Synthesis of Thiazolidin-4-Ones Derivatives, Evaluation of Conformation in Solution, Theoretical Isomerization Reaction Paths and Discovery of Potential Biological Targets

Thiazolin-4-ones and their derivatives represent important heterocyclic scaffolds with various applications in medicinal chemistry. For that reason, the synthesis of two 5-substituted thiazolidin-4-one derivatives was performed. Their structure assignment was conducted by NMR experiments (2D-COSY, 2D-NOESY, 2D-HSQC and 2D-HMBC) and conformational analysis was conducted through Density Functional Theory calculations and 2D-NOESY. Conformational analysis showed that these two molecules adopt exo conformation. Their global minimum structures have two double bonds (C=N, C=C) in Z conformation and the third double (C=N) in E. Our DFT results are in agreement with the 2D-NMR measurements. Furthermore, the reaction isomerization paths were studied via DFT to check the stability of the conformers. Finally, some potential targets were found through the SwissADME platform and docking experiments were performed. Both compounds bind strongly to five macromolecules (triazoloquinazolines, mglur3, Jak3, Danio rerio HDAC6 CD2, acetylcholinesterase) and via SwissADME it was found that these two molecules obey Lipinski’s Rule of Five.

The thiazolidin-4-one ring is susceptible to modifications in positions 2, 3 and 5.Such modifications capacitate the search for new compounds with desired activities.In particular, 5-substituted thiazolidin-4-ones derive from synthetic procedures applied to the methylene carbon.Knoevenagel reaction is one of them and has been applied to introduce a substituent in 5-position [8], introducing a double bond, usually of Z-configuration [9].
In our continuous effort towards the synthesis of new pharmacologically interesting scaffolds, we synthesized two 5-substituted thiazolidin-4-ones (DKI39, DKI40).The stereochemical outcome of these compounds, possessing three double bonds, was studied extensively with 2D NMR experiments and computational chemistry, leading to their unambiguous conformational characterization.Specifically, Density Functional Theory (DFT) was used to calculate the lowest energy conformer.The lowest energy conformer was superimposed with the experimental conformation derived using 2D NOESY spectroscopy.If the RMSD of the superimposition was found to be less than 2Å, then, experimentally, the two conformations were considered to be identical.Furthermore, the reaction isomerization paths were studied via DFT to check their stability.
Finally, as a continuation of our previous studies, where similar molecules presented strong binding affinity in macromolecules [10,11], the present compounds were tested through in silico molecular docking in different macromolecules to check if the present compounds are suitable for specific biological targets.The results gave promising bindings and SwissADME showed that both compounds obey Lipinski's Rule of Five.To the best of our knowledge, this is the first study towards this end.

Synthesis
Thiosemicarbazone DKI1 was synthesized from benzaldehyde and thiosemicarbazide under standard conditions and it was further transformed to thiazolidin-4-one DKI36 using methyl 2-chloroacetate, following our previously published conditions. 11Knoevenagel reaction between DKI36 and two aromatic aldehydes in the presence of piperidine provided 5-substituted thiazolidin-4-ones DK139 and DKI40.For DKI39, a similar synthetic strategy has been reported in the literature [12] using DKI36, benzaldehyde and NaOAc in refluxing glacial acetic acid at the final step, but in our hands, this reaction failed to give the desired product.Moreover, Kambe [13] reported a totally different DKI39 synthetic pathway using ethyl thiocyanoacetate as the starting material, providing the desired compound, albeit in very low yield (32%) (Scheme 1).
Molecules 2024, 29, x FOR PEER REVIEW 2 of 20 stereochemical outcome of these compounds, possessing three double bonds, was studied extensively with 2D NMR experiments and computational chemistry, leading to their unambiguous conformational characterization.Specifically, Density Functional Theory (DFT) was used to calculate the lowest energy conformer.The lowest energy conformer was superimposed with the experimental conformation derived using 2D NOESY spectroscopy.If the RMSD of the superimposition was found to be less than 2Å, then, experimentally, the two conformations were considered to be identical.Furthermore, the reaction isomerization paths were studied via DFT to check their stability.Finally, as a continuation of our previous studies, where similar molecules presented strong binding affinity in macromolecules [10,11], the present compounds were tested through in silico molecular docking in different macromolecules to check if the present compounds are suitable for specific biological targets.The results gave promising bindings and SwissADME showed that both compounds obey Lipinski's Rule of Five.To the best of our knowledge, this is the first study towards this end.

Synthesis
Thiosemicarbazone DKI1 was synthesized from benzaldehyde and thiosemicarbazide under standard conditions and it was further transformed to thiazolidin-4-one DKI36 using methyl 2-chloroacetate, following our previously published conditions. 11 Knoevenagel reaction between DKI36 and two aromatic aldehydes in the presence of piperidine provided 5-substituted thiazolidin-4-ones DK139 and DKI40.For DKI39, a similar synthetic strategy has been reported in the literature [12] using DKI36, benzaldehyde and NaOAc in refluxing glacial acetic acid at the final step, but in our hands, this reaction failed to give the desired product.Moreover, Kambe [13] reported a totally different DKI39 synthetic pathway using ethyl thiocyanoacetate as the starting material, providing the desired compound, albeit in very low yield (32%) (Scheme 1).

Structure Assignment
As a convenient starting point for structure elucidation of DKI39, the proton that resonates at 8.54 ppm was chosen.This peak corresponds to H-5, as it appears as a single peak, since it is not coupled with any other peak.Another important proton, the H-3 was identified as it gives a spatial correlation signal with H-5, through 2D-NOESY.Then, C-5 and C-3 were also identified, as they give 1 JC-H correlation with H-5 and H-3, respectively, through 2D-HSQC.Through 2D-COSY, H-2 and H-1 were identified as they give a correlation signal with H-3.Specifically, H-2 was identified because it gives triplet peak.C-2 and C-1 were then also identified, as they give 1 JC-H correlation with H-2 and H-1.H-12 was immediately identified, as it is the only proton that gives a correlation signal with C-10 of the carbonyl, through 2D-HMBC.After the identification of H-12, H-14 was identified as it gives a spatial correlation with H-12, through 2D-NOESY.H-15 and H-16 were

Structure Assignment
As a convenient starting point for structure elucidation of DKI39, the proton that resonates at 8.54 ppm was chosen.This peak corresponds to H-5, as it appears as a single peak, since it is not coupled with any other peak.Another important proton, the H-3 was identified as it gives a spatial correlation signal with H-5, through 2D-NOESY.Then, C-5 and C-3 were also identified, as they give 1 J C-H correlation with H-5 and H-3, respectively, through 2D-HSQC.Through 2D-COSY, H-2 and H-1 were identified as they give a correlation signal with H-3.Specifically, H-2 was identified because it gives triplet peak.C-2 and C-1 were then also identified, as they give 1 J C-H correlation with H-2 and H-1.H-12 was immediately identified, as it is the only proton that gives a correlation signal with C-10 of the carbonyl, through 2D-HMBC.After the identification of H-12, H-14 was identified as it gives a spatial correlation with H-12, through 2D-NOESY.H-15 and H-16 were also immediately identified, as they give a correlation signal through 2D-COSY.H-15 was identified, because it gives a pair of doublet peaks as it correlates with two protons.C-15, C-16, C-12 and C-14 were also identified, since they give a correlation signal with these protons.Through 2D-HSQC of the examined molecules, all carbons were identified except quaternary and carbonyl carbons.The remaining carbons could be identified through 2D-HMBC.Specifically, it was observed that H-5 showed 2 J C-H correlation with C-4, H-12 showed 2 J C-H and 3 J C-H correlation with C-10 and C-11, respectively, and H-15 showed 3 J C-H with C-13.
A similar procedure was carried out for thiazolidinone DKI40.The two structural identification strategies are shown in Supporting Information.The structures of DKI39 and DKI40 are shown in Scheme 2 with their numbering.The chemical shifts for both compounds are shown in Table 1.
Molecules 2024, 29, x FOR PEER REVIEW 3 of 20 also immediately identified, as they give a correlation signal through 2D-COSY.H-15 was identified, because it gives a pair of doublet peaks as it correlates with two protons.C-15, C-16, C-12 and C-14 were also identified, since they give a correlation signal with these protons.Through 2D-HSQC of the examined molecules, all carbons were identified except quaternary and carbonyl carbons.The remaining carbons could be identified through 2D-HMBC.Specifically, it was observed that H-5 showed 2 JC-H correlation with C-4, H-12 showed 2 JC-H and 3 JC-H correlation with C-10 and C-11, respectively, and H-15 showed 3 JC-H with C-13.
A similar procedure was carried out for thiazolidinone DKI40.The two structural identification strategies are shown in Supporting Information.The structures of DKI39 and DKI40 are shown in Scheme 2 with their numbering.The chemical shifts for both compounds are shown in Table 1.

Conformational DFT Analysis
DFT was used to predict the lowest minima for both compounds.Various initial structures were geometry-optimized in order to calculate the lowest energy isomers (E/Z) and their conformers (endo/exo).Exo conformers are shown in Scheme 2, while endo conformers are shown in Scheme 3. In the endo form, the double bond is between C8 and N next to the carbonyl, while in the exo form, the double bond is between C8 and N7.Three dihedral angles were selected for each compound.Specifically, for DKI39 the angles that were selected are formed by the following atoms: 4-5-6-7 (τ1), 10-11-12-13 (τ2) and 6-7-8-9 (τ3), and for DKI40: 4-5-6-7 (τ1′), 6-7-8-9 (τ2′) and 10-11-12-13 (τ3′).The conformations DKI39en_1 and DKI40en_1 were used as initial structures.Then, by rotating one dihedral angle each time by approximately 180°, the next conformation was occurred.For instance, starting from DKI39en_1, if τ1 is rotated by 180°, DKI39en_2 will be obtained (Scheme 3).Also, starting from DKI39en_2, if τ1 and τ2 is rotated by 180°, DKI30en_3 will be obtained, as it is shown in Scheme 3.All conformers were geometry-optimized.
Considering the endo compound for DKI39 and DKI40, the conformations that occurred are showed in Schemes 3 and 4. The double bond, which is considered as endo for both of the compounds, is shown in orange circle, and it is the same for the four different conformations.Considering the endo compound for DKI39 and DKI40, the conformations that occurred are showed in Schemes 3 and 4. The double bond, which is considered as endo for both of the compounds, is shown in orange circle, and it is the same for the four different conformations.
Molecules 2024, 29, x FOR PEER REVIEW 4 of 20 Considering the endo compound for DKI39 and DKI40, the conformations that occurred are showed in Schemes 3 and 4. The double bond, which is considered as endo for both of the compounds, is shown in orange circle, and it is the same for the four different conformations.The relative energies are given in Table 2 for the endo isomorph with the values of dihedral angles for both compounds.Considering the energetics values and the critical distances obtained from NMR experiments, the DKI39en_1 and DKI40en_1 structures are the most favorable isomers for the endo compounds.The structures are shown below in Scheme 5.
Table 2. Relative energies (∆E in kcal/mol) and dihedral angles (τ in degrees) of the minimized structures for DKI39 endo and DKI40 endo compounds a .The conformations DKI39ex_1 and DKI40ex_1 were the initial optimized structures.Then, by rotating the dihedral angles by 180°, additional conformations were obtained, which subsequently were geometry-optimized.For example, from DKI39ex_1 to DKI39ex_2, τ1 dihedral angle was rotated by 180° approximately.Considering the exo The conformations DKI39ex_1 and DKI40ex_1 were the initial optimized structures.Then, by rotating the dihedral angles by 180 • , additional conformations were obtained, which subsequently were geometry-optimized.For example, from DKI39ex_1 to DKI39ex_2, τ 1 dihedral angle was rotated by 180 • approximately.Considering the exo compounds, the conformations that occurred for DKI39 and DKI40 are shown in Schemes 6 and 7.The double bond, which is considered as exo for both compounds, is shown in a orange circle and it is the same for the eight different conformations.The relative energies are given in Table 3 for the exo isomorph, as well as the values of dihedral angles for both compounds.In the next section, the transition states are presented.
The dihedral angles of the minima were expected to be about 180 • and 0 • , however, there are deviations from these values due to stereochemical hindrances within the structures.The lowest energy conformations for DKI39 and DKI40 exo compounds are DKI39ex_1 and DKI40ex_1, respectively.To sum up, the DFT calculations show that the exo isomer is lower in energy than the endo one for both molecules.Specifically, among the exo calculated conformations, the DKI39ex_1 and DKI40ex_1 are the energetically more favorable isomers of the DKI39 and DKI40 compounds (Scheme 8).In particular, the endo conformer is higher than exo by 3 kcal/mol approximately.This was also confirmed because no spatial relationship was observed in 2D-NOESY between the NH group with the H-5 in both molecules (Figure 1).Also, in the exo conformers, there is an extended resonance which cannot be applied on the higher energy endo conformer.In Table 4, selected distances of the most energetical favorable conformations which have also been observed in 2D-NOESY are shown (Figure 1).The dihedral angles of the minima were expected to be about 180° and 0°, however, there are deviations from these values due to stereochemical hindrances within the structures.The lowest energy conformations for DKI39 and DKI40 exo compounds are DKI39ex_1 and DKI40ex_1, respectively.To sum up, the DFT calculations show that the exo isomer is lower in energy than the endo one for both molecules.Specifically, among the exo calculated conformations, the DKI39ex_1 and DKI40ex_1 are the energetically more The dihedral angles of the minima were expected to be about 180° and 0°, however, there are deviations from these values due to stereochemical hindrances within the structures.The lowest energy conformations for DKI39 and DKI40 exo compounds are DKI39ex_1 and DKI40ex_1, respectively.To sum up, the DFT calculations show that the exo isomer is lower in energy than the endo one for both molecules.Specifically, among the exo calculated conformations, the DKI39ex_1 and DKI40ex_1 are the energetically more

Isomerization Reaction Paths
The isomerization reaction paths, including the corresponding transition states that are involved in the reaction paths, were calculated [14].The cis-trans isomerization of DKI39 is plo ed in Figures 2 and 3

Isomerization Reaction Paths
The isomerization reaction paths, including the corresponding transition states that are involved in the reaction paths, were calculated [14].The cis-trans isomerization of DKI39 is plotted in Figures 2 and 3 and cis-trans isomerization of DKI40 is plotted in Figures 4 and 5.
The DFT calculations show that the energy gaps are quite high, i.e., larger than 20 kcal/mol, showing that there is only one dominant conformation for both compounds and it is difficult for the isomerization process from one isomer to other to occur.This conclusion is in agreement with our results obtained from our 2D-NMR spectra.

Molecular Binding
Molecular docking calculations were performed for both compounds in order to find some possible biological targets.From SwissADME [15], five possible targets were found.The grid parameters used were the same for all the substrates-X = 40, Y = 40, Z = 40 (default)-and the distance of the dots was 0.375 Å (default).
The coordinates for the active site for each macromolecule were: 2Y0J

Molecular Binding
Molecular docking calculations were performed for both compounds in order to find some possible biological targets.From SwissADME [15], five possible targets were found.The grid parameters used were the same for all the substrates-X = 40, Y = 40, Z = 40 (default)-and the distance of the dots was 0.375 Å (default).
The coordinates for the active site for each macromolecule were: 2Y0J  5. Regarding the results, it was observed that in all macromolecules, the compounds bind strongly to the active site.Only in macromolecule mglur3 compound did DKI40 not bind strongly to the active site.
In Figure 6, all binding modes are shown for both compounds.Firstly, DKI39 forms two hydrogen bonds with ILE682 and GLN716 and two π-π interactions with PHE719 and TYR683 of triazoloquinazolines. Also, DKI39 forms two hydrogen bonds with ARG68 and ARG277 and one π-π interaction with TYR222 of mglur3.Moreover, DKI39 forms one π-π interaction with TYR904 and two hydrogen bonds with LEU905 and GLU903 of Jak3.Furthermore, it forms two π-π interactions with HIS574 and PHE583 of Danio rerio HDAC6 CD2.Also, it forms one π-π interaction with TYR341 and two hydrogen bonds with VAL294 and GLU292 of acetylcholinesterase. and ARG277 and one π-π interaction with TYR222 of mglur3.Moreover, DKI39 forms one π-π interaction with TYR904 and two hydrogen bonds with LEU905 and GLU903 of Jak3.Furthermore, it forms two π-π interactions with HIS574 and PHE583 of Danio rerio HDAC6 CD2.Also, it forms one π-π interaction with TYR341 and two hydrogen bonds with VAL294 and GLU292 of acetylcholinesterase.
On the other hand, DKI40 forms two hydrogen bonds with GLN716 and HIS515 of triazoloquinazolines. Secondly, it forms two hydrogen bonds with ARG68 and ARG277 and one π-π interaction with TYR222 of mglur3.Furthermore, DKI40 forms two hydrogen bonds with VAL884 and LEU905 and one π-π interaction with TYR904 of Jak3.With the enzyme Danio rerio HDAC6 CD2, it does not form any interaction, while it forms one π-π interaction with TYR341 of acetylcholinesterase.

Molecular Dynamics
Molecular dynamics (MD) simulations were used to further elucidate binding of DKI39 and DKI40.Three replica simulations, each of duration of 300 ns, were performed for each system, starting from the optimized conformations derived from DFT calculations.Due to its wide applicability and efficiency, we used the end-point calculation with molecular mechanics/generalized born surface area (MM/GBSA) [21] method of the binding free energy difference between the ligand, the corresponding protein receptor and the On the other hand, DKI40 forms two hydrogen bonds with GLN716 and HIS515 of triazoloquinazolines. Secondly, it forms two hydrogen bonds with ARG68 and ARG277 and one π-π interaction with TYR222 of mglur3.Furthermore, DKI40 forms two hydrogen bonds with VAL884 and LEU905 and one π-π interaction with TYR904 of Jak3.With the enzyme Danio rerio HDAC6 CD2, it does not form any interaction, while it forms one π-π interaction with TYR341 of acetylcholinesterase.

Molecular Dynamics
Molecular dynamics (MD) simulations were used to further elucidate binding of DKI39 and DKI40.Three replica simulations, each of duration of 300 ns, were performed for each system, starting from the optimized conformations derived from DFT calculations.Due to its wide applicability and efficiency, we used the end-point calculation with molecular mechanics/generalized born surface area (MM/GBSA) [21] method of the binding free energy difference between the ligand, the corresponding protein receptor and the protein-ligand complex for both systems, ∆∆G bind = ∆G complex − ∆G protein − ∆G ligand , where ∆G complex , ∆G protein and ∆G ligand are solvation free energies of the complex, protein and ligand, respectively.∆∆Gbind values were derived for the sequential intervals of 100 ns for each simulation and are shown in Figure 7.According to the negative values of ∆∆G, the simulation results indicate stable binding in both systems during MD simulations with a bit stronger interaction for DKI40.

Pharmacokinetics and Toxicity of the Compounds
The results from the drug-likeness of the compounds are shown in Table 6.According to SwissADME and pkCSM, both compounds obey Lipinski's Rule of Five.[22] Due to the fact that rotable bonds are less than seven, the criterion for Veber's Rule [23] is also met.Both compounds can be easily absorbed from the body because lipophilicity is less than 5.

Pharmacokinetics and Toxicity of the Compounds
The results from the drug-likeness of the compounds are shown in Table 6.According to SwissADME and pkCSM, both compounds obey Lipinski's Rule of Five.[22] Due to the fact that rotable bonds are less than seven, the criterion for Veber's Rule [23] is also met.Both compounds can be easily absorbed from the body because lipophilicity is less than 5.Because the BBB [24] value (Table 7) is less than one, both compounds are inactive in the central nervous system (CNS).The value for human intestinal absorption is high for both compounds, and this signifies that these compounds might be better absorbed from the intestinal tract on oral administration.Both compounds are not inhibitors of CP isoenzymes and, therefore, are not toxic or do not exert other unwanted adverse effects.According to preADMET, these are not hepatotoxic [25] and they have no skin sensitivity.But, the AMES toxicity is positive for both of them and that means that they may be mutagenic, see Table 8.

Experimental
Reagents were purchased with the highest commercial quality from Aldrich (St. Louis, MI, USA), Acros (Geel, Belgium) and Fluka (Buchs, Switzerland) and were used without further purification.Reactions were monitored by thin-layer chromatography (TLC) carried out on 0.25 mm silica gel plates (E.Merck silica gel 60F254), and components were visualized by UV light absorbance.Purification of compounds was performed with filtration.Mobile phase for TLC experiments was conducted with an appropriate mixture of PE [refers to petroleum ether (40-60 • C)] and EA [ethyl acetate], as described for each compound.Electrospray ionization (ESI) mass spectral analyses were performed on a mass spectrometer MSQ Surveyor, Finnigan, using direct sample injection.Negative or positive ion ESI spectra were acquired by adjusting the needle and cone voltages accordingly.HRMS spectra were registered using a 4800 MALDI-TOF mass spectrometer (Applied Biosystems, Foster City, CA, USA) in the positive reflection mode in the m/z range of 100-700.

E-2-benzylidenehydrazinecarbothioamide, DKI1
To a mixture of benzaldehyde (4.0 mL, 40 mmol) in 95% ethanol (140 mL), thiosemicarbazide (3.65 g, 40 mmol) was added, followed by the addition of 4 drops of acetic acid.The reaction mixture was heated to 80 • C for 4 h.Then, to the reaction mixture, H 2 O (140 mL) was added and left to stand at 0 • C, the solid formed was filtered, washed with ethanol/water: 1/1 and dried to give DKI1 as white solid in 75% yield.R f (PE/EA: 2/1) = 0.69.Spectroscopic data are in accordance with the literature [26]. 1

(Z)-2-(((E)-benzylidene)hydrazono)thiazolidin-4-one, DKI39
To a mixture of thiazolidine-4-one DKI36 (219.3 mg, 1 mmol) in MeOH (2.5 mL), benzaldehyde (104 µL, 1 mmol) was added followed by piperidine (2 drops).The reaction mixture was warmed to 65 • C for 24 h, after which a new portion of benzaldehyde and piperidine was added, followed by 24 additional hours at 65 • C. The mixture was left to stand at room temperature, and the solid formed was filtered and washed with EtOH 95% to give DKI39 as a pale yellow solid in 84% yield.R f = 0.43 (PE/EA: 8:2).In a test tube, thiazolidinone DKI36 (219.3, 1 mmol), anisaldehyde (496 µL, 4 mmol) and piperidine (2 drops) were mixed and stirred for 24 h at room temperature.In 24 h, a new portion of anisaldehyde and piperidine were added and the mixture was stirred for 7 days.The new solid formed was filtered and washed with EtOH 95% to give DKI40 as a pale yellow solid in 31% yield.R f = 0.58 (PE/EA 2:1).

Structure NMR Elucidation
The identification for both molecules was carried out via NMR using a 400 MHz spectrometer.Specifically, 1D ( 1 H, 13 C) and 2D (2D-COSY, 2D-NOESY, 2D-HSQC, 2D-HMBC) NMR experiments were performed.All the experiments were conducted using DMSO-d 6 as a solvent at 25 • C. The pulse sequences were obtained from the library of the spectrometer and the process of the spectra was analyzed through MestreNova v. 14 program [28].

Computational Details
Both DKI39 and DKI40 molecules were energetically optimized.E and Z isomers, as well as endo and exo conformers, were computed employing the Density Functional Theory [29] at the B3LYP [30,31]/6-311 + G(d,p) [32] level of theory.All calculations were performed in DMSO solvent employing the polarizable continuum model (PCM) [33,34].It has been shown that the present DFT methodology is adequate for the prediction of different isomers and conformers [35][36][37][38].The transition states connecting the various minima structures were calculated using the synchronous transit-guided quasi-Newton (STQN) method [39].Vibrational analysis and IRC calculations confirmed that they are transition states.The calculations and the visualization of the results were performed through Gaussian 16 [40].

Molecular Binding
Docking [41] experiments were performed using Autodock [42] software and Lamarckian genetic algorithm.The crystal structures of the macromolecules were downloaded from the Protein Data Bank (PDB) platform and the compounds were minimized with MM2 force field.

Molecular Dynamics
Explicit solvent atomistic simulation systems were prepared from DFT-optimized structures of ligand-receptor complexes using the online server CHARMM-GUI [43].Both systems were electroneutralized in 150 mM NaCl, and the TIP3P [44] water model was used.We chose a cubic simulation box with the side length of 8.5 nm.Systems were energy-minimized using the steepest descent and the adaptive basis Newton-Raphson method.Both systems were exposed to an equilibration phase of 10 ns.All simulations were carried out on GPUs with the CUDA version of the NAMD [45] (ver.2.13) MD software suite.The CHARMM36 [46] force field was used, which included parameters for protein, water and ions.Missing parameters for the ligands were derived using the Paramchem [47] server, while the partial atomic charges were further refined by ab initio calculation performed by the Gaussian suit of programs [48].All production simulations used the constant number of particles, pressure and temperature (NPT) ensemble and each run was 300 ns long.Temperature was held constant (303.15K) using the Langevin thermostat with a dampening constant of 1 ps −1 , and the pressure was held constant at 1 bar using the Nose-Hoover Langevin piston for pressure control.The time-step of the production simulations was 2 fs and the SHAKE algorithm was used for hydrogen atoms.
The cutoff for non-bonded interactions was set to 12 Å, and electrostatic interactions were calculated using the Particle Mesh Ewald method [49].

Pharmacokinetic and Toxicity Properties
Both compounds were examined for their drug-likeness and toxicity through Swis-sADME [15], pkCSM [50] and preADMET [51].This is very important because many biological compounds fail due to unfavorable ADME results.

Conclusions
Scheme 9 includes all the research activities applied in the manuscript.In silico experiments of molecular docking were performed on five macromolecules, i.e., triazoloquinazolines, mglur3, Jak3, Danio rerio HDAC6 CD2 and acetylcholinesterase.The results showed that the compounds bind strongly to them and both compounds obey Lipinski's Rule of Five and they are not hepatotoxic.The molecular dynamics data showed that both compounds are stable in acetylcholinesterase and in metabotropic glutamate receptor 3 for DKI40 and DKI39, respectively.
To sum up, the present study showed that thiazolidine-4-ones are stable, safe and bioactive in various targets.This study aimed to investigate the physical and chemical properties of two synthetic analogues and examine the possibility of them serving as lead compounds for various diseases.These conclusions are quite important for synthetic organic scientists who synthesize similar derivatives and test them as potential biological targets.

Supplementary Materials:
The following supporting information can be downloaded at: www.mdpi.com/xxx/s1, Figure S1A 1S: Energetic isomerization between cis-trans for DKI39exo.;Table 2S: Energetic isomerization between cis-trans for DKI40exo.In particular, the present study focused on the synthesis of two thiazolidine-4-ones and on their study via DFT methodology and NMR spectroscopy.The aim of this study was to decipher which isomers and conformers are the most stable ones, as well as to check if the isomers or conformers can easily convert one to another.
DFT showed that these two molecules can form stable endo and exo conformations, but the exo ones are the most stable in energy, in agreement with our 2D-NMR measurements.The calculations of the reaction isomerization paths, and the calculation of the corresponding transition states, confirm our NMR results that for the exo compounds, only one conformation is synthesized.
In silico experiments of molecular docking were performed on five macromolecules, i.e., triazoloquinazolines, mglur3, Jak3, Danio rerio HDAC6 CD2 and acetylcholinesterase.The results showed that the compounds bind strongly to them and both compounds obey Lipinski's Rule of Five and they are not hepatotoxic.The molecular dynamics data showed that both compounds are stable in acetylcholinesterase and in metabotropic glutamate receptor 3 for DKI40 and DKI39, respectively.
To sum up, the present study showed that thiazolidine-4-ones are stable, safe and bioactive in various targets.This study aimed to investigate the physical and chemical properties of two synthetic analogues and examine the possibility of them serving as lead compounds for various diseases.These conclusions are quite important for synthetic organic scientists who synthesize similar derivatives and test them as potential biological targets.

Scheme 3 .
Scheme 3. Optimized conformations derived from DFT calculations for the DKI39 endo compound.In the Figure the dihedral angles rotated by 180° are shown.For instance, in DKI39en_1, the dihedral angle τ1 was rotated and DKI39en_2 was obtained.In the orange circle, the double bond that indicates the endo nature of compound is shown.

Scheme 4 .
Scheme 4. Optimized conformations derived from DFT calculations for the DKI40 endo compound.In the Figure, the dihedral angles rotated by 180° are shown.For instance, in DKI40en_1, the dihedral angle τ3′ was rotated and DKI39en_2 was obtained.In the orange circle, the double bond that indicates the endo nature of compound is shown.

Scheme 3 .
Scheme 3. Optimized conformations derived from DFT calculations for the DKI39 endo compound.In the Figure the dihedral angles rotated by 180° are shown.For instance, in DKI39en_1, the dihedral angle τ1 was rotated and DKI39en_2 was obtained.In the orange circle, the double bond that indicates the endo nature of compound is shown.

Scheme 4 .
Scheme 4. Optimized conformations derived from DFT calculations for the DKI40 endo compound.In the Figure, the dihedral angles rotated by 180° are shown.For instance, in DKI40en_1, the dihedral angle τ3′ was rotated and DKI39en_2 was obtained.In the orange circle, the double bond that indicates the endo nature of compound is shown.

Scheme 4 .
Scheme 4. Optimized conformations derived from DFT calculations for the DKI40 endo compound.In the Figure, the dihedral angles rotated by 180 • are shown.For instance, in DKI40en_1, the dihedral angle τ 3 ′ was rotated and DKI39en_2 was obtained.In the orange circle, the double bond that indicates the endo nature of compound is shown.

20 Scheme 6 .Scheme 7 .
Scheme 6. Optimized conformations derived from DFT calculations for DKI39 exo compound.In the Figure the dihedral angles rotated by 180° are shown.For instance, in DKI39ex_1, the dihedral angle τ3 was rotated and DKI39ex_2 was obtained.In the orange circle, the double bond that indicates the exo nature of compound is shown.

Scheme 6 . 20 Scheme 6 .Scheme 7 .
Scheme 6. Optimized conformations derived from DFT calculations for DKI39 exo compound.In the Figure the dihedral angles rotated by 180 • are shown.For instance, in DKI39ex_1, the dihedral angle τ 3 was rotated and DKI39ex_2 was obtained.In the orange circle, the double bond that indicates the exo nature of compound is shown.

Scheme 7 .Table 3 . 20 Figure 1 .
Scheme 7. Optimized conformations derived from DFT calculations for the DKI40 exo compound.In the Figure, the dihedral angles rotated by 180 • are shown.For instance, in DKI40ex_1, the dihedral angle τ 2 ′ was rotated and DKI40ex_2 was obtained.In the orange circle the double bond that indicates the exo nature of compound is shown.
and cis-trans isomerization of DKI40 is plo ed in Figures 4 and 5.

Figure 2 .
Figure 2. Schematic representation of the isomerization energy barrier between cis-trans for DKI39 (y = energy in kcal/mol, x = conformation).

Figure 2 .
Figure 2. Schematic representation of the isomerization energy barrier between cis-trans for DKI39 (y = energy in kcal/mol, x = conformation).

Figure 2 .
Figure 2. Schematic representation of the isomerization energy barrier between cis-trans for DKI39 (y = energy in kcal/mol, x = conformation).

Figure 3 .
Figure 3. Schematic representation of cis-trans kinetic isomerization for DKI39 exo compound (y = energy in kcal/mol, x = conformation).In each transition state, the dihedral angle, which is rotated by ~180°, is shown by an arrow.

.
Figure 3. Schematic representation of cis-trans kinetic isomerization for DKI39 exo compound (y = energy in kcal/mol, x = conformation).In each transition state, the dihedral angle, which is rotated by ~180 • , is shown by an arrow.Molecules 2024, 29, x FOR PEER REVIEW 10 of 20

Figure 4 .
Figure 4. Schematic representation of the isomerization energy barrier between cis −trans for DKI40 (y = energy in kcal/mol, x = conformation).

Figure 4 .
Figure 4. Schematic representation of the isomerization energy barrier between cis −trans for DKI40 (y = energy in kcal/mol, x = conformation).

Figure 4 .
Figure 4. Schematic representation of the isomerization energy barrier between cis −trans for DKI40 (y = energy in kcal/mol, x = conformation).

Figure 5 .
Figure 5. Schematic representation of cis-trans kinetic isomerization for DKI40 exo compound (y = Energy in kcal/mol, x = conformation).In each transition state, the dihedral angle, which is rotated by ~180°, is shown by arrow.

ConformationFigure 5 .
Figure 5. Schematic representation of cis-trans kinetic isomerization for DKI40 exo compound (y = Energy in kcal/mol, x = conformation).In each transition state, the dihedral angle, which is rotated by ~180 • , is shown by arrow.

Figure 7 .
Figure 7.Comparison of binding free energies during repeated simulations (3 replicas) of DKI39 and DKI40 complexes in aqueous solution.

Figure 7 .
Figure 7.Comparison of binding free energies during repeated simulations (3 replicas) of DKI39 and DKI40 complexes in aqueous solution.

Molecules 2024 ,Scheme 9 .
Scheme 9. Summary of the scientific work described in the manuscript.
: 1 H spectra of DKI39.The spectra were recorded in DMSO-d6 on a Bruker AC 800 MHz spectrometer at 25 °C.; Figure S2A: 2D-NOESY spectra of DKI39.The spectra were recorded in DMSO-d6 on a Bruker AC 800 MHz spectrometer at 25 °C.; Figure S3A: 13 C spectra of DKI39.The spectra were recorded in DMSO-d6 on a Bruker AC 800 MHz spectrometer at 25 °C.; Figure S4A: 2D-HSQC spectra of DKI39.The spectra were recorded in DMSO-d6 on a Bruker AC 800 MHz spectrometer at 25 °C.; Figure S5A: 2D-HSQC spectra of DKI39.The spectra were recorded in DMSO-d6 on a Bruker AC 800 MHz spectrometer at 25 °C.; Figure S1B: 1H spectra of DKI40.The spectra were recorded in DMSO-d6 on a Bruker AC 800 MHz spectrometer at 25 °C.; Figure S2B: 13C spectra of DKI40.The spectra were recorded in DMSO-d6 on a Bruker AC 800 MHz spectrometer at 25 °C.; Figure S3B: 2D-NOESY spectra of DKI40.The spectra were recorded in DMSO-d6 on a Bruker AC 800 MHz spectrometer at 25 °C.; Figure S4B: 2D-HSQC spectra of DKI40. Figure S5B: 2D-HMBC spectra of DKI40.The spectra were recorded in DMSO-d6 on a Bruker AC 800 MHz spectrometer at 25 °C.;Table

Scheme 9 .
Scheme 9. Summary of the scientific work described in the manuscript.

Table 1 .
Assignment of the experimental 1 H NMR spectra of

Table 1 .
Assignment of the experimental 1 H NMR spectra of

Table 4 .
Selected H-H distances in Å at DFT.
Scheme 8.The lowest in energy conformations for (a) DKI39 and (b) DKI40 exo compounds.

Table 4 .
Selected H-H distances in Å at DFT.

Table 5 .
Molecular docking results for each compound with the macromolecules.

Table 6 .
Drug-likeness of the compounds.

Table 6 .
Drug-likeness of the compounds.

Table 8 .
Toxicity results for both compounds according to pKCSm.