Development, Molecular Docking, and In Silico ADME Evaluation of Selective ALR2 Inhibitors for the Treatment of Diabetic Complications via Suppression of the Polyol Pathway.

Diabetic complications are associated with overexpression of aldose reductase, an enzyme that catalyzes the first step of the polyol pathway. Osmotic stress in the hyperglycemic state is linked with the intracellular accumulation of sorbitol along with the depletion of NADPH and eventually leads to oxidative stress via formation of reactive oxygen species and advanced glycation end products (AGEs). These kinds of mechanisms cause the development of various diabetic complications including neuropathy, nephropathy, retinopathy, and atherosclerotic plaque formation. Various aldose reductase inhibitors have been developed to date for the treatment of diabetic complications, but all have failed in different stages of clinical trials due to toxicity and poor pharmacokinetic profiles. This toxicity is rooted in a nonselective inhibition of both ALR2 and ALR1, homologous enzymes involved in the metabolism of toxic aldehydes such as methylglyoxal and 3-oxyglucosazone. In the present study, we developed a series of thiosemicarbazone derivatives as selective inhibitors of ALR2 with both antioxidant and antiglycation potential. Among the synthesized compounds, 3c exhibited strong and selective inhibition of ALR2 (IC50 1.42 μM) along with good antioxidant and antiglycative properties. The binding mode of 3c was assessed through molecular docking and cluster analysis via MD simulations, while in silico ADME evaluation studies predicted the compounds' druglike properties. Therefore, we report 3c as a drug candidate with promising antioxidant and antiglycative properties that may be useful for the treatment of diabetic complications through selective inhibition of ALR2.

Introduction

Normoglycemia is typified by low metabolic flux through the polyol pathway. This is because glucose-6-phosphate (the predominant glucose species under these conditions) has, relative to glucose, a poor affinity for aldose reductase (ALR2, AKR1B1; EC 1.1.1.21), the enzyme that catalyzes the first step of the pathway. This changes when an individual enters the hyperglycemic state, however, since high blood glucose concentrations provide ALR2 with a plentiful substrate. Upon entering the polyol pathway, glucose is first converted to sorbitol through the catalytic action of ALR2 and then subsequently converted to fructose through the action of a second enzyme, sorbitol dehydrogenase. The resulting increased production of sorbitol, an osmolyte, causes depletion of cellular stores of NADPH, which in turn increases the susceptibility of cells to damage by reactive oxygen species (ROS) and causes oxidative and osmotic stress. This, combined with the production of advanced glycation end products (AGEs) as driven by the increased production of fructose, can lead to diabetic complications including neuropathy, nephropathy, retinopathy, and atherosclerotic plaque formation.[1,2] ALR2 also plays a role in reducing various lipid peroxidation-derived aldehydes and the associated glutathione conjugates. Inflammatory signals initiated by endotoxins, cytokines, autoimmune reactions, growth factors, hyperglycemia, and allergens have been found to be prevented by inhibition of ALR2 in different cells and animal models.[3,4]

Various aldose reductase inhibitors have been investigated as potential drug candidates for the treatment of diabetic complications. However, a major problem discovered while developing these drugs is toxicity related to the nonselective inhibition of ALR2. This nonselective inhibition is associated with the interaction of these compounds with ALR1, another member of the aldo–keto reductase superfamily and an isoform of ALR2 with approximately 65% structural similarity that is involved in the metabolism of toxic lipid peroxidation products like methylglyoxal and 3-oxyglucosazone.[5] Thus, despite the development of several potent ALR2 inhibitors from both synthetic and natural sources,[6,7] few such compounds have entered clinical trials, and epalrestat remains the sole approved treatment for diabetic complications, approved for use in China, India, and Japan.[8,9]

The newly explored role of aldose reductase in various inflammatory disorders has sparked renewed interest in the development of ALR2 inhibitors.[10] As part of this exploration of novel ALR2 inhibitors, more emphasis is being placed on improving pharmacokinetic profiles and considering the bioavailability of the inhibitor at desired target sites.[11,12] Well-studied classes of the ALR2 inhibitor include acetic acid derivatives, spiro-hydantoins, and succinimide derivatives.[9,13,14]

The design and development of ALR2 inhibitors with good physicochemical parameters, higher target selectivity, and fewer adverse effects is the most demanding aspect of drug discovery for diabetic complications. We have previously reported selective inhibitors of ALR2 such as coumarin–thiosemicarbazone hybrids,[17] benzoxazinone–thiosemicarbazones,[19] and adamantyl–thiosemicarbazones.[18] Various substituted thiosemicarbazones and Schiff base derivatives have been reported for their potential therapeutic roles.[20−22] Therefore, in the present study, we have developed a series of phenol-substituted thiosemicarbazones as inhibitors of ALR2. The phenolic structure of these compounds was specifically incorporated with the intention of achieving good antioxidant properties to grant them an enhanced ability to treat diabetic complications.

Results and Discussion

Chemistry

A series of 16 novel 2-hydroxy-5-methylbenzaldehyde-based thiosemicarbazones (3a–p) bearing aryl and cyclohexyl substituents were synthesized by condensation of various N4-substituted thiosemicarbazides (1a–p) with 2-hydroxy-5-methylbenzaldehyde (2). The reaction conditions were optimized by treating N-phenylhydrazinecarbothioamide (1a) and 2 in an equimolar ratio using solvents of variable polarity such as methanol, ethanol, dimethyl sulfoxide (DMSO), tetrahydrofuran, and dichloromethane. Methanol was found to be the best solvent for the reaction along with a catalytic amount of glacial acetic acid. Target thiosemicarbazones 3a–p were purified by recrystallization from ethanol in good to excellent yields (78–90%) (Scheme ).

Scheme 1

Synthesis of Thiosemicarbazone Derivatives 3a–p

The structures of the newly synthesized derivatives 3a–p were confirmed using elemental analysis and various spectroscopic techniques including IR, 1H NMR, and 13C NMR. Data from the IR spectra (a new band due to the new azomethine linkage (C=N) observed in the range of 1577–1615 cm–1), the 1H NMR spectra (a new singlet due to the azomethine proton observed in the range of 7.19–8.47 ppm), and the 13C NMR spectra (a new peak observed at 155.2 ppm matching the chemical shift for an azomethine carbon) together indicated the successful formation of an azomethine group in each product. This approach was in contrast to our other recent publications where we used SC-XRD to confirm product structures.[23−25] Other notable IR spectrum observations were the C=S stretching in the range of 1192–1219 cm–1, N–H stretching in the range of 3210–3357 cm–1, and O–H stretching in the range of 3390–3455 cm–1. In 1H NMR, protons of the NH-CS and NH-N moieties were observed at variable chemical shifts (6.53–9.74 ppm and 6.94–10.05 ppm, respectively), and the most downfield signals in each spectrum were due to the OH groups (9.78–11.90 ppm). The 13C NMR spectral data also fully confirmed the structures of the target thiosemicarbazone derivatives (Figure ).

Figure 1

Classes of previously reported ALR2 inhibitors include the cyclic imides,[15] acetic acid derivatives,[11,16] and thiosemicarbazones.[17,18]

Biological Activities and Structure–Activity Relationship

ALR1 and ALR2 in vitro enzyme inhibition data for compounds 3a–p revealed different trends of enzyme inhibition. IC50 values and percent inhibition data are presented in Table . Compounds 3b, 3c, 3g, 3j, 3k, 3l, 3n, 3o, and 3p displayed strong and selective inhibition of ALR2, whereas 3a, 3i, 3h, and 3m exhibited strong but nonselective inhibition of ALR1 and ALR2. Among the selective inhibitors, 3c was the most potent and selective inhibitor of ALR2 with an IC50 value of 1.42 μM. Compounds 3d, 3e, and 3f were moderate inhibitors of both ALR1 and ALR2, and their percent inhibition of each isozyme at 100 μM was less than 50%.

Table 1

IC50 Values of 3a–p for Inhibition of ALR1 and ALR2

compounds	IC50 (μM) ALR2 ± SEM	IC50 (μM) ALR1 ± SEM	% FRSAc	% inhibition of AGEsd
3a	4.99 ± 0.035	5.13 ± 0.037	56.78	59.12
3b	2.55 ± 0.021	24.67%	63.24	50.78
3c	1.42 ± 0.024	38.23%	65.67	66.40
3d	28.45%	24.45%	61.34	62.81
3e	42.81%	31.66%	65.66	56.73
3f	41.46%	37.29%	57.84	59.86
3g	8.37 ± 0.062	23.82%	53.56	51.33
3h	3.8 ± 0.072	15.80 ± 0.027	64.94	46.33
3i	2.06 ± 0.057	2.14 ± 0.031	71.34	64.41
3j	4.95 ± 0.066	33.63%	66.57	61.60
3k	14.03 ± 0.047	17.78%	53.88	40.08
3l	3.80 ± 0.027	35.11%	62.44	56.61
3m	1.96 ± 0.029	4.57 ± 0.076	63.87	53.08
3n	3.11 ± 0.038	41.47%	55.93	68.66
3o	12.15 ± 0.083	39.85%	68.56	38.91
3p	10.67 ± 0.064	27.78%	71.53	64.81
Sorbinila	2.18 ± 0.002
valproic acidb		49.31 ± 0.005
6-AGe				79.34

Standard inhibitor of ALR2.

Standard inhibitor of ALR1.

Percent free-radical scavenging activity.

Percent inhibition of advanced glycation end-product formation.

6-Aminoguanidine (standard antiglycative agent).

From this data, the structure–activity relationship for the tested compounds was established. It was observed that substituents/moieties attached to the thiosemicarbazone backbone exerted varying effects on enzyme inhibition and selectivity. Compound 3a, which possesses an unsubstituted phenyl moiety, showed strong but nonselective inhibition of ALR1/ALR2, whereas compound 3b, which possesses two weakly electron-donating methyl groups at the 2- and 4-positions of the phenyl ring, displayed selective inhibition of ALR2 (IC50 2.55 μM). The most potent and selective ALR2 inhibitor of the series, compound 3c, possesses a 3-methoxy group on the phenyl ring (IC50 1.42 μM), while the potent nonselective ALR1/ALR2 inhibitors 3i (2,6-dimethyl substitution pattern, ALR2 IC50 2.06 μM, ALR1 IC50 2.14 μM; Figure ) and 3m (2-methyl substitution pattern, ALR2 IC50 1.96 μM, ALR1 IC50 4.57 μM) possess either one or two ortho methyl substituents. Compound 3c showed superior ALR2 inhibition to sorbinil (IC50 2.18 μM) with significant selectivity for ALR2 over ALR1, indicating that a strongly electron-donating methoxy group in a meta-position on the phenyl moiety was beneficial for achieving selective inhibition of ALR2. The weak inhibition of each enzyme was correlated with the presence of a benzyl moiety (compounds 3d [Figure ] and 3e) or cyclohexyl group (compound 3f) in place of the aforementioned phenyl ring.

Figure 2

Structure–activity relationship for the most potent/selective ALR2 inhibitor (3c), most potent nonselective ALR1/ALR2 inhibitor (3i), and weakest inhibitor (3d).

Free-Radical Scavenging and Antiglycation Activity

The percent free-radical scavenging activity of compounds 3a–p was determined by determining DPPH (2,2-diphenyl-1-picryl-hydrazyl-hydrate) activity. All compounds exhibited strong antioxidant activity with >50% free-radical scavenging activity (Table ), most likely due to the presence of a phenolic moiety in the chemical structures of 3a–p.

Antiglycation activity was determined using a bovine serum albumin glycation assay and following a previously reported method with few modifications.[26] The synthesized compounds 3a–p were assessed for antiglycation activity and the majority exhibited strong activity as shown in Table . Compounds 3h, 3k, and 3o showed moderate antiglycation activity with less than 50% inhibition of formation of advanced glycation end products (AGEs), whereas all other derivatives possessed more than 50% inhibition of formation of AGEs (Figure ).

Figure 3

Antiglycation activity of compounds 3a–p. AG (aminoguanidine) was used as a positive control (100 μM) and VC (vehicle control) as a negative control. The percent inhibition of AGEs is shown on the y-axis and compounds 3a–p are shown along the x-axis. The results are presented as the mean value ± S.E.M where n = 3 (three independent incubations). The Student t-test was applied; ****p ≤ 0.0001, ***p ≤ 0.001, **p ≤ 0.01, and *p ≤ 0.05 vs AG.

Molecular Docking Studies and Molecular Dynamics Simulations

Following the identification of compound 3c as the most potent selective inhibitor of ALR2 through the in vitro enzyme inhibition assay, molecular docking studies were carried out to investigate the interactions of 3c with the amino acid residues of the ALR2 active site.

Prior to running the docking simulation, redocking of ALR2 (PDB ID: 1US0) with a known inhibitor LDT320 (reference ligand) was carried out to validate the process (RMSD value 0.69 Å). Compound 3c was then docked against ALR2 and showed similar interactions with the ALR2 active pocket to the cocrystallized ligand (LDT320). A HYDE binding assessment was carried out and the binding free energy (ΔG) for 3c was calculated as −37 kJ mol–1.

As shown in Figure , compound 3c was predicted to show various types of interactions with the ALR2 active site including hydrophobic interactions and conventional hydrogen bonding. The −NH group of amino acid Trp111 was predicted to form a hydrogen bond with the phenolic oxygen of 3c while distinct hydrophobic interactions were predicted for the methylbenzene (with amino acid residues Val47, Tyr48, and Phe122) and methoxybenzene (with amino acid residues Trp111, Trp219, Cys298, Ala299, and Leu300) moieties of 3c. The Schiff base moiety (−C=N−) was predicted to form a van der Waals interaction with residue Trp79.

Figure 4

(a) Compound 3c docked within ALR2 active site. (b) Two-dimensional (2D) interactions of 3c with ALR2.

The top-ranked binding pose of compound 3c inside the ALR2 active site from the molecular docking study was subjected to molecular dynamics (MD) simulations. MD simulations were carried out in the presence of a cofactor. The overall protein–cofactor–inhibitor system was solvated and neutralized with counter ions using a periodic boundary condition (PBC) water box. Root-mean-square deviation (RMSD) values, clustering, and the binding free energies (using MMGBSA calculations) of the inhibitor were calculated.

A 50 ns simulation trajectory was visually observed using visual molecular dynamics. Trajectory analysis revealed different orientations of the inhibitor inside the active site of the enzyme. To determine the most probable pose, cluster analysis was performed as shown in Figure . Clustering of the inhibitor orientation revealed six different orientations inside the active pocket. The top-ranking pose from each individual cluster was visually observed (Figure ).

Figure 5

Cluster analysis of compound 3c inside the ALR2 active pocket.

Free energy of the ligand binding to the receptor was calculated using gmx_MMPBSA using the whole 50 ns trajectory. Contributions of the receptor, inhibitor, and the complex were determined and the change in binding free energy (ΔE) was determined. This revealed a varying pattern over the course of the trajectory. The varying pattern in the binding free energy corresponds to the varying inhibitor poses inside the active site throughout the simulation time and is evident from the clustering pattern. The average value of binding was found to be around −25 kJ mol–1 for the initial 5 ns of simulation, which corresponded to the poses obtained in the case of cluster 2. After around 5 ns of simulations, the pattern varied due to the varying conformations of the inhibitor pose, and the binding affinity was lower than that initially found for the second cluster (see Figure for energy contributions).

Figure 6

Binding free energy (ΔE) contributions of the (a) complex, (b) receptor, and (c) inhibitor ligand and (d) the change in binding free energy with time (ΔE).

The most probable binding mode of compound 3c was selected based on the cluster analysis and the binding free energy values obtained through MMGBSA calculations. Compound 3c was found to form hydrogen bonds with residue Trp111 of the anionic binding site and to form disulfide bridges with residue Cys303 through its sulfide group. van der Waals interactions between residue Cys80 and an −NH unit of the thiosemicarbazide moiety were also observed. Several hydrophobic interactions between 3c and the anionic binding pocket were also observed (Figure ).

Figure 7

(a) Binding pose of compound 3c inside the ALR2 active pocket (and the representative pose of cluster 2). (b) Compound 3c poses obtained through clustering the MD ensemble.

In Silico ADME Evaluation

SwissADME web-based software was used to predict ADME properties of compounds 3a–pin silico. As shown in Table , all compounds were compliant with Lipinski’s rule of five, each possessing 3 hydrogen bond donors, no more than 4 hydrogen bond donors, molecular weights less than 400 g mol–1, and log P values lower than 4. The topological polar surface area (TPSA) of the compounds was estimated to be 88.74–97.97 Å2, and these values were used to construct a boiled egg plot. All synthesized compounds were predicted to exhibit a druggable character and excellent gastrointestinal absorption properties.

Table 2

In Silico ADME Evaluation of Compound 3c

codes	MWa	H-bond acceptors	H-bond donors	TPSAb	WLOGP	GI absorption	BBB permeantc	Lipinski violations
3c	315.39	3	3	97.97	2.84	high	no	0

Molecular weight.

Topological polar surface area.

Blood brain barrier (BBB).

Conclusions

Thiosemicarbazone derivatives bearing phenolic moieties were synthesized as potential selective inhibitors of aldose reductase (ALR2). Among the tested inhibitors, compounds 3b, 3c, 3g, 3j, 3l, and 3n were identified as selective inhibitors for ALR2 over its isoform ALR1, and each exhibited IC50 values in the low micromolar range. 3c was found to be the most potent and selective inhibitor of the series with an ALR2 IC50 value of 1.42 μM. In addition, it exhibited strong antioxidant activity (65.67% free-radical scavenging activity) and antiglycation activity (66.40% inhibition of formation of advanced glycation end products). In vitro ALR2 inhibition results were further investigated through molecular docking and molecular dynamics simulations to assess binding interactions. Finally, an in silico evaluation of ADME properties predicted promising pharmacokinetic profiles for all synthesized compounds. Therefore, compound 3c can be considered a druggable lead candidate for a drug discovery program to identify a treatment for diabetic complications.

Experimental Work

Materials and Methods

For the preparation of enzyme ALR2, the expression plasmid (pDONR223_AKR1B1_WT) was obtained as a gift from Jesse Boehm, Matthew Meyerson, and David Root (www.addgene.org; Addgene plasmid # 82928; http://n2t.net/addgene:82928; RRID: Addgene_82928). In the enzyme inhibition studies, substrates for ALR2 (d,l-glyceraldehyde) and ALR1 (sodium-d-glucuronate) as well as the nicotinamide adenine dinucleotide phosphate (NADPH) cofactor were purchased from Sigma Aldrich. For the synthesis of phenolic-based thiosemicarbazones, all starting materials such as 2-hydroxy-5-methylbenzaldehyde were purchased from Sigma Aldrich. Solvents and chemicals including methanol, ethanol, petroleum ether, glacial acetic acid, and ethyl acetate were purchased from Merck and used in their original forms. Silica gel plates backed with aluminum were used to monitor the progress of reactions and product formation. A Bruker Vector-22 spectrometer was used for FTIR analysis of the synthesized compounds in the 4000–500 cm–1 range. A Bruker Ascend 400 MHz NMR spectrometer was used to obtain 1H and 13C NMR spectra in deuterated solvents like CDCl3 and DMSO-d6 at 25 °C (400 MHz for 1H and 100 MHz for 13C). NMR spectra were reported in the form of chemical shifts (ppm), and coupling constants (J) were expressed in Hertz (Hz) to detail signal multiplicity. LC-MS analysis was performed on an Agilent Infinity Lab LC/MSD System consisting of an Agilent 1290 Infinity II Analytical-Scale LC Purification System coupled to a 6120 Quadrupole mass spectrometer. The elemental analysis was performed on a 2400 CHNS Organic elemental analyzer 100 V (Perkin Elmer).

Synthesis of Thiosemicarbazones (3a–p)

Briefly, 1 mmol of 2-hydroxy-5-methylbenzaldehyde (2) was added to 10 mL of methanol in an oven-dried round-bottomed flask along with 1 mmol of the appropriate N4-substituted thiosemicarbazide (1a–p). A few drops of glacial acetic acid, a catalyst for the reaction, were added. This mixture was refluxed for 2–3 hours and monitored by thin-layer chromatography until the reaction was deemed complete. Then, the reaction mixture was cooled to room temperature, at which point the product precipitated out of the solution. This precipitate was filtered, washed with methanol ×3 and diethyl ether ×2, and then purified by recrystallization from ethanol to yield the thiosemicarbazone products 3a–p. Characterization data of each compound synthesized is given below.

2-(2-Hydroxy-5-methylbenzylidene)-N-phenylhydrazinecarbothioamide (3a)

Yellow solid, yield 82%; melting point (mp) 224–226 °C; IR υ (cm–1) 1198 (C=S), 1590 (C=N), 3295 (N–H), 3410 (OH); 1H NMR (400 MHz, DMSO-d6) δ 2.23 (s, 3H, CH3), 6.84 (d, 1H, J = 8.4 Hz, Ar-H), 6.99–7.00 (d, 1H, J = 2.0 Hz, Ar-H), 7.07–7.09 (m, 1H, Ar-H), 7.22–7.24 (m, 1H, Ar-H), 7.35 (t, 2H, J = 7.6 Hz, Ar-H), 7.50 (d, 2H, J = 7.2 Hz, Ar-H), 8.02 (s, 1H, N=CH), 8.40 (s, 1H, NH-CS), 9.04 (s, 1H, NH-N), 10.12 (bs, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 20.2 (CH3), 109.9, 116.5, 116.7, 125.0, 126.8, 129.0, 129.6, 131.7, 133.5, 137.3 (Ar-C), 155.2 (CH=N), 175.32 (C=S); Anal. calcd for C15H15N3OS (285.36); C, 63.13; H, 5.30; N, 14.73, found C, 63.21; H, 5.39; N, 14.64. LC-MS (m/z: ESI+) calcd for C15H15N3OS [M]+, 285.09 found 286.1 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 97.1%.

N-(2,4-Dimethylphenyl)-2-(2-hydroxy-5-methylbenzylidene)hydrazinecarbothioamide (3b)

White solid, yield 87%; mp 213–215 °C; IR υ (cm–1) 1212 (C=S), 1577 (C=N), 3271, 3321 (N–H), 3418 (OH); 1H NMR (DMSO-d6) δ 2.23 (s, 6H, CH3), 2.28 (s, 3H, CH3), 6.82 (d, 1H, J = 8.4 Hz, Ar-H), 6.99–7.09 (m, 4H, Ar-H), 7.32 (d, 1H, J = 7.6 Hz, Ar-H), 8.03 (s, 1H, N=CH), 8.10 (s, 1H, NH-CS), 9.19 (s, 1H, NH-N), 10.25 (s, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 17.9(CH3), 20.2 (CH3), 21.1 (CH3), 115.9, 116.6, 116.7, 122.2, 127.4, 127.7, 129.5, 131.6, 131.7, 133.4, 137.2, 149.2 (Ar-C), 155.2 (CH=N), 175.5 (C=S); Anal. calcd for C17H19N3OS (313.42); C, 65.15; H, 6.11; N, 13.41, found C, 65.24; H, 6.02; N, 13.49. LC-MS (m/z: ESI+) calcd for C17H19N3OS [M]+, 313.12 found 314.1 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 99.2%.

2-(2-Hydroxy-5-methylbenzylidene)-N-(3-methoxyphenyl)hydrazinecarbothioamide (3c)

Yellow solid, yield 83%; mp 234–236 °C, IR υ (cm–1) 1209 (C=S), 1615 (C=N), 3257, 3319 (N–H), 3390 (OH); 1H NMR (DMSO-d6) δ 2.22 (s, 3H, CH3), 3.75 (s, 3H, OCH3), 6.74–6.76 (m, 1H, Ar-H), 6.81 (d, 1H, J = 8.4 Hz, Ar-H), 6.99–7.02 (m, 2H, Ar-H), 7.06–7.09 (m, 1H, Ar-H), 7.21–7.25 (m, 2H, Ar-H), 8.01 (s, 1H, N=CH), 8.38 (s, 1H, NH-CS), 9.03 (s, 1H, NH-N), 10.13 (s, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 20.2 (CH3), 55.4(OCH3), 110.5, 112.6, 116.5, 116.7, 117.0, 129.2, 129.6, 129.6, 131.7, 133.5, 138.4, 160.0 (Ar-C), 155.2 (CH=N), 175.0 (C=S); Anal. calcd for C16H17N3O2S (315.39); C, 60.93; H, 5.43; N, 13.32, found C, 60.84; H, 5.52; N, 13.42.LC-MS (m/z: ESI+) calcd for C16H17N3O2S [M]+, 315.1 found 316.1 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 99.1%.

N-(4-Chlorobenzyl)-2-(2-hydroxy-5-methylbenzylidene)hydrazinecarbothioamide (3d)

White solid, yield 89%; mp 208–210 °C, IR υ (cm–1) 1192 (C=S), 1591 (C=N), 3348 (N–H), 3398 (OH); 1H NMR (CDCl3) δ 2.21 (s, 3H, CH3), 4.84 (d, 2H, J = 5.6 Hz, CH2), 6.78 (d, 1H, J = 8.4 Hz, Ar-H), 6.95 (s, 1H, NH-CS), 7.05 (d, 1H, J = 8.4 Hz, Ar-H), 7.19–7.21 (m, 2H, Ar-H), 7.24–7.27 (m, 3H, Ar-H), 7.93 (s, 1H, N=CH), 8.98 (s, 1H, NH-N), 9.78 (s, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 20.2 (CH3), 47.9 (CH2), 116.4, 116.6, 129.0, 129.0, 129.6, 131.6, 133.5, 147.5, 153.5, 162.6 (Ar-C), 155.0 (CH=N), 177.0 (C=S); Anal. calcd for C16H16ClN3OS (333.84); C, 57.56; H, 4.83; N, 12.59, found C, 57.65; H, 5.74; N, 12.67. LC-MS (m/z: ESI+) calcd for C16H16ClN3OS [M]+, 333.14 found 334.0 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 98.6%.

N-Benzyl-2-(2-hydroxy-5-methylbenzylidene)hydrazinecarbothioamide (3e)

White solid, yield 82%; mp 218–220 °C, IR υ (cm–1) 1227 (C=S), 1608 (C=N), 3210, 3341 (N–H), 3435 (OH); 1H NMR (DMSO-d6) δ 2.20 (s, 3H, CH3), 4.86 (d, 2H, J = 4.8 Hz, CH2), 6.76 (d, 1H, J = 8.4 Hz, Ar-H), 6.94 (s, 1H, NH-CS), 6.97 (s, 1H, Ar-H), 7.02–7.05 (m, 1H, Ar-H), 7.18–7.28 (m, 5H, Ar-H), 8.00 (s, 1H, N=CH), 9.11 (s, 1H, NH-N), 10.27 (s, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 20.2 (CH3), 48.6 (CH2), 116.5, 116.6, 127.6, 127.9, 128.9, 129.5, 131.7, 133.3, 136.9, 147.7 (Ar-C), 155.0 (CH=N), 177.7 (C=S); Anal. calcd for C16H17N3OS (299.39); C, 64.19; H, 5.72; N, 14.04, found C, 64.28; H, 5.64; N, 14.11. LC-MS (m/z: ESI+) calcd for C16H17N3OS [M]+, 299.11 found 300.1 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 99.8%.

N-Cyclohexyl-2-(2-hydroxy-5-methylbenzylidene)hydrazinecarbothioamide (3f)

White solid, yield 78%; mp 229–231 °C, IR υ (cm–1) 1206 (C=S), 1596 (C=N), 3347 (N–H), 3452 (OH); 1H NMR (DMSO-d6) δ 1.15–1.29 (m, 3H, cyclohexyl), 1.32–1.42 (m, 2H, cyclohexyl), 1.54–1.59 (m, 1H, cyclohexyl), 1.64–1.69 (m, 2H, cyclohexyl), 2.00–2.04 (m, 2H, cyclohexyl), 2.21 (s, 3H, CH3), 4.20–4.22 (m, 1H, cyclohexyl), 6.53 (s, 1H, NH-CS), 6.79 (d, 1H, J = 8.4 Hz, Ar-H), 6.97–6.98 (m, 1H, Ar-H), 7.03–7.06 (m, 1H, Ar-H), 7.19 (s, 1H, N=CH), 9.28 (s, 1H, NH-N), 10.27 (s, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 20.2 (CH3), 24.6, 25.4, 32.6, 53.2 (cyclohexyl-C), 116.5, 116.8, 129.4, 131.7, 133.1, 147.2 (Ar-C), 155.1 (CH=N), 174.9 (C=S); Anal. calcd for C15H21N3OS (299.41); C, 61.82; H, 7.26; N, 14.42, found C, 61.90; H, 7.20; N, 14.35. LC-MS (m/z: ESI+) calcd for C15H21N3OS [M]+, 291.14 found 292.1 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 93.51%.

N-(4-Bromophenyl)-2-(2-hydroxy-5-methylbenzylidene)hydrazinecarbothioamide (3g)

White solid, yield 83%; mp 243–245 °C, IR υ (cm–1) 1218 (C=S), 1582 (C=N), 3328 (N–H), 3440 (OH); 1H NMR (DMSO-d6) δ 2.23 (s, 3H, CH3), 6.80 (d, 1H, J = 8.4 Hz, Ar-H), 7.06 (dd, 1H, J = 1.6 Hz, 8.4 Hz, Ar-H), 7.43 (dd, 2H, J = 2.0 Hz, 8.8 Hz, Ar-H), 7.61 (dd, 2H, J = 2.0 Hz, 8.8 Hz, Ar-H), 7.86 (s, 1H, Ar-H), 8.47 (s, 1H, N=CH), 9.74 (s, 1H, NH-CS), 10.05 (s, 1H, NH-N), 11.82 (s, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 20.1 (CH3), 115.9, 119.7, 126.8, 127.5, 127.8, 127.9, 129.2, 132.1, 138.2, 140.6 (Ar-C), 154.5 (CH=N), 175.7 (C=S); Anal. calcd for C15H14BrN3OS (364.26); C, 49.46; H, 3.87; N, 11.54, found C, 49.54; H, 3.94; N, 11.48. LC-MS (m/z: ESI+) calcd for C15H14BrN3OS [M]+, 363.0 found 364.1 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 99.8%.

2-(2-Hydroxy-5-methylbenzylidene)-N-(p-tolyl)hydrazinecarbothioamide (3h)

White solid, yield 87%; mp 224–226 °C, IR υ (cm–1) 1199 (C=S), 1595 (C=N), 3281, 3335 (N–H), 3412 (OH); 1H NMR (DMSO-d6) δ 2.21 (s, 3H, CH3), 2.29 (s, 3H, CH3), 6.80 (d, 1H, J = 8.4 Hz, Ar-H), 6.97 (d, 1H, J = 1.6 Hz, Ar-H), 7.05–7.09 (m, 1H, Ar-H), 7.12 (d, 2H, J = 8.4 Hz, Ar-H), 7.31(d, 2H, J = 8.0 Hz, Ar-H), 8.02 (s, 1H, N=CH), 8.33 (s, 1H, NH-CS), 9.03 (s, 1H, NH-N), 10.35 (s, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 20.2 (CH3), 21.1 (CH3), 116.6, 125.3, 129.6, 131.81, 132.8, 134.3, 134.6, 136.9, 157.6, 164.5 (Ar-C), 155.2 (CH=N), 175.3 (C=S); Anal. calcd for C16H17N3OS (299.39); C, 64.19; H, 5.72; N, 14.04, found C, 64.11; H, 5.66; N, 14.12. LC-MS (m/z: ESI+) calcd for C16H17N3OS [M]+, 299.11 found 300.1 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 93.4%.

N-(2,6-Dimethylphenyl)-2-(2-hydroxy-5-methylbenzylidene)hydrazinecarbothioamide (3i)

White solid, yield 90%; mp 237–239 °C, IR υ (cm–1) 1203 (C=S), 1583 (C=N), 3261, 3329 (N–H), 3426 (OH); 1H NMR (DMSO-d6) δ 2.19 (s, 6H, CH3), 2.20 (s, 3H, CH3), 6.75 (d, 1H, J = 8.0 Hz, Ar-H), 7.02 (dd, 1H, J = 1.6 Hz, 8.0 Hz, Ar-H), 7.08–7.12 (m, 3H, Ar-H), 7.90 (s, 1H, Ar-H), 8.42 (s, 1H, N=CH), 9.65 (s, 1H, NH-CS), 9.74 (s, 1H, NH-N), 11.63 (s, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 18.5 (CH3), 20.5 (CH3), 116.3, 120.4, 127.3, 127.3, 128.0, 128.2, 132.2, 137.0, 137.77, 140.0 (Ar-C), 154.8 (CH=N), 176.9 (C=S); Anal. calcd for C17H19N3OS (313.42); C, 65.15; H, 6.11; N, 13.41, found C, 65.10; H, 6.19; N, 13.34. LC-MS (m/z: ESI+) calcd for C17H19N3OS [M]+, 313.12 found 314.1 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 99.2%.

N-(2,4-Difluorophenyl)-2-(2-hydroxy-5-methylbenzylidene)hydrazinecarbothioamide (3j)

Yellow solid, yield 79%; mp 215–217 °C, IR υ (cm–1) 1197 (C=S), 1588 (C=N), 3299, 3354 (N–H), 3447 (OH); 1H NMR (DMSO-d6) δ 2.21 (s, 3H, CH3), 6.77 (d, 1H, J = 8.4 Hz, Ar-H), 7.02–7.13 (m, 2H, Ar-H), 7.33 (ddd, 1H, J = 2.8 Hz, 8.8 Hz, Ar-H), 7.47–7.53 (m, 1H, Ar-H), 7.86 (s, 1H, Ar-H), 8.45 (s, 1H, N=CH), 9.70 (s, 1H, NH-CS), 9.82 (s, 1H, NH-N), 11.90 (s, 1H, OH); Anal. calcd for C15H13F2N3OS (321.35); C, 56.06; H, 4.08; N, 13.08, found C, 56.14; H, 4.15; N, 13.12. LC-MS (m/z: ESI+) calcd for C15H13F2N3OS [M]+, 321.07 found 322.1 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 95.6%.

N-(3-Fluorophenyl)-2-(2-hydroxy-5-methylbenzylidene)hydrazinecarbothioamide (3k)

Yellow solid, yield 78%; mp 228–230 °C, IR υ (cm–1) 1219 (C=S), 1591 (C=N), 3324 (N–H), 3431 (OH); 1H NMR (DMSO-d6) δ 2.23 (s, 3H, CH3), 6.77 (d, 1H, J = 8.4 Hz, Ar-H), 7.02–7.05 (m, 2H, Ar-H), 7.37–7.48 (m, 2H, Ar-H), 7.60 (d, 1H, J = 6.8 Hz, Ar-H), 7.85 (s, 1H, Ar-H), 8.47 (s, 1H, N=CH), 9.74 (s, 1H, NH-CS), 10.07 (s, 1H, NH-N), 11.85 (s, 1H, OH); Anal. calcd for C15H14FN3OS (303.36); C, 59.39; H, 4.65; N, 13.85, found C, 59.45; H, 4.58; N, 13.76 LC-MS (m/z: ESI+) for C15H14FN3OS [M]+, 303.08 found 309.1 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 96.5%.

N-(4-Chlorophenyl)-2-(2-hydroxy-5-methylbenzylidene)hydrazinecarbothioamide (3l)

White solid, yield 83%; mp 214–216 °C, IR υ (cm–1) 1205 (C=S), 1582 (C=N), 3268, 3357 (N–H), 3450 (OH); 1H NMR (DMSO-d6) δ 2.22 (s, 3H, CH3), 6.77 (d, 1H, J = 8.4 Hz, Ar-H), 7.03 (dd, 1H, J = 1.6 Hz, 8.4 Hz, Ar-H), 7.40–7.43 (dd, 2H, J = 2.0 Hz, 8.8 Hz, Ar-H), 7.60 (dd, 2H, J = 2.0 Hz, 8.8 Hz, Ar-H), 7.85 (s, 1H, Ar-H), 8.46 (s, 1H, N=CH), 9.73 (s, 1H, NH-CS), 10.04 (s, 1H, NH-N), 11.81 (s, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 20.5 (CH3), 116.4, 120.1, 127.3, 128.02, 128.3, 128.3, 129.6, 132.6, 138.6, 141.0 (Ar-C), 155.0 (CH=N), 176.1 (C=S); Anal. calcd for C15H14ClN3OS (319.81); C, 56.33; H, 4.41; N, 13.14, found C, 56.40; H, 4.34; N, 13.22. LC-MS (m/z: ESI+) calcd for C15H14ClN3OS [M]+, 319.05 found 320.0 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 99.4%.

2-(2-Hydroxy-5-methylbenzylidene)-N-(o-tolyl)hydrazinecarbothioamide (3m)

White solid, yield 80%; mp 223–225 °C, IR υ (cm–1) 1195 (C=S), 1596 (C=N), 3298, 3336 (N–H), 3399 (OH); 1H NMR (DMSO-d6) δ 2.26 (s, 3H, CH3), 2.28 (s, 3H, CH3), 6.88 (d, 1H, J = 8.4 Hz, Ar-H), 7.04 (d, 1H, J = 7.2 Hz, Ar-H), 7.12 (s, 1H, Ar-H), 7.19 (d, 2H, J = 8.2 Hz, Ar-H), 7.39 (d, 2H, J = 8.0 Hz, Ar-H), 8.09 (s, 1H, N=CH), 8.40 (s, 1H, NH-CS), 8.64 (s, 1H, NH-N), 10.43 (s, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 20.4 (CH3), 21.2 (CH3), 116.7, 117.0, 125.5, 129.7, 131.9, 132.6, 133.6, 134.4, 134.7, 137.0, 157.7, 164.7 (Ar-C), 155.3 (CH=N), 175.5 (C=S); Anal. calcd for C16H17N3OS (299.39); C, 64.19; H, 5.72; N, 14.04, found C, 64.12; H, 5.76; N, 14.14. LC-MS (m/z: ESI+) calcd for C16H17N3OS [M]+, 299.09 found 300.1 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 99.3%.

2-(2-Hydroxy-5-methylbenzylidene)-N-(4-isopropylphenyl)hydrazinecarbothioamide (3n)

Green solid, yield 86%; mp 244–246 °C, IR υ (cm–1) 1214 (C=S), 1592 (C=N), 3274, 3316 (N–H), 3429 (OH); 1H NMR (DMSO-d6) δ 1.20 (d, 6H, J = 2.2 Hz, isopropyl CH3), 2.22 (s, 3H, CH3), 2.86–2.92 (m, 1H, isopropyl CH), 6.78 (d, 1H, J = 8.4 Hz, Ar-H), 7.03 (d, 1H, J = 7.2 Hz, Ar-H), 7.21 (d, 2H, J = 8.4 Hz, Ar-H), 7.44 (d, 2H, J = 8.4 Hz, Ar-H), 7.86 (s, 1H, Ar-H), 8.47 (s, 1H, N=CH), 9.71 (s, 1H, NH-CS), 9.94 (s, 1H, NH-N), 11.71 (s, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 20.1 (CH3), 24.0 (isopropyl CH3), 33.1 (isopropyl CH), 116.0, 119.9, 125.95, 126.0, 126.9, 127.9, 132.0, 136.9, 140.3, 145.5 (Ar-C), 154.6 (CH=N), 175.8 (C=S); Anal. calcd for C18H21N3OS (327.14); C, 66.02; H, 6.46; N, 12.83, found C, 66.12; H, 6.52; N, 12.76. LC-MS (m/z: ESI+) calcd for C18H21N3OS [M]+, 327.14 found 328.1 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 94.5%.

2-(2-Hydroxy-5-methylbenzylidene)-N-phenethylhydrazinecarbothioamide (3o)

White solid, yield 88%; mp 233–235 °C, IR υ (cm–1) 1217 (C=S), 1581 (C=N), 3257, 3341 (N–H), 3455 (OH); 1H NMR (DMSO-d6) δ 2.21 (s, 3H, CH3), 2.92 (t, 2H, J = 7.2 Hz, CH2), 3.88 (q, 2H, J = 6.8 Hz, CH2), 6.70 (s, 1H, NH-N), 6.79 (d, 1H, J = 8.4 Hz, Ar-H), 6.94 (s, 1H, NH-CS), 7.04 (dd, 1H, J = 2.0 Hz, 8.4 Hz, Ar-H), 7.17–7.19 (m, 4H, Ar-H), 7.26 (d, 2H, J = 7.6 Hz, Ar-H), 7.82 (s, 1H, N=CH), 9.91 (s, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 20.2 (CH3), 35.1 (CH2), 46.0 (CH2), 116.6, 116.6, 126.8, 128.7, 129.0, 129.4, 131.6, 133.2, 138.2, 147.3 (Ar-C), 155.16 (CH=N), 176.47 (C=S); Anal. calcd for C17H19N3OS (313.12); C, 65.15; H, 6.11; N, 13.41, found C, 65.23; H, 6.02; N, 13.32. LC-MS (m/z: ESI+) calcd for C17H19N3OS [M]+, 313.12 found 314.10 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 97.5%.

N-(4-(Dimethylamino)phenyl)-2-(2-hydroxy-5-methylbenzylidene)hydrazinecarbothioamide (3p)

White solid, yield 85%; mp 230–232 °C, IR υ (cm–1) 1210 (C=S), 1605 (C=N), 3276, 3325 (N–H), 3440 (OH); 1H NMR (DMSO-d6) δ 2.22 (s, 3H, CH3), 2.89 (s, 6H, 2 × CH3), 6.69 (d, 2H, J = 9.2 Hz, Ar-H), 6.76 (d, 1H, J = 8.0 Hz, Ar-H), 7.02 (dd, 1H, J = 2.0 Hz, 8.4 Hz, Ar-H), 7.25 (d, 2H, J = 8.8 Hz, Ar-H), 7.85 (s, 1H, Ar-H), 8.43 (s, 1H, N=CH), 9.68 (s, 1H, NH-CS), 9.80 (s, 1H, NH-N), 11.57 (s, 1H, OH); 13C NMR (100 MHz, DMSO-d6) δ 20.1 (CH3), 40.4 (2×CH3), 111.9, 116.0, 120.0, 126.9, 127.3, 127.9, 128.4, 131.9, 139.8, 148.5 (Ar-C), 154.4 (CH=N), 176.2(C=S); Anal. calcd for C17H20N4OS (328.4); C, 62.17; H, 6.14; N, 17.06, found C, 62.25; H, 6.08; N, 17.13. LC-MS (m/z: ESI+) calcd for C17H20N4OS [M]+, 328.14 found 329.10 [M + H]+. Purity determined by HPLC-UV (254 nm)-ESI-MS: 93.5%.

Enzyme Inhibition Assays

Inhibition of ALR1 and ALR2 by the synthesized compounds was determined by estimating the decrease in absorbance at 340 nm on a spectrophotometer (FLUOstar Omega BMG LABTECH, Germany). The protocols for the enzyme inhibition assay were adopted from our previous study with slight modifications.[17] In the assay, 100 μL of the total reaction mixture was composed of 20 μL of 100 mM sodium phosphate buffer (pH 6.2), 30 μL of the enzyme (expressed in the bacterial system; protein concentration 12 μg mL–1), 20 μL of dl-glyceraldehyde as a substrate (1 mM), 10 μL of the test compound (1 mM), and 20 μL of the NADPH cofactor (0.1 mM). First, the reaction mixture was incubated at 32 °C for 10 min without NADPH. Then, the reaction was initiated with the addition of the cofactor (NADPH) and monitored for 5 min. The change in absorbance was measured as preread (without a cofactor) and after reading (with a cofactor), the percentage inhibition was measured for test compounds and standard inhibitors. For ALR1 (aldehyde reductase), sodium-d-glucuronate was employed as a substrate and valproic acid as a standard inhibitor. For ALR2, DL-glyceraldehyde was used as the substrate and sorbinil as the standard inhibitor. Otherwise, identical protocols were adopted for the ALR1 and ALR2 assays. The method for expression of aldose reductase in Escherichia coli BL21 (DE3) has been added to the supporting information.

First, the inhibitor to be tested was dissolved in DMSO (100%) and dilutions were prepared with deionized water to keep the concentration of DMSO at 0.1% during the assay. Test compounds were prepared as 100 μM solutions; these compounds that exhibited greater than 50% inhibition were further analyzed to establish their IC50 values using different dilutions up to 10 nM. The logarithms of the inhibitor concentration were plotted versus the remaining activity of the enzyme, and IC50 values were calculated using nonlinear regression analysis in GraphPad Prism Version 8.

Methods for Docking and Molecular Dynamics Simulations

Molecular docking and molecular dynamics studies were carried out for the most potent inhibitors synthesized. LeadIT software from BioSolveIT was used and its FlexX utility was exploited for docking purposes.[27] The crystal structure of ALR2 (PDB ID 1US0) was selected due to its better resolution and aldose reductase inhibitor presence for easy identification of the active pocket. The ALR2 crystal structure was subjected to default preparation and docking parameters of the software.[28] Docking studies were performed in the presence of a cofactor (NADPH). By redocking the cognate ligand, the docking protocol was validated. The structures of the inhibitor were sketched, and the energy minimized with Chem3D v20.0. prior to the docking studies. Scoring and ranking of the docked poses were performed using the FlexX utility with a hybrid enthalpy and entropy approach. The pose with the highest HYDE score was selected for subsequent studies.[29,30]

GROMACS v.2020 was used for the MD simulations of inhibitor 3c inside the ALR2 active site.[31,32] The Charmm36 force field with TIP3P as the explicit water model was used.[33] The docked pose of inhibitor 3c previously selected by the HYDE assessment was used. The topology and parameter files of the inhibitor were obtained using the web-based server (https://cgenff.umaryland.edu) of the Charmm General Force Field (CGENFF). An initial system of the enzyme–cofactor–inhibitor complex was prepared. The system was wrapped in the TIP3P water box and neutralized with Na+ and Cl- counter ions. The energy of the complex system was then minimized using 500 maximum steps of steepest descent and 500 maximum steps of the conjugate gradient method. The system was then subjected to 100 ps of an isothermal–isochoric ensemble using a velocity rescaling Berendsen thermostat and an isothermal–isobaric ensemble using a Berendsen barostat. 50 ns of the production run was carried out at 300K and 1 atmosphere pressure.

To calculate the binding free energy of the protein–ligand complex, the gmx_MMPBSA tool was used.[34] Snapshots were taken every 10 ps from the entire MD ensemble. The binding free energy was estimated by taking into account the vacuum potential energy inclusive of both the bonded as well as the nonbonded terms. Solving the Poisson–Boltzmann equation was used to determine the polar solvation term, while the solvent-accessible surface area (SASA) method was used to determine the nonpolar solvation energy terms.

To determine the heterogeneity of conformations obtained by the MD ensemble, K-means, the clustering method embedded in GROMACS v.2020 was used to perform the cluster analysis of compound 3c.[35,36] An RMSD cutoff value of 0.1 nm was used for clustering the poses and assigning them to individual clusters. Similar poses were clustered together if their root-mean-square deviation (RMSD) was less than the defined cutoff value.

Antiglycation Assay

The glycation of bovine serum albumin (BSA) was performed by following a previously reported method with slight modifications.[26] BSA (1 mg mL–1) was incubated with 0.5 M glucose in 0.1 M PBS (phosphate-buffered saline) and the test compound (1 mM) at pH 7.4 and 50 °C for 4 days in the dark. Aminoguanidine was used as a reference glycation inhibitor. The glycation of BSA was measured by determining the fluorescence intensity at excitation and emission wavelengths of 350 nm and 460 nm, respectively. The percentage of AGE (advanced glycation end product) inhibition was calculated using the following formulawhere Fcontrol – Fcontrolblank is the difference between the fluorescence intensity of BSA incubated with or without glucose and Ftestcompounds – Ftestblank is the difference between the fluorescence intensity of BSA and glucose incubated with or without the test compounds.

Antioxidant Assay (Free-Radical Scavenging Activity)

The antioxidant potential of compounds 3a–p was investigated in this study with ascorbic acid used as a positive control. DPPH was used to assess the free-radical quenching ability of newly synthesized compounds to detect their antioxidant potential according to a previously reported protocol with slight modifications.[37] The percent radical scavenging activity (%RSA) of the compounds was determined by spectrophotometric analysis at 517 nm in which a homogeneous mixture composed of a methanolic solution of DPPH (0.025 mg mL–1) and a 100 μM solution of the test compound was used. The following formula was used to calculate the percent free-radical scavenging activitywhere % FRSA is the percent free-radical scavenging activity and “Abs” is an abbreviation of “absorption”.