Synthesis, Evaluation and Molecular Docking of Thiazolopyrimidine Derivatives as Dipeptidyl Peptidase IV Inhibitors


  • Mani Sharma,

    Corresponding author
    1. Department of Pharmaceutical Chemistry, Delhi Institute of Pharmaceutical Sciences and Research, Pushp Vihar, Sector-3, M B Road, New Delhi 110017, India
      Corresponding authors: Mani Sharma,; Monica Gupta,
    Search for more papers by this author
  • Monica Gupta,

    Corresponding author
    1. Department of Pharmaceutical Chemistry, Delhi Institute of Pharmaceutical Sciences and Research, Pushp Vihar, Sector-3, M B Road, New Delhi 110017, India
      Corresponding authors: Mani Sharma,; Monica Gupta,
    Search for more papers by this author
  • Divya Singh,

    1. Department of Pharmaceutical Chemistry, Delhi Institute of Pharmaceutical Sciences and Research, Pushp Vihar, Sector-3, M B Road, New Delhi 110017, India
    Search for more papers by this author
  • Manoj Kumar,

    1. Department of Biophysics, All India Institute of Medical Sciences, New Delhi 110029, India
    Search for more papers by this author
  • Punit Kaur

    1. Department of Biophysics, All India Institute of Medical Sciences, New Delhi 110029, India
    Search for more papers by this author

Corresponding authors: Mani Sharma,; Monica Gupta,


A series of thiazolopyrimidine derivatives was designed, synthesized and screened for in-vitro inhibition of Dipeptidyl Peptidase IV (DPP IV). The SAR study indicated the influence of substituted chemical modifications on thiazolopyrimidine scaffold. Compound 9 (IC50 = 0.489 μm) and 10 (IC50 = 0.329 μm) having heterocyclic-substituted piperazine with acetamide linker resulted as most potent DPP IV inhibitors among all the compounds screened. Single dose (10 mg/kg) of both the compounds 9 and 10 significantly reduced glucose excursion during oral glucose tolerance test in streptozotocin induced diabetic rat model. Molecular docking studies illustrated the probable binding mode and interactions of thiazolopyrimidine nucleus and its derivatives at binding site of receptor. The binding site for DPP IV is composed of active site region (catalytic triad of Ser630, Asp708 and His740) including S1 and S2 sub-pocket. The aryl moiety of compounds 9, 10 and 11 were observed to occupy S2 binding pocket and interacted with aromatic ring of Tyr662 and Tyr666 acquired through π-π interaction. Thus, it is indicated that occupancy of the highly hydrophobic S2 pocket is more important for DPP IV inhibitory activity. The present study on substituted thiazolopyrimidine derivatives shows good to moderate inhibitory potential of DPP IV enzyme.

Diabetes mellitus (T2DM) is slowly but gradually growing metabolic disorder affecting 90–95% of population worldwide (1). The continued global burden of diabetes is speculated to be double within the next 25 years and will affect 380 million people by 2025 (2). Patients with diabetes experiences severe pancreatic ß-cell dysfunction and a progressive loss of insulin secretion ultimately leads to hyperglycaemia (3). The treatment of diabetes includes modification in lifestyle and use of anti-hyperglycaemic agents to maintain the adequate glucose level (4). Nevertheless, therapeutically available anti-diabetic agents are associated with risks of adverse effects such as gastrointestinal toxicities, hypoglycaemia and weight gain. As a result, novel therapies having long-term efficacy and improved tolerability are urgently required to attain glycemic control. Amongst the various strategies of treatment, DPP IV inhibitor appears to be novel class of antidiabetic drugs, as they are observed with improved insulin secretion in type-2 diabetic patients (5).

DPP IV is a 766-amino acid serine protease which belongs to prolyl oligopeptidase family. It is highly specific enzyme, removes N-terminal of dipeptide from amino terminus preferentially having proline or alanine residues at penultimate position (6). DPP IV rapidly inactivates incretin hormones (<2 min), secreted by intestinal endocrine cells in response to meal intake (7,8). The incretin hormone includes glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide responsible for 50–70% of postprandial insulin release (9). GLP-1 is a naturally occurring peptide hormone and allows regeneration of insulin stimulating pancreatic ß-cell function. It also regulates glucose homoeostasis by potentiating glucose-induced insulin secretion and reducing production of glucagon (10). The inhibition of DPP IV will restore the level of endogenous incretin hormones and might delay or prevent the onset of diabetes (11). DPP IV inhibitors possess advantages over existing antidiabetic agents as increase in weight gain and hypoglycaemia is not seen (5,12). Therefore, DPP IV inhibitors have emerged as novel class of antidiabetic drugs which prevent proteolytic degradation of incretin hormones and stimulate glucose-dependent insulin secretion (5).

These findings encouraged us to explore targeted DPP IV inhibitors which may serve as better approach to surmount the problems consorted by clinically used antidiabetic agent. Earlier studies shows that heterocyclic structures such as fused azolopyrimidines (13), thienopyrimidine [14], benzimidazole (15), quinoxalinedione (16), imidazopyrazinone (17), imidazopyridazinones (18) and xanthine (19) have played a vital role in providing series of potent, selective and novel DPP IV inhibitors. Taking the earlier consideration in fact, maintaining cyclic integrity of the pharmacophore may prove to be an attractive strategy for development of DPP IV inhibitors with desirable activity.

The structural frame work of molecules was designed using thiazolopyrimidine ring as core bioactive scaffold. Accordingly, the chemical modifications were carried out on scaffold to increase its affinity and selectivity towards DPP IV inhibition. However, the modified pharmacophore closely showed resemblance with the various potent and selective DPP IV inhibitors (Figure 1) (20,21).

Figure 1.

 Dipeptidyl Peptidase IV inhibitors.

Thus, the present study discloses synthesis of series of thiazolopyrimidine derivatives, structure-activity relationship (SAR) and in-vitro inhibitory activity against DPP IV enzyme. Among the synthesized derivatives, two compounds 9 and 10 showed the most potent DPP IV inhibition and excellent anti-hyperglycaemic activity in streptozotocin induced diabetic rat model. Molecular docking study has been carried out for the insight of binding mode and interaction pattern of designed inhibitors of DPP IV using crystal structure of DPP IV.

Methods and Materials


All commercial chemicals and solvents were purchased from Sigma-Aldrich Chemical Company, St. Louis, MO, USA and Spectrochem Pvt. Ltd. Chemicals, India and solvents were used as such without further purification. 1H spectra were recorded on Bruker NMR spectrometer operating at 300 MHz in CDCl3 or DMSO-d6 with tetramethylsilane as an internal standard. Chemical shifts are expressed as parts per million (p.p.m., δ), and signals are reported as s (singlet), d (doublet), t (triplet), q (quartet) and m (multiplet). All the Fourier Transform Infrared (FT-IR) spectra were recorded on Jasco-410 spectrophotometer. Elemental analysis was carried out using Vario EL III elemental analyser. Melting points of compounds were determined in open capillary tubes using Hicon melting point instrument and are uncorrected. All reactions were carried out under dried condition and performed using oven-dried glassware.

Step I: synthesis of 2-amino-4-methyl-4H-thiazolo [4,5-d]pyrimidine-5,7-dione (1)

The compound 1 is prepared by previously reported procedure (22). The compound obtained was authenticated by further characterization. White solid; Yield: 89%; mp: 104–110 °C; Rf: 0.35; 1H NMR (300MHz, DMSO-d6) δ p.p.m.: 3.4 (s, 3H), 6.55 (s, 2H); FT-IR (νmax; /cm, KBr) 3313 (N-H stretch), 3100 (C-H stretch), 1140 (C-N stretch), 1233 (C-S stretch); Anal. Calcd. For C6H6N4O2S: C, 36.36; H, 3.05; N, 28.27; Found: C, 36.31; H, 3.09; N, 28.25.

Step II: general procedure for synthesis of N-substituted chloracetamides (27)

Secondary amine (2.9 mol) was taken in 1, 4-dioxan (10 mL) maintaining temperature 50–60 °C. The solution of chloroacetyl chloride (3.33 mol) in 1, 4-dioxan (2 mL) was added drop wise under continuous stirring. After reflux of 2 h, cooled mixture to room temperature and stirred for 30 min. The completion of reaction was monitored by thin-layer chromatography (TLC) using mobile phase, chloroform:methanol (9.5:0.5). The reaction mixture was concentrated under reduced pressure and the resulting mixture was diluted with water and neutralized using saturated bicarbonate solution. The solution was further diluted and partitioned between water and chloroform. The organic layer was washed twice with water, brine and dried over anhydrous sodium sulphate. The organic layer was evaporated to the dryness under reduced pressure and resultant intermediates (2–7) were obtained in excellent yields.

1-(4-aminopiperidin-1-yl)-2-chloroethanone (2)

Yellow oil; Yield: 75%; Rf: 0.69; 1H NMR (300MHz, CDCl3) δ p.p.m.: 2.00 (q, = 12, 4H), 3.2 (t, = 9, 4H), 4.46 (d, = 12, 2H), 4.8 (s, 2H); FT-IR (νmax; /cm, neat): 2936 (C-H stretch), 1649 (C=O), 1169 (C-N stretch), 654 (C-Cl stretch); Anal. Calcd. For C7H13ClN2O: C, 47.60; H, 7.42; N, 15.86; Found: C, 47.62; H, 7.45; N, 15.89.

2-chloro-1-[4-(pyrimidin-2-yl) piperazin-1-yl]ethanone (3)

Yellow oil; Yield: 87%; Rf: 0.71; 1H NMR (300MHz, CDCl3) δ p.p.m.: 3.70–3.64 (m, 4H), 3.91–3.73 (m, 4H), 4.28 (s, 2H), 6.66 (t, = 9, 1H), 8.38 (d, = 9, 2H); FT-IR (νmax; /cm, neat): 2931 (C-H stretch), 1629 (C=O), 1493 (C=N), 640 (C-Cl stretch); Anal. Calcd. For C10H13ClN4O: C, 49.90; H, 5.44; N, 23.28; Found: C, 49.94; H, 5.39; N, 23.24.

2-chloro-1-[4-(pyridin-2-yl) piperazin-1-yl]ethanone (4)

Brown oil; Yield: 87%; Rf: 0.70; 1H NMR (300MHz, CDCl3) δ p.p.m.: 3.71–3.66 (m, 4H), 3.91–3.73 (m, 4H), 4.43 (s, 2H), 6.73 (t, = 12 Hz, 1H), 6.99 (d, = 9, 1H), 7.71 (t, = 15.3, 1H), 8.17 (s, 1H); FT-IR (νmax; /cm, neat): 3004 (C-H stretch), 2920 (C-H stretch), 1647 (C=O), 1541 (C=N), 775 (C-Cl stretch); Anal. Calcd. For C11H14ClN3O: C, 55.12; H, 5.89; N, 17.53; Found: C, 55.17; H, 5.85; N, 17.58.

2-chloro-1-[4-(4-chlorophenyl) piperazin-1-yl]ethanone (5)

Yellow solid; Yield: 84%; mp: 129–133 °C; Rf: 0.71; 1H NMR (300MHz, CDCl3) δ p.p.m.: 3.23–3.15 (m, 4H), 3.72–3.68 (m, 4H), 4.12 (s, 2H), 6.88 (d, = 6 Hz, 2H), 7.28 (d, = 6 Hz, 2H); FT-IR (νmax; /cm, KBr): 2917 (C-H stretch), 1594 (C=O), 1497 (C=C ring stretch), 667 (Ar-Cl stretch), 624 (C-Clstretch); Anal. Calcd. For C12H14Cl2N2O: C, 52.76; H, 5.17; N, 10.26; Found: C, 52.79; H, 5.19; N, 10.29.

1-(azepan-1-yl)-2-chloroethanone (6)

Light brown oil; Yield: 78%; Rf: 0.69; 1H NMR (300MHz, CDCl3) δ p.p.m.: 1.79–1.73 (m, 4H), 1.87–1.82 (m, 4H), 3.58–3.42 (m, 4H), 4.11 (s, 2H); FT-IR (νmax; /cm, neat): 1647 (C=O), 1099 (C-N stretch), 651 (C-Cl stretch); Anal. Calcd. For C8H14ClNO: C, 54.70; H, 8.03; N, 7.97; Found: C, 54.74; H, 8.05; N, 7.99.

2-chloro-1-(piperidin-1-yl) ethanone (7)

Yellow oil; Yield: 77%; Rf: 0.69; 1H NMR (300MHz, CDCl3) δ p.p.m.: 2.00–1.89 (m, 4H), 3.48 (t, = 18 Hz, 4H), 4.0 (s, 2H); FT-IR (νmax; /cm, neat): 1647 (C=O), 1096 (C-N stretch), 802 (C-Cl stretch); Anal. Calcd. For C7H12ClNO: C, 52.02; H, 7.48; N, 8.67; Found: C, 52.07; H, 7.52; N, 8.71.

Step III: General procedure for synthesis of compounds (813)

To a solution of 2-amino-4-methyl-4H-thiazolo [4, 5-d] pyrimidine-5,7-dione (0.25 mmol) in dry DMF (1 ml), added anhydrous potassium carbonate (0.75 mmol) and mixture was heated initially at 70 °C for 2 h under continuous stirring. At 70 °C drop wise added the solution of N-substituted chloracetamide derivative (0.25 mmol) in dry DMF to the reaction mixture. The reaction mixture was heated at 70–80 °C for 3 h, and completion of reaction was monitored through TLC using mobile phase as chloroform:methanol (9.5:0.5). The reaction mixture was brought to room temperature, filtered to remove insoluble particulates. Filtrate was diluted with ice-cold water and partitioned between water and ethyl acetate layer. The extraction process was repeated twice, combined organic layers were washed with water and brine. The organic layers were dried over anhydrous sodium sulphate and concentrated under reduced pressure to give the final compounds. Compounds obtained as solid were recrystallized using appropriate solvent, and the oily compounds were purified by column chromatography (5% MeOH in DCM). The final synthetic compounds were obtained in considerable yields.

2-amino-6-[2-(4-aminopiperidin-1-yl)-2-oxoethyl]-4-methyl[1,3]thiazolo[4,5-d]pyrimidine- 5,7(4H,6H)-dione (8)

Yellow oil; Yield: 64%; Rf: 0.65; 1H NMR (300MHz, CDCl3) δ p.p.m.: 2.00 (q, = 12, 4H), 3.2 (t, = 9, 4H), 3.55 (s, 3H), 4.46 (d, = 12, 2H), 4.8 (s, 2H); FT-IR (νmax; /cm, neat): 3313 (N-H stretch), 1649 (C=O), 1233(C-S stretch), 1169 (C-N stretch) Anal. Calcd. For C13H18N6O3S: C, 46.14; H, 5.36; N, 24.84; Found: C, 46.17; H, 5.31; N, 24.83.

2-amino-6-{2-[4-(4-chlorophenyl)piperazin-1-yl]-2-oxoethyl}-4-methyl[1,3]thiazolo[4,5-d]pyrimidine-5,7(4H,6H)-dione (9)

Yellow solid; Yield: 67%; mp: 145–148°C; Rf: 0.69; 1H NMR (300MHz, CDCl3) δ p.p.m.: 3.25–3.13 (m, 4H), 3.56 (s, 3H), 3.68–3.57 (m, 4H), 4.760 (s, 2H), 6.99 (d, = 13.2 Hz, 2H), 7.269 (d, = 8.7, 2H); FT-IR (νmax; /cm, KBr): 3446 (N-H stretch), 2917 (C-H stretch), 1653 (C=O), 1495 (C=C stretch), 1233 (C-S stretch), 669 (Ar-Cl stretch); Anal. Calcd. For C18H19ClN6O3S: C, 49.71; H, 4.40; N, 19.32; Found: C, 49.69; H, 4.43; N, 19.37.

2-amino-4-methyl-6-{2-oxo-2-[4-(pyridin-2-yl)piperazin-1-yl]ethyl}[1,3]thiazolo[4,5-d]pyrimidine-5,7(4H,6H)-dione (10)

Yellow oil; Yield: 64%; Rf: 0.69; 1H NMR (300MHz, CDCl3) δ p.p.m.: 3.56 (s, 3H), 3.70–3.64 (m, 4H), 3.91–3.73 (m, 4H), 4.85 (s, 2H), 6.73 (q, = 6 Hz, 2H), 7.52 (d, = 6 Hz, 1H), 8.20 (s, 1H); FT-IR (νmax; /cm, KBr): 3462 (N-H stretch), 2926 (C-H stretch), 1652 (C=O), 1479 (C=C stretch), 1234 (C-S stretch), Anal. Calcd. For C17H19N7O3S: C, 50.86; H, 4.77; N, 24.43; Found: C, 50.88; H, 4.73; N, 24.39.

2-amino-4-methyl-6-{2-oxo-2-[4-(pyrimidin-2-yl)piperazin-1-yl]ethyl}[1,3]thiazolo[4,5-d]pyrimidine-5,7(4H,6H)-dione (11)

Yellow solid; Yield: 64%; mp: 134–140 °C; Rf: 0.69; 1H NMR (300MHz, CDCl3) δ p.p.m.: 3.56 (s, 3H), 3.70–3.64 (m, 4H), 3.91–3.73 (m, 4H), 4.28 (s, 2H), 6.66 (t, = 9 Hz, 1H), 8.41 (d, = 9 Hz, 2H); FT-IR (νmax; /cm, KBr): 3444(N-H stretch), 2924 (C-H stretch), 1493 (C=N), 1650 (C=O), 1236 (C-S stretch); Anal. Calcd. For C16H18N8O3S: C, 47.75; H, 4.51; N, 27.84; Found: C, 47.78; H, 4.48; N, 27.89.

2-amino-6-[2-(azepan-1-yl)-2-oxoethyl]-4-methyl[1,3]thiazolo[4,5-d]pyrimidine-5,7(4H,6H)-dione (12)

Brown oil; Yield: 64%; Rf: 0.65; 1H NMR (300MHz, CDCl3) δ p.p.m.: 1.62–1.47 (m, 4H), 1.85–1.74 (m, 4H), 3.55 (s, 3H), 4.15- 4.04 (m, 4H), 4.65 (s, 2H), 5.82 (s, 2H); FT-IR (νmax; /cm, KBr): 3312 (N-H stretch), 2925 (C-H stretch), 1651 (C=O); Anal. Calcd. For C14H19N5O3S: C, 49.84; H, 5.68; N, 20.76; Found: C, 49.87; H, 5.62; N, 20.78.

2-amino-4-methyl-6-[2-oxo-2-(piperidin-1-yl)ethyl][1,3]thiazolo[4,5-d]pyrimidine-5,7(4H,6H)-dione (13)

Yellow oil; Yield: 68%; Rf: 0.65; 1H NMR (300MHz, CDCl3) δ p.p.m.: 1.28–1.25 (m, 2H), 2.65–2.52 (m, 4H), 3.55 (s, 3H), 3.70 (t, = 12 Hz, 4H), 4.17 (s, 2H), 6.55 (s, 2H); FT-IR (νmax; /cm, KBr): 3475 (N-H stretch), 2927 (C-H stretch), 1654 (C=O); Anal. Calcd. For C13H17N5O3S: C, 48.28; H, 5.30; N, 21.66; Found: C, 48.23; H, 5.32; N, 21.67.

General procedure for synthesis of compounds (1418)

To a solution of 2-amino-4-methyl-4H-thiazolo [4,5-d]pyrimidine-5,7-dione (0.25 mmol) in dry DMF (1 ml) was added anhydrous potassium carbonate (0.75 mmol) under vigorous stirring. A solution of alkyl chloride (0.25 mmol) in dry DMF was added drop wise to the reaction mixture. The reaction mixture was heated for 2 h at 70° C and status of reaction was monitored by TLC using ethylacetate: hexane as mobile phase (6:4). The reaction mixture was cooled to room temperature. The product was filtered and washed twice with cold water. The final compounds obtained were dried under vacuum. Recrystallization was carried out using appropriate solvent.

2-amino-6-benzyl-4-methyl[1,3]thiazolo[4,5-d]pyrimidine-5,7(4H,6H)-dione (14)

Yellow solid; Yield: 74%; mp: 142–146 °C; Rf: 0.65; 1H NMR (300MHz, CDCl3) δ p.p.m.: 3.54 (s, 3H), 5.19 (s, 2H), 5.74 (s, 2H), 7.30–7.26 (m, 3H), 7.55 (t, J = 15, 2H); FT-IR (νmax; /cm, KBr): 3440 (N-Hstretch), 1643 (C=O), 1429 (C-N); Anal. Calcd. For C13H12N4O2S: C, 54.15; H, 4.20; N, 19.43; Found: C, 54.17; H, 4.23; N, 19.47.

2-amino-4-methyl-6-[2-(piperidin-1-yl)ethyl][1,3]thiazolo[4,5-d]pyrimidine-5,7(4H,6H)-dione (15)

Yellow solid; Yield: 78%; mp:145–147 °C; Rf: 0.65; 1H NMR (300MHz, CDCl3) δ p.p.m.: 1.28–1.25 (m, 2H), 2.65–2.52 (m, 4H), 3.41 (s, 3H), 3.70 (t, = 12 Hz, 4H), 4.17 (t, = 12 Hz, 2H), 5.68 (s, 2H); FT-IR (νmax; /cm, KBr): 3321(N-H stretch), 2925 (C-H stretch), 1646 (C=O), 1440 (C-N); Anal. Calcd. For C13H19N5O2S: C, 50.47; H, 6.19; N, 22.64; Found: C, 50.49; H, 6.22; N, 22.61.

2-amino-4-methyl-6-[2-(morpholin-4-yl)ethyl][1,3]thiazolo[4,5-d]pyrimidine-5,7(4H,6H)-dione (16)

Light brown solid; Yield: 78%; mp:150–156 °C; Rf: 0.65; 1H NMR (300MHz, CDCl3) δ p.p.m.: 2.65 (t, = 18 Hz, 4H), 3.52 (s, 3H), 3.70–3.61 (m, 4H), 4.17 (t, = 15 Hz, 4H), 5.68 (s, 2H); FT-IR (νmax; /cm, KBr): 3475 (N-Hstretch), 2927 (C-Hstretch), 1654 (C=O), 1457 (C-N); Anal. Calcd. For C12H17N5O3S: C, 46.29; H, 5.50; N, 22.49; Found: C, 46.31; H, 5.53; N, 22.50.

2-amino-6-[2-(diethylamino)ethyl]-4-methyl[1,3]thiazolo[4,5-d]pyrimidine-5,7(4H,6H)-dione (17)

Yellow solid; Yield: 75%; mp: 147–150 °C; Rf: 0.65; 1H NMR (300MHz, CDCl3) δ p.p.m.: 1.09 (t, = 24 Hz, 6H), 2.70 (q, = 27 Hz, 4H), 3.56 (s, 3H), 4.13 (t, = 15 Hz, 4H), 5.68 (s, 2H); FT-IR (νmax; /cm, KBr): 3447 (N-Hstretch), 2925 (C-Hstretch), 1457 (C-N), 1648 (C=O); Anal. Calcd. For C12H19N5O2S: C, 48.47; H, 6.44; N, 23.55; Found: C, 48.43; H, 6.42; N, 23.58.

2-amino-6-[2-(dimethylamino)ethyl]-4-methyl[1,3]thiazolo[4,5-d]pyrimidine-5,7(4H,6H)-dione (18)

White solid; Yield: 75%; mp:149–153 °C; Rf: 0.65; 1H NMR (300MHz, CDCl3) δ p.p.m.: 1.25 (s, 6H), 3.56 (s, 3H), 4.17 (t, = 15 Hz, 4H), 5.69 (s, 2H); FT-IR (νmax; /cm, KBr): 3447(N-H stretch), 2925 (C-H stretch), 1457 (C-N), 1648 (C=O); Anal. Calcd. For C10H15N5O2S: C, 44.60; H, 5.61; N, 26.00; Found: C, 44.63; H, 5.58; N, 26.04.

Pharmacological activity

The Institutional Animal Ethical Committee (IAEC) approved the experiments for oral glucose tolerance test (OGTT). Healthy Wistar rats used for the study were obtained from Animal House of Delhi Institute of Pharmaceutical Sciences and Research. Rats were housed in colony cages (3 female and 1 male rats per cage for breeding), at an ambient temperature of 25 °C with 12 h light: 12 h dark cycle. Animals were kept in polypropylene cages and were fed with rat food as pellets and water ad libitum unrestrictedly. All experimental procedures were performed according to IAEC (Protocol No.1/DIPSAR/IAEC/2010) and CPCSEA guidelines.

Induction of experimental diabetes (23,24)

Induction of diabetes was carried out using freshly prepared solution of streptozotocin (stz, Sigma-Aldrich) at the dose level of 90 mg/kg (I.P.) in 0.1 m freshly prepared citrated buffer pH 4.5 to 2 days old neonatal pups. A total of 24 diabetic and 6 non-diabetic male rats were used in the study. After 6 weeks of injection, animals were screened for fasting blood glucose level. Diabetic rats were further divided into four groups with six rats per group. Animals showing fasting glucose levels >150 mg/dL were considered diabetic and selected for screening of compounds.

Antihyperglycemic activity

Male Wistar rats weighing 150–200 g having blood glucose level >150 mg/dL were selected for OGTT. Animals of diabetic control group received vehicle (0.5% w/v methylcellulose) and standard group received vildagliptin (10 mg/kg). Animals of experimental group were administered single dose of test compounds (suspension in 0.5% w/v methylcellulose) of 10 mg/kg body weight (25,26). A glucose load (2 g/kg) was given orally to over night fasting rats after 30 min prior to administration of the test compound/vehicle. Blood samples were taken from tail vein at regular intervals of 0, 30, 60, 90, 120 min after glucose load. The blood glucose level was monitored using glucometer (ACCU-CHEK, Active; Roche Diagonistics, Mannheim, Germany).

Statistical evaluation

The statistical data were expressed as mean ± SEM IC50 values were determined using non-linear regression analysis. Statistical evaluation was performed by one-way anova followed by Dunnett’s post-test. Statistical studies and data analyses were performed using graphpad prism Version 5.0 (GraphPad Software Inc., San Diego, CA, USA).

In-vitro DPP IV enzyme inhibition assay

All the synthesized compounds were evaluated for in-vitro DPP IV enzyme inhibition. The enzyme assay was performed using DPP IV drug discovery kit (BML-AK 499; Enzo Life Sciences, Plymouth Meeting, PA, USA). The activity of test compounds was assayed using human recombinant DPP IV enzyme, chromogenic substrate (H-Gly-Pro-AMC, Km 114 μm), DPP IV inhibitor (P32/98), assay buffer and calibration standard as provided in the kit. The assay was performed using 96-well flat-bottomed microtiter plate followed by addition of assay buffer, DPP IV enzyme and chromogenic substrate (HGly-Pro-pNA). The assay principle and procedure was followed as per manufacturer’s guidelines. Solutions of the test compounds were made in dimethyl sulfoxide (DMSO) at different concentrations of 25, 50, 100, 200 μg/mL and 20 μL were added to each well after further dilutions. The plate was incubated at 37 °C for 10 min to allow enzyme-inhibitor reaction and read continuously at A-405 nm using Bio-Rad Elisa Plate Reader, Philadelphia, PA, USA.

Modelling studies

The binding mode and molecular interactions of the selected thizolopyrimidine derivatives were analysed using computational modelling. The structures of designed inhibitors were built and minimized, and subsequently simulated annealing was carried out to obtain the probable three-dimensional (3-D) conformation for these compounds with the help of CHARMm (27,28) force field parameters and partial charges based on MMFF94 (29). The crystal structure of DPP IV complexed with a structurally similar compound, a triazolopyridine derivative, (PDB Id: 2FJP) was taken as the template model (30). The binding site for docking studies was defined around the active site region (catalytic triad of Ser630, Asp708 and His740) including the S1 and S2 sub-pocket. The side chains of the binding site residues were kept flexible, during this analysis to allow for local perturbations in this region. The CHARMm force field parameters were assigned to the protein atoms of DPP IV. The obtained structure of thizolopyrimidine derivatives from above was docked into the defined binding site using LigandFit (31) docking protocol. The LigandFit docking algorithm combines a shape comparison filter with a Monte Carlo conformational search to generate docked poses consistent with the binding site shape. The docked poses have been scored using Dock Score scoring function. It is a force field-based scoring function that is combination of internal energy of the ligand and interaction energy between ligand and receptor in the docked complex. The docked pose with the highest Dock Score has been taken as the most probable binding conformation of the ligand. Discovery Studio (DS) 2.0 (Accelrys Inc., San Diego, CA, USA) and the protocols available in it were used for this (32).

Results and Discussion


A series of 2-amino-4-methyl-4H-thiazolo [4,5-d]pyrimidine-5,7-dione derivatives as DPP IV inhibitors were designed and synthetic strategy followed is outlined in Scheme 1. The preparation of 2-amino-4-methyl-4H-thiazolo [4,5-d]pyrimidine-5,7-dione (1) was easily synthesized in two steps via 6-amino-1-methyl-2,4-dioxo-1,2,3,4-tetrahydropyrimidin-5-yl-thiocyanate followed by cyclization, according to the previously reported procedure (22). The other part of synthesis involved preparation of N-substituted chloracetamides by nucleophilic substitution reaction of amines and chloroacetyl chloride [20]. Starting material used were various substituted secondary amines. These were subsequently reacted with chloroacetyl chloride to provide N-substituted chloracetamide intermediates in an excellent yield. Resulting intermediates and few commercially available alkyl chlorides were again condensed via N-3 alkylation of 2-amino-4-methyl-4H-thiazolo [4,5-d] pyrimidine-5,7-dione using anhydrous potassium carbonate as base at 70–80 °C. The title analogues were obtained in considerable yields after purification. The synthesis of proposed structures was found in conformity with spectral analysis.

                Scheme 1:

 (a) Br2/KSCN, DMF, (b) DMF, 80–120 °C, (c) NHR (a–h) secondary amines, diisopropylethylamine, 1,4-dioxan, reflux, 2 h, (d) 2–7, anhydrous K2CO3, DMF, 70–80 °C, (e) RCl (alkyl chlorides), anhydrous K2CO3, DMF, 70 °C.

In-vitro DPP IV activity

The synthesized thiazolopyrimidine derivatives (818) were evaluated in-vitro for DPP IV enzyme inhibitory activity and results obtained are reported as micromolar inhibitory concentration, IC50m). The DPP IV inhibition was determined using human DPP IV, and the inhibitory activity was obtained for various substitutions around thiazolopyrimidine scaffold. The unsubstituted thiazolopyrimidine 1 showed micromolar inhibition (IC50 = 11.87 μm). Further, we focused our optimization strategy towards effect of chemical substitutions on thiazolopyrimidine ring. The influence of three different substitution patterns having methylene, ethylene and acetamide linker was studied against their specificity towards DPP IV inhibition. Insertion of an acetamide linker produced potent inhibitors, which might have allowed conformational changes and may be responsible for increase in inhibitory effect (Table 1). The alkyl chlorides having ethylene and methylene linker (1418) caused considerably lower potency than N-substituted chloracetamide (813) derivatives.

Table 1.   DPP IV inhibitory effect of thiazolopyrimidine derivatives 1, 813 Thumbnail image of

N-substituted chloracetamide derivatives (27) which were synthesized from secondary amines and chloroacetyl chloride, have been explored due to their substrate specificity for DPP IV inhibition (20). An introduction of 4-aminopiperidine substituent 8 (IC50 = 0.98 μm) showed higher inhibition against DPP IV enzyme. Further, effect of substitutions on DPP IV inhibitory activity was evaluated using 6 or 7 membered cyclic amines. The azepanyl derivative 12 (IC50 = 38.27 μm) preserved the activity as compared to less bulky group as piperidine 13 (IC50 = 91.18 μm). Among the cyclic analogues, good correlation with variation in ring size was observed indicating minimum activity for piperidine analogue. Comparison of potency indicates that bulkier group can be well tolerated and decrease in bulkiness may be responsible for diminished inhibitory activity. Alternative approach for improving the inhibitory effect was explored by using bicyclic analogues such as substituted piperazine which offered highly potent compounds of series as 9, 10 and 11. Dramatic rise in potency was seen with substitution of chlorophenyl piperazine 9 (IC50 = 0.489 μm). It was observed that introduction of electron withdrawing chlorine atom at 4-position of phenyl ring is desirable for enhancement of activity. These results prompted us to introduce pyridyl piperazine 10 (IC50 = 0.329 μm) having an electron deficient aromatic pyridyl moiety which led the most potent compound of series. Conversely, no beneficial modification was seen with pyrimidyl piperazine 11 (IC50 = 1.494 μm) which showed weak inhibition. However, pyrimidyl piperazine derivative provided an inhibitor with rise in potency as compared to non-substituted scaffold 1.

The effect of alkyl chain length using methylene and ethylene linker was assessed to understand the alterations in activity as a replacement for acetamide linker (Table 2). The SAR revealed that benzylic moiety having methylene linker 14 (IC50 = 34.61 μm) may be tolerated as substitution. The incorporation of ethylpiperidine 15 (IC50 = 32.69 μm) showed comparable potency with that of compound 14. The potency was retained when N,N-diethylethamine 17 (IC50 = 2.46 μm) was introduced and replacement by ethylmorpholine analogue 16 (IC50 = 67.21 μm) leads to the drastic reduction in inhibitory potency. Disappointingly, inhibition was decreased sharply when N,N-dimethylethamine 18 (IC50 = 103.9 μm) was substituted, which suggest that decrease in alkyl chain length and replacement by methyl group may lead to decrease in inhibitory activity.

Table 2.   DPP IV inhibitory effect of thiazolopyrimidine derivatives, 1418 Thumbnail image of

In general, substantial decrease in potency was observed in few cases with removal of ring strain and reduction in alkyl chain length. Thus, we can say that incorporation of acetamide linker having carboxyl group attached next to heteroatom can effectively contribute to higher DPP IV enzyme inhibitory activity, which has validated our rational of study design.

Antihyperglycemic activity

The most active compounds 9 and 10 were chosen for in-vivo study on the basis of in-vitro DPP IV enzyme inhibitory activity. These compounds were orally administered at the dose of 10 mg/kg. In single dose study, both the compounds lead to lower plasma glucose levels beginning 30 min after glucose loading and 10 mg/kg dose significantly suppressed hyperglycaemia as compared with control group. The results indicate that area under blood glucose concentration-time curve (AUC) and OGTT significantly reduced (<0.0001) glucose excursion during 0–2 h. Vildagliptin showed 40.97% reduction of blood glucose at time of OGTT (at 1.5 h) at 10mg/kg. At the same time, compound 9 and 10 also showed glucose lowering up to 32.81% and 31.27%, respectively, at same dose. These results showed that the in-vivo efficacy of both the compounds was comparable to vildagliptin at 10 mg/kg as shown in Figure 2A,B.

Figure 2.

 Effect of compound 9 and 10 on blood glucose level in an oral glucose tolerance test (OGTT) using streptozotocin induced diabetic rats: (A) time course of changes in blood glucose level and (B) area under the blood glucose concentration-time curve (AUC) for the period of 0–2 h during OGTT. Data are expressed as mean ± SEM for 6 animals in each group. ***p < 0.0001 versus diabetic group.

Oral dosing of compounds 9 and 10 significantly decreased elevated glucose challenge and possesses therapeutic efficacy, thus improving glycemic control.

Modelling studies

Our concerned study is also focused to understand the binding mode of thiazolopyrimidne moiety and to investigate the necessary interactions required for DPP IV inhibition. On the basis of results obtained from in-vitro screening, compounds 1, 8, 9, 10, 11 and 18 were further evaluated for in silico studies. DPP IV molecule consists of two domains, an eight stranded beta-propeller domain at the N-terminus and a serine protease domain on the C-terminal end. The active site is located in the serine protease domain and comprises the catalytic triad Ser630, Asp708 and His740 which lies in a cave-like pocket bounded by hydrophobic residues. The protein around this active site comprises two subsites, a S1 pocket comprising residues Tyr631, Val656, Trp659, Tyr662, Tyr666 and Val711 and a S2 pocket lined by the side chains of residues Tyr547, Tyr631 and Pro550. The thiazolopyrimidine group of the derivatives docked into the binding site pocket occupying a position akin to that observed for triazolopyridine ring in biarylphenylalanine fluoropyrrolidine amide (compound 23) (PDB id: 2FJP) (28) (Figure S1). Moreover, the orientation of the thiazolopyrimidine nucleus in all the derivatives is found to be oriented perpendicular to the triazolopyridine ring of compound 23 (Figure 3). This orientation brings the aromatic interactions with Phe357 in an edge to face rather than parallel manner. Concurrently, it facilitates the formation of intermolecular hydrogen bonds between the nitrogen atoms of the thiazolopyrimidine ring and the hydroxyl group of Ser209 (Figure 4A–D) which has been implicated to be essential for DPP IV inhibitors (33).

Figure 3.

 The binding conformation of docked compounds in stick rendering and DPP IV (GRASP surface rendering) along with selected side chains in stick) (PDB id: 2FJP). The compound 1 is shown in grey colour, 8 in green, 9 in cyan, 10 in magenta, 11 in yellow and 17 in orange.

Figure 4.

 The docked complex of (A) compound 8, (B) compound 9, (C) compound 10 and (D) compound 11 in the binding pocket of DPP IV. The DPP IV is shown in cartoon rendering with selected binding site residues in (stick rendering in green) and compounds in (stick rendering in cyan).

The compound 8 has a piperidine ring, which does not occupy the S1 pocket. It lies near the catalytic triad which brings the primary amino group of piperidine within hydrogen bonding distance from the side chain atoms of the catalytic triad residues (Figure 4A). As a result, compound 8 acts as a potent inhibitor. The compound 9 has is bulkier, having chlorophenyl substituent at piperazine ring, in comparison with compound 8, and consequently, the thiazolopyrimidine ring moves towards Glu361 where it interacts both with the carboxylate group of this residue together with Arg358 (Figure 4B). However, the piperazine moiety in compound 9 unlike compound 8 lacks the capability to form hydrogen bonded interactions with the protein. The chlorophenyl group in compound 9 lies in the S1 pocket and interacts with the aromatic ring side chains of Tyr662 and Tyr666 (Figure 4B). This leads to a twofold increase in inhibitory activity of compound 9 as compared to compound 8 indicating that the occupation of the hydrophobic S1 pocket plays an important role in the inhibitory activity. This is also supported by a higher steric interaction energy (more negative) of compound 9 (−38.55 kcal/mol) in comparison with compound 8 (−29.21 kcal/mol) (Table 3).

Table 3.   Docking results of compounds (1, 811, 17) with DPP IV
LigandDocking energy (kcal/mol)Interacting residues with docked ligand (hydrogen-bonded residues are highlighted in bold)
Compound 8−29.21−21.67−50.88R125, E205, E206, S209, F357, R358, Y547, S630, Y662, Y666, R669, N710, H740
Compound 9−38.55−13.04−51.59R125, E205, E206, S209, F357, R358, E361, Y547, V656, S630, Y631, Y662, Y666, N710, V711
Compound 10−32.43−21.64−54.07R125, E205, E206, S209, F357, R358, E361, Y547, S630, Y631, Y662, Y666, N710, H740
Compound 11−33.19−15.74−48.93R125, E205, E206, S209, F357, R358, E361, Y547, S630, Y631, Y662, Y666, N710, H740
Compound 17−24.64−20.70−45.34 R125, E205, E206, V207, S209, F357, Y547, S630, Y631, Y662, Y666, R669
Compound 1−15.83−15.84−31.67E205, E206, V207, S209, F357, R358

In compounds 10 and 11, the aryl moiety (pyridine/pyrimidine) at the piperazine ring also resides in the S1 pocket and results in aromatic interactions resembling the chlorophenyl group of compound 9 (Figure 4C,D). The nitrogen (N23) atom in the pyridine/pyrimidine aryl moiety in compound 10 and 11 forms a hydrogen bond with hydroxyl group of Tyr666 (Figure 4C,D). However, the other nitrogen (N25) atom in pyrimidine aryl moiety of compound 11 results in a nearly fourfold lower activity as compared to pyridine containing compound 10. This is due to the resulting repulsive force existing between the similarly charged N25 atom in the ring and Nz atom of Arg125 as reflected by decrease in electrostatic interaction energy (less negative) (Table 3).

The compound 10 is more potent than compound 9 due to presence of an additional hydrogen bonded interaction between the nitrogen atom of pyridyl group and the side chain of Tyr666 (Figure 4C). This is also corroborated by better electrostatic interaction energy of compound 10 in comparison with compound 9, in spite of compound 9 having higher steric interaction energy (Table 3). However, compound 9 is more potent than compound 11 because of greater occupancy of S1 pocket as evident from increase in steric interaction energy (Table 3). The compound 17 has the same thiazolopyrimidine nucleus but lacks piperazine ring. Its ethyl substituent lies in the S1 pocket and thiazolopyrimidine nucleus occupies the adjacent pocket where it interacts with Arg125, Glu206 and Ser209. This is less potent than compound 8, 9, 10 and 11 ethyl substituent occupies S1 pocket less optimally as evident from lesser steric interaction energy (Table 3). The compound 1, thiazolopyrimidine nucleus, occupies the binding position similar to the other compounds in the series but in the opposite orientation. It forms strong face to face π-π interaction with aromatic ring of Phe357 in addition to hydrogen bonded interactions with Glu206 and Arg358.

The identified potent inhibitors of DPP IV, compounds 8, 9, 10, 11 and 17, have predicted to have good ADMET property. All of these satisfy four criteria of Lipinski’s rule of five for drug like compound. Their molecular weight ranges from 338 Da to 434 Da and calculated LogP value lies between 1.6 to −1.5. The maximum number of hydrogen bond acceptor is 8 in compound 11 and the number of hydrogen bond donor is either 1 or 2. Hence, it indicates that they should have good oral bioavailability. They also have moderate to good predicted score for blood–brain barrier penetration, CYP2D6 interactions and hepatotoxicity. Hence, they are very good drug like compounds that can provide good drug candidate.


In conclusion, we synthesized series of thiazolopyrimidine derivatives evaluated for inhibitory activity against DPP IV enzyme. Among these, compounds 9 (IC50 = 0.489 μm) and 10 (IC50 = 0.329 μm) showed excellent DPP IV inhibition and exhibited pronounced in-vivo antihyperglycemic activity in streptozotocin induced diabetic rat model. Furthermore, heterocyclic-substituted piperazines with an acetamide linker were indicated to offer the best potent DPP IV inhibitors. Molecular docking studies of active compounds displayed the probable binding interactions in the binding site of DPP IV. Thus, thiazolopyrimidine nucleus demonstrated a potential bioactive core for design of DPP IV inhibitors with desirable bioactivity for type 2 diabetes.


The funding of this study was supported by Government of NCT Delhi, India. One of the author, MS is also pleased to acknowledge Udaya P. Singh, SHIATS for his critical discussion and suggestions on the manuscript. The support from the Indian council of Medical Research in the form of the Bio-Medical Informatics Centre is gratefully acknowledged. The authors acknowledge Novartis Switzerland for providing vildagliptin.

Conflict of Interest