797 complex was found to have a binding energy of -8.95 kcal/mol; whereas 8 hydrogen bonds were formed with the amino acids (ASP 156, ASN 344, THR 320, ALA 146, and GLN 228), where 1 π − π interaction was observed with the PHE 180. The FOXO1 (PDB ID: 5DUI) exhibited a docking score of -8.48 kcal/mol, and 5 hydrogen bond amino acid (SER 184, GLU 188, LYS 192, LYS 233) interactions were obtained with the IMPHY000797. The PPARGC1A (PDB ID:1XB7) was found to have a docking score of -7.217 kcal/mol; whereas it predicted 5 amino acids interactions (GLU 512, GLN 353, LYS 340) as shown in Fig. 8. The results suggested different hydrogen and hydrophobic amino acid interactions with the compound IMPHY000797; whereas more hydrophobicity of the compound could improve the brain permeability.Fig. 82D and 3D representation of the interaction between protein/compound via Schrodinger suite (glide). (A) 2D and 3D interactions showing multiple hydrogens and π − π interaction with the SIRT3 (PDB ID:4FVT) and IMPHY000797 compound. (B) 2D and 3D interactions showing multiple hydrogens and π − π interaction with the FOXO1 (PDB ID:5DUI) and IMPHY000797 compound. (C) 2D and 3D interactions showing multiple hydrogens and π − π interaction with the PPARGC1A (PDB ID:1XB7).Molecular dynamics simulation studiesSIRT3 and IMPHY000797The stability between the SIRT3 and IMPHY000797 was determined using 100 nanoseconds (ns) simulation studies. In contrast, there was no effective interaction from 0 to 18 ns, and few amino acids, such as VAL 292, HIS 248, PHE 157, and PHE 157, were found to have interactions with the compound IMPHY000797. The mild interactions were observed from 20 to 40 ns, whereas the amino acid interacting was found to be VAL 292, HIS 248, PHE 157, and PHE 157. The root means square deviation (RMSD) between the protein/IMPHY000797 was 2.8/4.5 Å. Continuous stability was observed from 50 to 100 ns. The protein/ligand visualized 12 hydrogen bond formation with the amino acids ALA 146, ASP 156, PHE 157, SER 160, SER 162, TYR 165, GLU 177, GLN 228, ASN 229, HIS 248, VAL 292, and GLU 295. The hydrophobic interaction was visualized with the amino acids PHE 157, PRO 160, PRO 176, PHE 180, LEU 199, ILE 230, HIS 248, and PHE 294, as shown in Fig. 9. The hydrophobic amino acids signify their role in the good binding affinity with the amino acids/proteins.FOXO1 and IMPHY000797The stability between the FOXO1 and IMPHY000797 was determined using 100 ns simulation studies. In contrast, there was no effective interaction from 0 to 25 ns, and few amino acids such as ALA 146, ASP 156, GLN 228, HIS 248, THR 320, and SER 321 were found to have interactions with the compound IMPHY000797. The strong interactions were observed from 40 to 100 ns, whereas the amino acid interacting was found to be VAL 292, HIS 248, PHE 157, and PHE 157. Continuous stability was observed from 40 to 100 ns. The RMSD between the protein/IMPHY000797 was 2.8/4.8 Å. The protein/ligand visualized 15 hydrogen bond formation with the amino acids ALA 146, GLY 147, THR 150, ASP 156, ARG 158, GLU 177, GLN 228, ASN 229, ASP 231, HIS 248, VAL 292, THR 320, SER 321, LEU 322, ASN 344 and ARG 345. The hydrophobic interaction was visualized with the amino acids PHE 157, PHE 180, ILE 230, HIS 248, PHE 294, and VAL 234, as shown in Fig. 9. The hydrophobic amino acids signify their role in the good binding affinity with the amino acids/proteins.PPARGC1A and IMPHY000797The stability between the PPARGC1A and IMPHY000797 was determined using 100 ns simulation studies. There was no effective interaction from 0 to 60 ns, and strong interactions were observed from 70 to 80 ns. Continuous stability was observed from 70 to 100 with an RMSD of 4.0–9 Å. The RMSD predicted the fluctuation with the complex molecule (PPARGC1A/IMPHY000797), as shown in Fig. 9.Fig. 9Simulation studies for the top 3 proteins and IMPHY000797 compound to identify the stability of the protein/compound complex at 100 nanoseconds. 1(A) PL-RMSD between the SIRT3 (PDB ID: 4FVT) with the IMPHY000797 compound. (B) The PL-contact timeline indicates the stable interaction of amino acids with the compound EGGC. (C) The PL-histogram plot represents the number of hydrogen, water, and hydrophobic interactions between the amino acids and the IMPHY000797 compound. 2(A) PL-RMSD between the FOXO1 (PDB ID:5DUI) with the IMPHY000797 compound. (B) The PL-contact timeline indicates the stable interaction of amino acids with the IMPHY000797 compound. (C) PL-histogram plot representing the number of hydrogens, water, and hydrophobic interactions between the amino acids and the IMPHY000797 compound. 3(A) PL-RMSD between the PPARGC1A (PDB ID:1XB7) with the compound EGGC. (B) The PL-contact timeline indicates the stable interaction of amino acids with the IMPHY000797 compound. (C) PL-histogram plot representing the number of hydrogens, water, and hydrophobic interactions between the amino acids and the IMPHY000797 compound.Rat acute toxicityTo unearth the bad upshots that may result from the unintended/determined short-term exposure, a compound acute toxicity should be investigated24. Long-term toxicity studies and animal model assessments should be done to choose the dose of a substance. These acute toxicity findings can be further used to determine the substance’s toxicity status25. The computational model of way2drug (PASS) software was used in the prediction of the possible toxicity for the compound IMPHY000797. The bulkiness of the IMPHY000797 compound was determined using the QSAR applicability domain (AD). The bulkiness of the compound was determined using parameters such as Intraperitoneal route of administration (IP), Intravenous route of administration (IV), Oral route of administration, and Subcutaneous route of administration (SC). The rat IP LD50 was found to have class 4, which indicated that the compound was within the AD of the predicted models. The rat IV50 was found to have class 5, which stated that the compound was within the AD of the predicted models. The rat oral LD50 was classified under class 4, and the rat SC LD50 was classified as non-toxic, where both of the parameters were found in the AD of the predicted models, as shown in Table 1.Table 1 Acute toxicity prediction of IMPHY000797 compound (IMPHY000797).
