Diabetes mellitus is a spectrum of metabolic disorders resulting from myriad pathogenic mechanisms, all of which lead to hyperglycemia ^{[ 1 ]}. In recent years, diabetes has become one of the leading causes of death worldwide. In 2016, the World Health Organisation reported around 1.6 million deaths globally due to diabetes. The number of diabetes cases worldwide in 2019 was 463 million and by 2045 it could increase to 629 million. In India, 77 million suffer from diabetes and could be further increased and India has the second highest number of cases in the world ^{[ 2 ]}. Oxidative stress plays an important role in the pathogenesis of typical long-term diabetic complications such as neuropathy and microangiopathy. Further, dynamic modification of intracellular redox sensors by way of reactive oxygen species in addition to other post-translational modifications such as protein phosphorylation, acetylation or ubiquitination also play a role in the development of the diabetes ^{[ 3 ]}. Thiazolidinone is one of the important and fundamental central structures for the synthesis of biologically active compounds ^{[ 4 ]}. Thiazolidin-4-one is a well-known scaffold for producing the desired biological activities: Antidiabetic ^{[ 5 ]}, antioxidant ^{[ 6 ]}, anticancer ^{[ 7 ]}, antitubercular ^{[ 8 ]}, antimicrobial ^{[ 9 ]}, anticonvulsant ^{[ 10 ]}, analgesic and anti-inflammatory ^{[ 11 ]} and antihyperlipidemic ^{[ 12 ]}.

The abovementioned facts triggered an interest in taking up this investigation, wherein molecular architecture having a 2,3-disubstituted 4-thiazolidinone core derived from sulfanilamide was designed with an anticipation of good antidiabetic activity. Spectral characterisation, docking studies with peroxisome proliferative activated receptor-y (PPAR-y), and in silico ADME parameter studies were also undertaken.

Chemicals employed were of laboratory reagent grade and used as received. Sulphanilamide, 1,1-diphenyl-2-picryl-hydrazyl (DPPH), and p-hydroxybenzaldehyde were purchased from Loba Chemie Pvt Limited, Mumbai, India. The other substituted aldehydes, 1,4-dioxane, thioglycolic acid, anhy-drous zinc chloride, ascorbic acid, and D-fructose were procured from SD Fine Chemicals, Mumbai, India. p-Tolualdehyde was purchased from Sigma Aldrich, Steinheim, USA. p-Fluorobenzaldehyde was obtained from Spectrochem, Mumbai, India; and 2,4-dichlorobenzaldehyde, 2,6-dichlo-robenzaldehyde, and p-bromobenzaldehyde from HiMedia Laboratories Pvt. Ltd. Mumbai, India. The glucose, cholesterol, and triglyceride kits were purchased from Erba Mannheim, Sikkim, India. Pioglitazone hydrochloride was provided by USV Private Limited, Solan, India. Thin layer chromatography using pre-coated aluminium sheets with silica gel 60 F₂₅₄ (Merck, Darmstadt, Germany) was performed using a combination of sol-vents in different ratios as mobile phase. Spots were visualised by iodine vapor and/or UV light. Yields refer to crude products.

Melting points were determined on an open capillary apparatus (Veego VMP-DS Mumbai, India) and are uncorrected. The compounds were characterised by ultraviolet (UV) spectra (UV 1601 Shimadzu, Kyoto, Japan) and infrared (IR) spectra as KBr pellets (Shimadzu and Thermo Scientific Nicolet iS5, Japan). Proton nuclear magnetic resonance (1H NMR) was recorded by means of DMSO-d6 (Bruker Avance Neo 500 MHz, Germany) while 13C NMR was recorded at 125 MHz. Mass spectra (MS) were registered by chemical ionisation technique (Waters-SynaptG2, Massachusetts, USA).

Sprague Dawley rats of either sex weighing between 250 and 350 g and Swiss albino female mice weighing from 25 to 40 g were obtained from the central animal house of the HSK College of Pharmacy, Bagalkote. The animals were housed under standard conditions (temperature 25 ±1°C, relative humidity (50-55%)) for 12 h dark and light cycle. They were given standard laboratory feed (Pranava Agro Industries Ltd., Sangli, India) and water ad libitum and acclimatised for 7 days before experimentation. Food was withdrawn 18 h before the experiment, but free access to water was allowed. The experiments were performed on randomly formed groups during the light phase of the cycle and the animals were used for one experiment only. The test compounds and standard drug were administered p.o. as aqueous suspension in 0.5% Tween-80. The animals in the vehicle control group received 0.5% aqueous Tween-80. Blood samples were collected from retro orbital flexus under ether anaesthesia. The study was approved and conducted under the guidelines of institutional animal ethics committee constituted as per committee for the control and supervision of experiments on animals, Ministry of Environment and Forestry, Government of India (F.No. HSK College of Pharmacy Bagalkot/IAEC//HSKCOP/AUG2021/PG13).

The hypoglycemic side effect of antimicrobial sulfonamides led to the development of the sulfonyl urea class of hypoglycemic agents such as tolbutamide and chlorpropamide. Glitazone antidiabetics such as rosiglitazone and pioglitazone show their activity due to the presence of acidic thiazolidinedione head. Enthused by this fact, the title compounds which have both the features of sulfonamide and thiazolidin-4-one (that closely resembles the thiazolidinedione present in glitazone antidiabetics) in the same molecule were designed by way of the molecular hybridisation strategy ^{[ 28 ]} with an anticipation of better hypoglycemic activity ^{[ 29 ]}.

Compounds 4e, 4h, 4i, and 4n were found to be safe at the dose of 2000 mg/ kg b.w. Thus, they were tested at 200 mg/kg b.w. and compound 4e at 50 mg/kg b.w. orally for in vivo antidiabetic activity.

It is a well-known fact that, diabetes mellitus is associated with oxidative stress ^{[ 30 ]} and compounds having the antioxidant property will be of great value in the treatment of diabetes. Therefore, the thiazolidin-4-ones that have exhibited good antioxidant activity in the DPPH assay namely, 4e, 4h, 4i and 4n were selected for antidiabetic activity. The antidiabetic activity was carried out by estimating the serum glucose level as per equation 2 in rats with fructose induced diabetes. The estimation of the serum glucose level was carried out by Trinder's method in which, glucose oxidase reacts with glucose, water, and oxygen to form gluconic acid and hydrogen peroxide. The enzyme peroxidase catalyses the oxidative coupling of 4-aminoantipyrine with phenol in the presence of hydrogen peroxide to yield a purple coloured quinoneimine complex with absorbance proportional to the concentration of glucose^{[ 31 ]}.

The administration of fructose resulted in the induction of diabetes in rats when compared to the normal group. The tested compounds exhibited highly significant antidiabetic activity when compared with the control group. Compound 4e with p-hydroxy substitution reduced serum glucose by 67.93% followed by compound 4h (p-Cl) by 66.03%, compound 4i (2,4-Cl₂) by 65.29%, and compound 4n by 62.68% on the 21st day of the treatment. All the compounds exhibited higher glucose lowering activity than the standard drug pioglitazone. Contrary to DPPH scavenging activity, p-OH substitution seems to favour the activity more than p-Cl and 2,4-Cl₂. The reduction in the serum glucose level by way of compounds was found to be significant when compared with pioglitazone on the 3^rd day, compound 4e exhibited a significant reduction in the serum glucose level on the 14^th day. The results are tabulated on Table 2.

Table 2

All the values are expressed as mean ± SEM, n = 6, ^a p < 0.001 as compared to the normal group, ***p < 0.001 as compared to the control group and ^### p < 0.001, ^## p < 0.01 and ^# p < 0.05 as compared to the standard group [ANOVA followed by multiple comparison Dunnett's test].

Cholesterol esterase catalyses the conversion of cholesterol ester to cholesterol in the presence of water. The released cholesterol is then converted into cholest-4-en-3-one and hydrogen peroxide in the presence of oxygen by cholesterol oxidase. The released hydrogen peroxide reacts with 4-aminoantipyrine and phenol to yield a purple complex of quinoneimine that was measured spectrophotometrically ^{[ 19 ]}. The serum cholesterol was calculated as per equation 3. The tested compounds exhibited better and significant serum cholesterol lowering activity than pioglitazone, when compared with the control group. Compound 4e possessed the highest activity followed by 4h, 4i, and 4n. The activity pattern displayed by the compounds is the same as that of antidiabetic activity. The reduction in total cholesterol by tested compounds was found to be insignificant when compared with pioglitazone; only compound 4e has shown significant activity on the 21^st day. The results are tabulated on Table 3.

Table 3

All the values are expressed as mean ± SEM, n = 6, ^a p < 0.001 as compared to the normal group, ***p < 0.001 as compared to the control group and ^##p < 0.01 as compared to the standard group. [One way ANOVA followed by multiple comparison Dunnett's test].

The histopathological investigation of the pancreas of rats belonging to the normal group showed a normal architecture of islets of Langerhans with pale, rounded and ovoid p cells in the centre embedded in the exocrine portion of the pancreas. Whereas, shrinkage of islets of Langerhans with degeneration and necrosis of cells with their nucleus appearing densely basophilic with predominant karyolysis was observed in rats of the control group. The pancreas of diabetic rats treated with pioglitazone showed a normal architecture of islets of Langerhans with pale, large, round to ovoid shape containing cells embedded in the exocrine portion of the pancreas. In the case of a pancreas treated with compounds 4h and 4i normal sized islets of Langerhans but some degeneration of p cells in the centre were noticed, suggesting a protective role of compounds 4h and 4i on the islets of Langerhans. The histopathological changes are presented in Figure 2.

Figure 2 Normal: Pancreatic islet of rat showed intact cluster of β-cells and architecture of islets of Langerhans. Control: Showed destruction of pancreatic islets of β-cells, islet shape and number of islets are reduced. Standard: Showed apparently reduced degranulation architecture, moderately restored pancreatic cells, islet shape and increased number of islets. Treatment: 4e, 4h, 4i and 4n treated showing regeneration of β-cells, restoration of architecture, shape and increase in the number of islets.

The compounds that were tested for antidiabetic activity were docked with PPAR-y (PDB ID: 4PRG) using autodock tools 1.5.6 to explore their interaction. The PPAR-y plays an important role in the regulation of glucose and homeostasis and was used as a molecular target for the 2,4-thiazolidinedione class (glitazones) of antidiabetic agents ^{[ 32 ]}. The binding energies and amino acid residues with which compounds 4e, 4h, 4i, 4n and pioglitazone interacted are presented on Table 5 and in Figure 3. The compounds exhibited more binding energy with PPAR-y than with pioglitazone.

Table 5

Figure 3

It is interesting that, the tested compounds exhibited higher binding energy with PPAR-y. The compound 4n demonstrated the highest binding energy with -10.3 kcal/mol among the tested compounds and formed two hydrogen bonds with amino acid residues ILE (A:267) and ARG (A:280), and also formed other bonds by interacting with other amino acid residues PHE (A:264), LYS (A:265), and PHE (A:287). Likewise, compound 4i showed good binding energy of -8.3 kcal/mol and formed two hydrogen bonds with ASP (B:475) and ILE (B:279), and also formed another bond with GLU (B:460). Compound 4e displayed good binding energy of -7.6 kcal/mol and formed three hydrogen bonds with GLU (B:343), ARG (B:288), and LEU (B:340), and formed other bonds with LEU (B:228) and LEU (B:330). Compound 4h also showed binding energy of -7.3 kcal/mol and formed a hydrogen bond with GLU (B:460), and other bonds with LEU (B:453) and VAL (B:450). Compound 4n interacted with the A chain while, others with the B chain of PPAR-y. All the results are compared with pioglitazone which had binding energy of -7.12 kcal/mol and formed two hydrogen bonds with GLU (B:343) and LYS (C:263), and other bonds with PHE (C:287), LEU (C:255), MET (C:256), ILE (C:281), PHE (C:264), and ARG (C:280) with the B and C chains. Thus, compound 4n exhibited excellent binding energy and interaction with PPAR-y followed by 4i, 4e and 4h when compared with pioglitazone. It is interesting to note that, the compounds which showed good antidiabetic activity interacted with a lower binding energy with PPAR-y.

For a molecule to be effective as a potent drug, it must reach its target in the body in sufficient concentration and stay there in a bioactive form long enough for the expected biologic events to occur. Drug development involves assessments of pharmacokinetic properties such as absorption, distribution, metabolism, and excretion (ADME); in this context computer models constitute valid alternatives to experiments. The Swiss ADME web tool provides easy efficient inputs and free access to a pool of past yet robust predictive models for the physicochemical properties, pharmacokinetics, and drug likeness of compounds. The ADME and drug likeness of 2,3-disubstitued thiazolidin-4-ones (4a-n) were determined by the Swiss ADME web tool and the results are tabulated on Table 6.

Table 6

MW: Molecular Weight, Fsp3: Fraction of sp3 Carbon atoms, RB: Rotatable Bonds, HBA: Hydrogen Bond Acceptor, HBD: Hydrogen Bond Donor, MR: Molar Refractivity, TPSA: Topological Polar Surface Area, LogP: Indicator of Lipophilicity, Log KP: Skin Permeation, F: Bioavailability Score

The filters used to evaluate the drug likeness of the title compounds were i) Lipinski filter: MW≤500, MLogP≤ 4.15, HBA≤ 10, HBD≤ 5 ^{[ 33 ]} ii) Ghose filter: 160≤ MW≤ 480, -0.4≤ WlogP≤ 5.6, 40≤ MR≤ 130, 20≤ atoms≤ 70 ^{[ 34 ]} iii) Veber (GSK) filter: RB≤ 10, TPSA≤ 140 ^{[ 35 ]} iv) Egan (Pharmacia) filters: WLogP≤ 5.88, TPSA≤ 131.6 ^{[ 36 ]} v) Muegge (Bayer) filter : 200≤ MW≤ 600, -2≤ XLogP≤ 5, TPSA≤ 157; HBD≤ 5, RB≤ 15, number of rings ≤ 7; number of carbons >4, number of heteroatoms ≤ 1 ^{[ 37 ]}. The compounds did not violate the Lipinski and Ghose rule but, some violated the Veber, Egan, and Muegge rules. To summarise, 2,3-disubstituted thiazolidin-4-ones met the criteria of drug likeness and exhibited satisfactory ADME parameters.

A series of 2,3-disubstituted thiazolidinones 4a-n were synthesised from Schiff base intermediates 3a-n that have been derived from sulfanilamide. The final products were characterised by physicochemical parameters and spectral data.

Compounds 4e, 4h, 4i, and 4n exhibited weak DPPH radical scavenging activity. The acute toxicity study revealed the safety of compounds 4e, 4h, 4i and 4n at the dose of 2000 mg/kg b.w. p.o. Compounds 4h, 4i and 4n were tested at 200 mg/kg b.w. and compound 4e at 50 mg/kg b.w. orally for in vivo antidiabetic activity. Compounds 4e with p-hydroxy substitution, 4h with p-chloro substitution, 4i with 2,4-dichloro substitution, and 4n with vinyl substitution exhibited highly significant antidiabetic activity when compared with the control group. The tested compounds exhibited better and significant serum cholesterol lowering activity with reference to the control group. Compound 4e possessed the highest cholesterol lowering activity followed by 4h, 4i, and 4n. The tested compounds also reduced the serum triglyceride level after the 21^st day, however this was found to be insignificant in comparison to the control group. Compound 4h exhibited the highest serum triglyceride reducing activity followed by 4n and 4e. The tested compounds reduced elevated body weight but it was found to be insignificant when compared with the control group. The histopathology demonstrated the protective role of compounds 4h and 4i on the p cells of islets of Langerhans. The molecular docking showed compound 4n binds with high energy more than pioglitazone with PPAR-y. The physicochemical, ADME, and drug likeness properties of most of the synthesised compounds were found to be within permitted limits. Thus, the molecular architecture developed and synthesised in this study would be of great interest with good antidiabetic potential warranting further work.