The synthesized g-C3N4·OH underwent comprehensive characterization through FTIR, XRD, FE-SEM, EDS and TGA/DTA studies, aligning with previous literature findings and affirming the successful synthesis of the catalyst20,21,39. XRD analysis revealed characteristic peaks: for g-C3N4, peaks were observed at 2θ = 13.25 (100 plane) and 27.12 (002 plane). Conversely, g-C3N4·OH displayed peaks at 2θ = 12.08 (100 plane) and 27.88 (002 plane), matching JCPDS card no. 87-1526, indicating that functionalization did not disrupt the CN wedge. The addition of hydroxyl groups exhibits negligible impact on the peak intensity observed in the XRD pattern of the catalyst. However, there is a discernible shift towards higher angles for the (002) crystal plane. This shift suggests that hydroxyl grafting results in a reduction of the interlayer spacing within the catalyst. Such a denser structure is likely attributed to electron localization and enhanced binding between the layers Fig. 2. FT-IR spectrum confirmed the presence of functional groups in g-C3N4·OH, with a notable triazine network stretching frequency at 807.76 cm–1 and peaks at 2950-3450 cm–1 corresponding to NH, NH2, and OH frequencies, ensuring g-C3N4·OH formation Fig. 3.Figure 2Comparative XRD spectra of the synthesized g-C3N4 and g-C3N4·OH.Figure 3Comparative analysis of FT-IR spectra between synthesized g-C3N4 and g-C3N4·OH.FE-SEM analysis at varying magnifications revealed a loosely wrinkled porous sheet-like structure, indicative of increased surface area, void presence, and stacked thin layers or sheets Fig. 4. EDX spectrum analysis of g-C3N4·OH exhibited elemental constituents: nitrogen (58.75), carbon (38.86), and oxygen (2.39), consistent with previous literature20,39, establishing a congruent composition of N, C, and O as main elements Fig. 5.Figure 4FE-SEM images showcasing synthesized g-C3N4·OH at varying magnifications.Figure 5Elemental composition analysis using EDX spectrum.As depicted in Fig. 6 the TGA curve for g-C3N4·OH shows distinct weight loss stages. Initially, a 4.99% weight loss occurs around 120 °C, likely due to the evaporation of physically adsorbed water and volatile impurities. Between 120 °C and 440 °C, an additional 2.37% weight loss is observed, which can be attributed to the removal of hydroxyl groups (–OH) and decomposition of organic components or residual solvents. From 440 °C to 600 °C, a significant 10.01% weight loss occurs, indicative of the thermal decomposition of g-C3N4·OH itself, including the breakdown of its structure and loss of more stable organic components. Notably, only 17.37% of the compound was decomposed up to 600 °C, highlighting the excellent thermal stability of g-C3N4·OH. The DTA plot (Fig. 7) for g-C3N4·OH demonstrates an initial increase in the DTA signal from room temperature to 325 °C, indicating endothermic processes such as the evaporation of water and the decomposition of surface groups. Following this, the DTA signal decreases from 325 °C to 600 °C, indicating exothermic reactions likely due to the thermal breakdown of the material. This behavior suggests that the material absorbs heat initially for the removal of moisture and volatile components, and subsequently, it releases heat as it undergoes thermal degradation. The increase to 10 μV/mg followed by a decrease to 0 μV/mg highlights the distinct endothermic and exothermic phases of the thermal behavior of g-C3N4·OH.Figure 6TGA analysis of g-C3N4·OH in N2 atmosphere at 10 ◦C/min up to 600 °C.Figure 7DTA analysis of g-C3N4·OH conducted in N2 atmosphere at 10 ◦C/min up to 600 °C.To refine the reaction conditions, we specifically selected 4-methylthiazole-5-carboxaldehyde, ethyl acetoacetate, and hydroxylamine hydrochloride as the key components for synthesizing 4-methyl thiazole-substituted methylene isoxazole-5-one, serving as our model reaction. Initially, we commenced the reaction without any catalyst at room temperature in water and also without any solvent but observed only minimal product formation (Table 1; entry 1, 2). Subsequently, we embarked on catalyst optimization, exploring a variety of readily available catalysts, including both acid and base types, with a 20 mg loading, while employing water as the solvent at ambient temperature (Table 1; entries 3-10). Unfortunately, the results did not meet our expectations. Turning our attention to synthesized g-C3N4 and g-C3N4.SO3H, previously synthesized in our laboratory, we achieved yields of 75% and 87%, respectively, under identical reaction conditions (Table 1; entry 11-12). Remarkably, utilizing g-C3N4·OH led to a substantial increase in product yield, reaching 91% within a 30-minute reaction time, marking our most successful outcome so far (Table 1; entry 13). Subsequent optimization of catalyst quantity, temperature, and solvent choice with g-C3N4·OH (Table 1; entry 14-20) revealed the optimal conditions: 15 mg catalyst loading at room temperature in water (Table 1; entry 14). Utilizing these optimized conditions, we proceeded to explore the substrate scope, incorporating a range of aliphatic and aromatic aldehydes. Subsequently, we diversified the hydroxylamine source, substituting it with phenyl-hydrazine and hydrazine hydrate, resulting in excellent yields. The notable advantages of g-C3N4·OH catalyst include its high catalytic efficiency, ease of synthesis, and compatibility with aqueous media, making it a promising candidate for sustainable and environmentally friendly synthetic methodologies.
Table 1 Optimization of reaction conditions for the synthesis of 4-methyl thiazole-substituted methylene isoxazole-5-one as a model reaction. Significant values are in bold.Following synthesis, the products underwent purification via column chromatography and were subsequently characterized using 1H and 13C NMR spectroscopy to confirm the structure of the compounds, ensuring the integrity and accuracy of our findings.Following the optimization of all reactions, we investigated the reusability of the catalyst in the model reactions over multiple runs. The results are presented in Fig. 8. Remarkably, the synthesized g-C3N4·OH demonstrated impressive reusability, maintaining its catalytic activity over six consecutive cycles without any discernible loss, as confirmed by FT-IR and XRD studies (Fig. 9). To explore the versatility of the optimized reaction, its efficiency was further evaluated for synthesizing various derivatives as shown in Table 2. The efficacy of the protocol was further examined through a comparative analysis between the current study and prior research efforts as shown in Table 3.Figure 8Assessing the sustainability of the g-C3N4·OH catalyst through recyclability studies.Figure 9The comparative spectral analysis of FT-IR and XRD of both freshly synthesized and recycled catalyst.Table 2 Compilation of synthesized isoxazol-5-one/pyrazol-3-one derivatives.Table 3 The comparative investigation of synthesizing the Isoxazol-5-one/Pyrazol-3-one scaffold using the current method with reported methods.Significant values are in bold.Based on the literature40,45, a plausible mechanism for the proposed reaction is depicted in Scheme 1 g-C3N4·OH in water served as a catalyst, facilitating the generation of a hydronium ion that accelerated the reaction, likely through the formation of an active cationic species. In this proposed mechanism, ethyl acetoacetate and a NH2-bearing compound (second reactant) underwent reaction in water solvent, leading to the expected formation of an oxime intermediate (I). This intermediate subsequently underwent Knoevenagel condensation with an activated aldehyde, yielding intermediate (II). Following this, intramolecular cyclization among hydroxyl and carbonyl groups occured, resulting in the formation of intermediate (III). Finally, intermediate (III) eliminated an ethanol molecule to yield the final desired product.Scheme 1Proposed reaction mechanism for the synthesis of isoxazole-5-one and pyrazol-3-one scaffolds.Docking analysisThe synthesized molecules were docked in the proteins predicted by the PASS online software (Table 4) to check their binding with the respective enzymes. The three proteins of insulin-degrading enzyme (IDE) with PDB ID 3E4A, agonist-boundi G-protein couple receptor (GPCR) with PDB ID 6KPC, and Fibronectin type-III (FNIII) domain of human with PDB ID 2CRM were retrieved from the RCSB database. The IDE is responsible for the proteolytic inactivation and degradation of the insulin46. Insulin belongs to the class of the family that regulate the peptide hormone that are involved in various physiological process like homeostasis of glucose and energy to cognition and memory47. Inhibition of IDE can reduce the catabolism of the insulin, by potentiate the insulin signaling within the cell48. GPCR protein takes part in many crucial physiological functions, which makes them a pharmaceutically suitable druggable target49. FNIII is a glycoprotein that act as a crucial link within cells and their outer matrices50. This target is recognized as a target for many bacterial proteins51.
Table 4 Additional evaluation of the biological activity prediction for 5J using the PASS online program (with similar findings across other results).Based on the number of enzymes, against which our ten synthesized ligands displayed the best predicted activity, three set of docking calculations were conducted. The outcome of the calculations is reported in the Tables 5, 6, 7, 8 and 9 and Figs. 10, 11 and 12. Tables 5, 7 and 9 represent the docking score along with other parameters. Tables 6, 8, and 10 displayed the crucial interactions formed during the interaction study. Figures 10, 11 and 12 represent the 3D and 2D interaction plots of the eight ligands with 3E4A, 6KPC, and 2CRM respectively. Interestingly, it was observed that all the three set of calculations, molecules 5B and 5G were not docked in the selected PDBs.
Table 5 List of the docking score of the synthesized eight molecules with insulin-degrading enzyme (3E4A).Table 6 List of the docked candidates with their docking score and amino acid interaction with insulin-degrading enzyme (3E4A).Table 7 List of the docking score of the synthesized eight molecules with cannabinoid receptor-GiComplex (6KPC).Table 8 List of the docked candidates with their docking score and amino acid interaction with cannabinoid receptor-Gi complex (6KPC).Table 9 List of the docking score of the synthesized eight molecules with forth FNIII domain (2CRM).Figure 10The 3D and 2D interaction plots of the synthesized molecules with insulin-degrading enzyme (3E4A).Figure 11The 3D and 2D interaction plots of the synthesized compounds with cannabinoid receptor-Gi complex (6KPC).Figure 12The 3D and 2D interaction plots of the synthesized molecules with forth FNIII domain (2CRM).Table 10 List of the docked candidates with their docking score and amino acid interaction with forth FNIII domain (2CRM).3E4A dockingOn observing the docking score of the eight compounds with 3E4A, it was observed that 5C displayed the higher docking score (− 23.47 kcal/mol) followed by 5A, 5D, 5J, 5I, 5H, 5E, and 5F (decreasing order). The results of best four docked molecules are tabulated in the form of Table 5.On observing the 2D docking interaction pattern, it was observed that in all the docked complexes, the crucial interactions were present48. Moreover, it is clearly visible from the Fig. 10 and Table 6, that top three candidates displayed crucial interaction with the Zn(II) atom. Apart from the interaction with Zn(II), interaction with important residues like, His108, Gln111, Asn139, Arg824, and Tyr831 was reported in the eight docked complexes (Fig. 10 and Table 6). It is well reported in the literature that IDE is an metalloenzyme, and the interaction with Zn(II) metal ion can cause the inhibition in the proper functioning of the protein48.6KPC dockingOn observing the docking score of the eight compounds with 6KPC, it was observed that 5C displayed the higher docking score (-33.96 kcal/mol) followed by 5I, 5H, 5I, 5D, 5A, 5E, and 5F (decreasing order). The results of best docked molecules are tabulated in the form of Table 7.On observing the 2D interaction plots of 6KPC docked complexes, the crucial interactions were witnessed52. Moreover, it is clearly visible from the Fig. 11 and Table 8, that the eight docked candidates displayed crucial interactions with the residues like, His95, Lys109. Ser285. More specifically, in all the complexes, a common interaction with the residue His95 was witnessed, and the interaction with this amino acid is considered important to inhibit the proper function of the enzyme52.2CRM dockingOn observing the docking score of the eight compounds with 6KPC, it was observed that 5C displayed the higher docking score (− 18.19 kcal/mol) followed by 5H, 5J, 5D, 5E, 5F, 5A, and 5I (decreasing order). The results are tabulated in the form of Table 9.On observing the 2D interaction plots of 2CRM complexes, all the reported crucial interactions were observed53. Moreover, it is clearly visible from the Fig. 12 and Table 10, that the eight docked candidates display hydrogen bonding interactions with the amino acids residues like, Ser61, Asn62, and Phe87. All the eight docked candidates displayed at least two hydrogen bonding. More specifically, in all the complexes, a common interaction with the residue Ser61 (except 5F) was witnessed. The successful interaction of these molecules within the binding domain of 2CRM may cause inhibition of the enzyme. The docking scores of all active compounds against proteins 3E4A, 6KPC, and 2CRM are illustrated in a visual format in Fig. 13, providing a clearer understanding.Figure 13Visual depiction of the docking of eight active molecules within the binding pocket of proteins 3E4A, 6KPC, and 2CRM.ADMET/pharmacokinetics and physicochemical propertiesThe common docked candidates, 5C, 5A, 5D, 5J, 5I, 5H, 5E, and 5F were selected to check their ADMET (absorption, distribution, metabolism, excretion, and toxicity) properties. This step is considered the most crucial step in the drug design and development. Therefore, the physicochemical and ADMET properties of the eight shortlisted compounds were carefully examined using online web-based tools. The predicted physicochemical parameters from SwissADME are reported in Table S1. These parameters help to evaluate the drug-likeness rules, as per Lipinski’s rule of 554, and Veber’s rule55. These compounds show molecular weight less than 500 Dalton, H-bond acceptors less than 10, H-bond donors less than 5, and MLogP (Lipophilicity threshold) less than 5. Moreover, all eight molecules successfully follow the drug-like properties as per the Ghose model56, Egan model57, and Muegge model58. The selected compounds did not violate the drug-likeness rules. This suggests that all compounds possess drug-like attributes and, thus, can be considered as biologically active molecules.Figure 14 shows the Egan BOILED-Egg plot of the selected eight molecules. The BOILED-Egg plot studies the two pharmacokinetic behaviors and predicts gastrointestinal absorption and brain penetration (BBB) of the molecules. The yellow region in the plot corresponds to the physicochemical space of molecules with the highest probability of being permitted to the brain. It represents the hydrophobicity (WLOGP) and polarity (TPSA), which allow the graph to present good gastrointestinal absorption. The white region corresponds to the space with the highest probability of being absorbed by the gastrointestinal tract, i.e. molecules that possess only good gastrointestinal absorption. The potential substrate of P-gp (PGP+), and non-substrate (PGP-) are represented by blue, and red dots respectively. The BOILED-Egg plot reveals that all compounds can be passively absorbed by the gastrointestinal tract. However, compounds 5C, 5H, 5E, 5F, and 5D lie within the yellow region, indicating that these molecules are highly likely to be absorbed by the gastrointestinal tract and permeate into the brain. Also, all the molecules were predicted to be non-substrate, suggesting that, they have the potential for higher absorption in the gastrointestinal tract, increased BBB penetration, higher bioavailability, and better distribution within tissues. Figure S17 represents the bioavailability radar chart. It is a graphical method used to evaluate the drug-likeness of a molecule with the help of various physicochemical properties: (1) LIPO (Lipophilicity): It measures the ability of the molecules to dissolve in non-polar solvents, (2) size (molecular weight): It indicates the size of the molecule, (3) POLAR (Total polar surface area): predicts the ability of the molecule to form H-bonding, (4) INSOLU (Solubility): predicts the solubility of the molecule in water, (5) INSATU (Insaturation): indicates the proportion of unsaturated carbons in the molecules, and (6) FLEX (Flexibility): indicates the number of rotatable bonds responsible for molecular flexibility. It is clear from the figure that all the compounds exhibit the most physicochemical properties in the acceptable pink region while following Lipinski’s rule of five and Veber’s rule. This suggests that the molecules are well-optimized for drug-likeness, like, compound 5I. However, a minor deviation is seen in the INSATU parameter of compounds 5A, 5D, 5J, and 5E, and a large deviation is seen in compounds 5C and 5F. Minor deviation in INSATU suggests that the molecule has a slight imbalance in its degree of saturation, but may not significantly impact the drug-likeness of the molecules. Thus, compounds, 5I, 5A, 5D, 5J, and 5E can act as promising drug-like candidates. The predicted ADMET properties of the eight molecules are present in Table 11. These properties are based on the predicted parameters from pkCSM. The reported absorption parameters include water solubility (logS in mol/L), Caco2 permeability (logPapp in cm/s), Intestinal absorption (% Absorption), Skin Permeability (log Kp), and P-glycoprotein II inhibitor. From the water solubility parameter, compounds 5A, 5D, 5J, 5I, 5E, and 5F show better solubility, with 5J being the most soluble. The compounds 5C and 5H show moderate solubility. In Caco2 permeability, compounds 5C, 5D, 5H, 5E, and 5F shows good permeability, whereas, compound 5A, 5J and 5I shows lower permeability. Except for 5J, all the remaining compounds show high intestinal absorption in the % absorption. Also, 5C and 5E show 100% absorption. In skin permeability, compound 5A, 5J, 5I, 5C, and 5F shows good skin permeability, whereas, 5D, 5H, and 5E are on borderline. Distribution parameters contain BBB Permeability (log BB), and CNS Permeability (log PS). From BBB Permeability (log BB), molecules 5C, 5D, 5H, and 5E show good BBB permeability, indicating the potential for central nervous system (CNS) activity. Compounds 5A, 5J, 5I, and 5F have lower BBB permeability, which may limit their CNS activity. The CNS permeability parameter shows that compounds 5C and 5H have the highest CNS permeability, indicating potential for CNS activity. The molecules, 5D, 5E, and 5J have moderate CNS permeability, and compounds 5A, 5I, and 5F show the lowest CNS permeability, thus, limiting CNS penetration. From the metabolism parameters, compounds 5J and 5F are the best candidates for metabolic safety, as they do not inhibit any of the major CYP enzymes, thus, reducing the interactions with the drugs. Compounds 5A, 5D, 5I, and 5E, inhibit only CYP1A2, whereas, compounds 5C and 5H may cause significant metabolic interactions due to their broad inhibition profile of various CYP enzymes. The Excretion parameters include Total Clearance (in L/h/kg), and Renal OCT2 Substrate. From the Total Clearance parameter, molecules 5C, 5A, 5D, 5J, 5I, and 5E show moderate clearance values, favoring maintaining effective drug levels. Molecules 5H and 5F show low clearance, suggesting potential issues with drug accumulation and toxicity. In Renal OCT2, none of the compounds are substrates of OCT2 transporter, thus, indicating a reduced risk of renal drug-drug interactions through this pathway. On analyzing the toxicity parameters, compounds 5A, 5D, 5J, and 5I are negative, making them the safest option, whereas, compounds 5C and 5F may possess the risk of mutagenicity and liver toxicity, compound 5H possess the risk of hepatotoxicity, and compound 5E shows the risk of skin sensitization. Based on a balance among the ADMET parameters, the compounds 5A, 5D, 5J, and 5I can be further recommended for drug development. However, among the shortlisted four compounds, the 5D compound emerges as the best candidate that passes all the key ADMET properties and toxicity tests. Compound 5D possesses the following characteristics: (1) Good permeability and intestine absorption, (2) Good BBB permeability and moderate CNS permeability, (3) only inhibits CYP1A2, thus, reducing the risk of drug-drug interactions, (4) moderate clearance, suggesting balanced profile, and (5) negative for all tested toxicity parameters 9AMES toxicity, hERG I inhibition, hepatotoxicity, and skin sensitization). These properties demonstrate that compound (5D) can be the most promising candidate for further drug development. Based on the ADMET profile, compound 5D has been selected for the molecular dynamic simulation studies with the selected enzyme 3E4A, 6KPC, and 2CRM (3E4A-5D, 6KPC-5D, and 2CRM-5D).Figure 14BOILED-Egg diagram evaluating passive gastrointestinal absorption (HIA) and Blood-brain barrier penetration (BBB) with WLOGP vs TPSA for the eight common docked candidates after drug-likeness and lead-likeness studies generated from the SwissADME59 web tool.Table 11 List of the predicted ADME/pharmacokinetics properties of the eight common docked inhibitors retrieved from web tool pkCSM60.Molecular dynamics simulation studiesMolecular dynamics simulations were conducted for 300 ns on the ligand 5D in complex with 3EA4, 6KPC, and 2CRM to check the stability of 5D within the biological environment of the three proteins. The stability of these complexes was evaluated via plots like RMSD (protein and ligand), RMSF (protein and ligand), Rg, and H-bonding as shown in Figs. 15 and S18. The average values of protein-ligand RMSD, RMSF, Rg, and H-bonding are shown in Table S2. Protein-RMSD of complexes was carried out to check the structural deviations throughout the time duration of 300 ns. The average RMSD values for the 3E4A, 6KPC, and 2CRM were found to be 0.18 ± 0.02 nm, 0.59 ± 0.14 nm, and 0.61 ± 0.10 nm respectively. We observed from Fig. 15 that the stability of the protein backbone has not been impacted by the docked compound bound in the active site of the respective protein. The stability of the ligand in the active site of the respective protein is analyzed by the ligand-RMSD plot. The average ligand-RMSD values for 3E4A-5D, 6KPC-5D, and 2CRM-5D are 1.16 ± 0.40 nm, 0.05 ± 0.04 nm, and 0.05 ± 0.01 nm respectively. The outcome from ligand-RMSD plots revealed that molecule 5D displays a similar kind of fluctuation behaviour in 6KPC and 2CRM, indicating consistent binding stability and interaction pattern in these complexes. In contrast, the higher RMSD value for 3E4A-5D suggests greater fluctuation in the complex. The RMSD analysis confirmed that all the docked candidates have greater stability than the reference in the CDK5 binding site. Overall, the protein-ligand RMSD analysis indicates that molecule 5D exhibits stable binding in 6KPC and 2CRM, it shows greater fluctuation in 3E4A despite the protein fluctuation in 3E4A being lower. Lower protein RMSD and higher ligand RMSD depict that while the overall protein structure remains stable, the ligand shows significant fluctuations within the binding site of the protein, suggesting less stable binding of the ligand to the protein. The Rg measures the compactness of the protein structure and provides insight into the overall shape and folding of the protein. The average Rg values of 3E4A, 6KPC, and 2CRM are 2.95 ± 0.01 nm, 2.99 ± 0.04 nm, and 1.71 ± 0.07 nm respectively. On comparing Rg with the protein RMSD, the 3E4A shows lower protein RMSD, indicating overall structural stability. The average Rg value of 3E4A indicates a relatively compact structure, which might contribute to the observed stability of the protein. In the case of proteins, 6KPC, and 2CRM, higher average protein RMSD values indicate more structural deviation over time. However, the Rg value of 6KPC and 2CRM indicate that 6KPC is similar in compactness as in 3E4A, while 2CRM is more compact, possibly contributing to the more stable ligand binding in the 2CRM protein. Overall, the structural stability and compactness of the protein correlate with the observed binding stability of the ligands, indicating that a stable protein structure ensures stable ligand binding as in the case of 6KPC and 2CRM. However, in 3E4A, the protein deviation pattern highlights a stable protein behaviour, but, shows significant fluctuations within the active site of the protein.Figure 15Graphical representation of protein-RMSD, ligand-RMSD, and Rg of 5D-6KPC, 5D-3EA4, and 5D-2CRM formed during the molecular dynamic simulations of 300ns.Further to check the flexibility of residues in the protein backbone, protein RMSF plots were generated, as shown in Fig. S18. The average protein-RMSF of 3E4A, 6KPC, and 2CRM is 0.10 ± 0.05 nm, 0.27 ± 0.10 nm, and 0.75 ± 0.46 nm respectively. The average protein RMSF for 3E4A is relatively low, indicating that the individual residues of the protein are stable and exhibit minimal fluctuation. In the case of 6KPC and 2CRM, the average RMSF values are higher, indicating that the residues in these protein chains are more flexible. Ligand- RMSF measures the atom’s deviation from its average position during the simulation run. The average ligand-RMSF of 5D in 3E4A, 6KPC, and 2CRM is 0.10 ± 0.06 nm, 0.05 ± 0.04 nm, and 0.05 ± 0.04 nm respectively. The average ligand RMSF of 5D is similar in 6KPC, and 2CRM, indicating their similar behaviour in the binding site of the respective proteins. A similar pattern is also observed in the ligand RMSD plots of 5D, which show consistent binding stability of 5D in the two proteins 6KPC, and 2CRM. The average ligand RMSF of 5D in 3E4A indicates moderate fluctuations in the position of the ligand, correlating with its fluctuation observed in the ligand RMSD. Additionally, average H-bonds were evaluated from the three complexes, as shown in Fig. 7. Hydrogen bonding reflects bonding between the ligand and protein formed within the complex during the simulation run. The average H-bond value of 3E4A, 6KPC, and 2CRM is 0.16 ± 0.47, 0.40 ± 0.55, and 0.16 ± 0.45 respectively. The average number of H-bonds formed indicates the involvement of these bonds in stabilizing the complex with the ligand 5D and the respective proteins. Overall from all the studied properties, we can conclude that 6KPC and 2CRM displayed slightly higher protein fluctuations with stable ligand binding, supported by a more compact structure (2CRM) and moderate H-bonding. Protein 3E4A displays a stable protein structure but exhibits fluctuations in ligand binding, possibly affecting its interaction dynamics.Gram scale synthesisWe showcased the robustness of our methodology for industrial-scale applications through a compelling demonstration of gram-scale synthesis. Initially, 4-methylthiazole-5-carboxaldehyde (1.271 gm), ethyl acetoacetate (1.263 gm), and hydroxylamine hydrochloride (0.695 gm), were reacted for the synthesis of 4-methyl thiazole-substituted methylene isoxazole-5-one (1.861 gm). This synthesis was seamlessly executed with the assistance of a mere 15 mg of the g-C3N4·OH catalyst in water at ambient temperature, culminating in an impressive 89% product yield achieved within a mere 45 minutes. The validation of reaction completion was confirmed through TLC analysis, ensuring precision and reliability. Subsequently, a pivotal step in our process was the efficient recovery of the catalyst utilizing methanol further enhancing the sustainability and cost-effectiveness of our approach. Following this, the crude product underwent a rigorous purification process, elevating its quality to unparalleled levels. This meticulous purification yielded the product in its purest form, boasting exceptional yields that underscore the viability of our methodology for large-scale industrial applications. This demonstration represents a significant leap forward in industrial chemistry, promising enhanced efficiency, sustainability, and economic viability (Scheme 2).Scheme 2Synthesis of 3-methyl-4-((4-methylthiazol-5-yl)methylene)isoxazol-5(4H)-one on a gram scale.Green chemistry matrixIn recent years, there has been a notable shift towards developing environmentally friendly and sustainable methods for synthesizing organic compounds. One such pioneering approach is green chemistry, which provides a comprehensive framework for evaluating the environmental impact of chemical reactions61,62,63. An exemplary illustration of this approach is the utilization of g-C3N4·OH as a catalyst in the synthesis of isoxazol-one and pyrazol-one derivatives. This methodology has exhibited remarkable attributes, including a low E-factor (0.28–0.72), high atom economy (63.63–81.06), superior reaction mass efficiency (57.96–77.97), elevated process mass intensity (1.30–1.75), and an impressive eco-score (74.66–78.71). These compelling findings underscore the viability of employing g-C3N4·OH catalysts in the synthesis of these scaffolds as a sustainable and eco-conscious strategy, with minimal adverse effects on the environment. [Detailed data and calculations are provided in the SI File].
Highly efficient synthesis of isoxazolones and pyrazolones using g-C3N4·OH nanocomposite with their in silico molecular docking, pharmacokinetics and simulation studies
