2-(4-(4-(1H-indol-3-yl)butyl)piperazin-1-yl) Results and Discussion -N-cyclohexyl-N-(2-methoxyphenyl) acetamide(11o) Pharmacophore-based virtual screening White solid (135 mg, yield 37.4%). H NMR(400 MHz, The obtained pharmacophore model was shown in CDCl3)8 8.13(, 1H),7.57(d, J=7.8 Hz, 1H),7.36- Figure 1A. As a result of our virtual screening protocol, 16 7.32(, 2H), 7.16 (t, J=7.1 Hz, 1H), 7.08 (t, compounds were selected to purchase and submitted to J=7. 4 Hz, 1H), 7.04(dd, J=7.6, 1.6 Hz, 1H), 6.97- pharmacological experiments(Tables S1 and S2). To our 6.92(m, 3H), 4.52(tt, J=12.0, 3.5 Hz, 1H), 3.78(s, delight, three of them revealed moderate D3R activities 3H), 2.78-266(m, 4H), 2.49-2. 35(m, 10H),1.94-1.91 Their chemical structures and corresponding binding (m,1h),1.80(d,j=12.3Hz,1h),1.72(d,J=7.4Hz,assaysweresummarizedinFigure1bandTable1.com 1H),1.67(d, J=7.6 Hz, 1H),1.61-1.53(m, 4H), 1.44- pound 11a, with a high fit value and a core structure of in- 1. 24(m, 3H), 1.17(ddd, J= 24.7, 12.2, 3.6 Hz, 1H), dolepropylamine and N-phenylacetamide, matches the 0.96-0.88(m, 1H),0.81(ddd, J=25.1, 12.6, 3.6 Hz, pharmacophore model quite well(Figure S1)and repre 1H).13C-NMR(100 MHz, CDCl3)8 169.10, 156. 20, sents a novel class of D3R ligands. It was identified to 136.33, 131.38, 129.69, 127.46, 127.31, 121.66, bind hD3R with 2161 nM affinity and was thus chosen as 121.26, 120.54, 118.85, 116.34, 111.53, 111.01, 59.73, the lead compound for further optimization 5848,5521,54.96,53.11,5291,48.58,32.04,29.62 28.04, 26.52, 25.80. 25.76, 25.49, 25.02. ESI-MS m/z Rational design and structure-activity 034M+H relationships The structural analysis and the ligand-receptor interaction elucidated by molecular docking were investigated to N-cyclohexyl-2-(4-(4-(5-fluoro-1H-indol-3-yl)butyl) guide the structure modification and optimization of com piperazin-1-yl)-N-phenylacetamide(11p) pound 11a. Compound 11a is characterized by an indole White solid (110 mg, yield 29.5%). H NMR(400 MHz, head, a linear alkyl linker and the N-phenylacetamide tail CDCl3)8 8.16(s, 1H), 7. 38(dd, J=5.0, 1.7 Hz, 3H), connected to a piperazine moiety To rationalize the design 7.24 (dd, J=8.8, 4.4 Hz, 1H), 7.19(dd, J=9.7, of the derivatives, the structural model of the complex 2.3 Hz, 1H), 7.09-707(m, 2H), 6.98(s, 1H), 6.90(td, D3R-11a was constructed by combining molecular dock =9.0, 2.4 Hz, 1H),4.57(tt, J=12.0, 3.3 Hz, 1H), ing and all available experimental data(Figure 2A).Three 2.73(s, 2H), 2.69 (t, J=7. 3 Hz, 2H), 2.50-2.38(m, important interactions were identified in the D3R-11a 10H),1.81(d, J=10.8 Hz, 2H), 1.72-1.63(m, 2H), model: the conserved salt bridge interaction between the 1.60-1.54(m, 3H), 1.42-1.35 (m, 3H), 1.01(ddd, protonated nitrogen atom(N1)of 11a and the carboxylate J=249,123,32Hz,3H),0.950.85(m,2H) group of D3. 32; the cation-Tt contact between the proton NMR (100 MHZ, CDCl3)8169352, 168.48, 158.71, ated nitrogen atom and F6.51; and the hydrogen bond 156.39, 138.23, 132.79, 130.39, 129.14, 128.37, formed by the oxygen atom of carbonyl group in 11a and 127.84,127.74,12308,116.52,114.73,111 Y7. 35 in D3R. It indicates that the piperazine ring and the 11.56, 110.22, 109.96, 103.85, 103.62, 60.17, 58.25, carbonyl group are critical to the activity, as these indis 54.12, 52.80, 31.44, 29.70, 27. 72, 26.13, 25.71, 25.32, pensable interactions determined the binding orientation of 2487.ES-MSm/z4914M+H the head down into the orthosteric binding site(OBS enclosed by TM-Ill, -V, -V, -vIn)and the tail up to the sec. ond binding pocket (SBP; comprised of ECL2 and the N-cyclohexyl-2-(4-(4-(5-fluoro-1H-indol-3-yl)butyl) extracellular segments of TM-lll, -)(18). The hollow piperazin-1-yl)-N-(2-methoxyphenyl)acetamide space was found in the OBS and SBP(Figure 2A), sug (11q) gesting that 11a could be optimized by appending larger White solid (135 mg, yield 37.4%). H NMR(400 MHz, groups in the head and tail or lengthening the linker CDCl3)87.34(td, J=8.1, 1.7 Hz, 1H), 7.24(dd, J=8.8, Therefore, a series of IBA derivatives were designed, syn 4.4 Hz, 1H), 7.18(dd, J=9.7, 2.1 Hz, 1H), 7.03(dd, thesized, and bioassyed for D3R activity with the aim to J=7.6, 1.7 HZ, 1H), 6.99-686(m, 4H), 4.51(tt, improve the potency of this series of ligands (Table 2) J=120.35Hz,H,3.77(s,3H,2.95(s,1H),2.88(s, 1H), 2.78-267(m, 4H), 2.53-2.38(m, 9H), 1.91(d, As our lead compound 11a already carries an aromatic J=11. 8 Hz, 1H), 1.79(d, J=12.3 Hz, 1H), 1.72-1.52 head and a bulky tail, the length of the linker was first con- 3H), 1.16(ddd, J=24.7, 12.2, sidered to be incremented to fill the hollow space in the 6.6 Hz, 1H), 0.95-0.85(m, 1H), 0.80(ddd, J=25. 1, 12.5, active site and 11p was obtained, providing a delightful 3.6 Hz, 1H). C-NMR(100 MHZ, CDCl3)8 169.02, 158.69, improvement in the binding affinity(;=636 nM). To verify 156.18, 132.82, 131.35, 129.72, 127. 73, 127.26, 123.12, our predicted binding mode, the effect of the length of the 120.56, 116.41, 111.68, 111.55, 110.16, 109.89, 103.81, linker (n= 2-4)on affinity was further examined. Indeed, 103.58, 59.62, 58.25, 55.21, 55.01, 52.79, 32.03, 29.62, the affinity of 11o with a 4-carbon linker is superior to 27.74. 26.14. 25.79. 25.75. 25.48. 24.88. ESI-Ms m/z those of 11m with 2-carbon and/or 11k with 3-carbon 521.4M+H linkers. As predicted, it proved that the longer linker could Chem Biol Drug Des 2013: 82: 326-3352-(4-(4-(1H-indol-3-yl)butyl)piperazin-1-yl) -N-cyclohexyl-N-(2-methoxyphenyl) acetamide (11o) White solid (135 mg, yield 37.4%). 1 H NMR (400 MHz, CDCl3) d 8.13 (s, 1H), 7.57 (d, J = 7.8 Hz, 1H), 7.36– 7.32 (m, 2H), 7.16 (t, J = 7.1 Hz, 1H), 7.08 (t, J = 7.4 Hz, 1H), 7.04 (dd, J = 7.6, 1.6 Hz, 1H), 6.97– 6.92 (m, 3H), 4.52 (tt, J = 12.0, 3.5 Hz, 1H), 3.78 (s, 3H), 2.78–2.66 (m, 4H), 2.49–2.35 (m, 10H), 1.94–1.91 (m, 1H), 1.80 (d, J = 12.3 Hz, 1H), 1.72 (d, J = 7.4 Hz, 1H), 1.67 (d, J = 7.6 Hz, 1H), 1.61–1.53 (m, 4H), 1.44– 1.24 (m, 3H), 1.17 (ddd, J = 24.7, 12.2, 3.6 Hz, 1H), 0.96–0.88 (m, 1H), 0.81 (ddd, J = 25.1, 12.6, 3.6 Hz, 1H). 13C-NMR (100 MHz, CDCl3)d 169.10, 156.20, 136.33, 131.38, 129.69, 127.46, 127.31, 121.66, 121.26, 120.54, 118.85, 116.34, 111.53, 111.01, 59.73, 58.48, 55.21, 54.96, 53.11, 52.91, 48.58, 32.04, 29.62, 28.04, 26.52, 25.80, 25.76, 25.49, 25.02. ESI-MS m/z 503.4 [M + H]+ . N-cyclohexyl-2-(4-(4-(5-fluoro-1H-indol-3-yl)butyl) piperazin-1-yl)-N-phenylacetamide (11p) White solid (110 mg, yield 29.5%). 1 H NMR (400 MHz, CDCl3) d 8.16 (s, 1H), 7.38 (dd, J = 5.0, 1.7 Hz, 3H), 7.24 (dd, J = 8.8, 4.4 Hz, 1H), 7.19 (dd, J = 9.7, 2.3 Hz, 1H), 7.09–7.07 (m, 2H), 6.98 (s, 1H), 6.90 (td, J = 9.0, 2.4 Hz, 1H), 4.57 (tt, J = 12.0, 3.3 Hz, 1H), 2.73 (s, 2H), 2.69 (t, J = 7.3 Hz, 2H), 2.50–2.38 (m, 10H), 1.81 (d, J = 10.8 Hz, 2H), 1.72–1.63 (m, 2H), 1.60–1.54 (m, 3H), 1.42–1.35 (m, 3H), 1.01 (ddd, J = 24.9, 12.3, 3.2 Hz, 3H), 0.95–0.85 (m, 2H). 13C￾NMR (100 MHz, CDCl3) d169352, 168.48, 158.71, 156.39, 138.23, 132.79, 130.39, 129.14, 128.37, 127.84, 127.74, 123.08, 116.52, 114.73, 111.65, 111.56, 110.22, 109.96, 103.85, 103.62, 60.17, 58.25, 54.12, 52.80, 31.44, 29.70, 27.72, 26.13, 25.71, 25.32, 24.87. ESI-MS m/z 491.4 [M + H]+ . N-cyclohexyl-2-(4-(4-(5-fluoro-1H-indol-3-yl)butyl) piperazin-1-yl)-N-(2-methoxyphenyl)acetamide (11q) White solid (135 mg, yield 37.4%). 1 H NMR (400 MHz, CDCl3) d7.34 (td, J = 8.1, 1.7 Hz, 1H), 7.24 (dd, J = 8.8, 4.4 Hz, 1H), 7.18 (dd, J = 9.7, 2.1 Hz, 1H), 7.03 (dd, J = 7.6, 1.7 Hz, 1H), 6.99–6.86 (m, 4H), 4.51 (tt, J = 12.0, 3.5 Hz, 1H), 3.77 (s, 3H), 2.95 (s, 1H), 2.88 (s, 1H), 2.78–2.67 (m, 4H), 2.53–2.38 (m, 9H), 1.91 (d, J = 11.8 Hz, 1H), 1.79 (d, J = 12.3 Hz, 1H), 1.72–1.52 (m, 5H), 1.44–1.22 (m, 3H), 1.16 (ddd, J = 24.7, 12.2, 3.6 Hz, 1H), 0.95–0.85 (m, 1H), 0.80 (ddd, J = 25.1, 12.5, 3.6 Hz, 1H).13C-NMR (100 MHz, CDCl3)d 169.02, 158.69, 156.18, 132.82, 131.35, 129.72, 127.73, 127.26, 123.12, 120.56, 116.41, 111.68, 111.55, 110.16, 109.89, 103.81, 103.58, 59.62, 58.25, 55.21, 55.01, 52.79, 32.03, 29.62, 27.74, 26.14, 25.79, 25.75, 25.48, 24.88. ESI-MS m/z 521.4 [M + H]+ . Results and Discussion Pharmacophore-based virtual screening The obtained pharmacophore model was shown in Figure 1A. As a result of our virtual screening protocol, 16 compounds were selected to purchase and submitted to pharmacological experiments (Tables S1 and S2). To our delight, three of them revealed moderate D3R activities. Their chemical structures and corresponding binding assays were summarized in Figure 1B and Table 1. Com￾pound 11a, with a high fit value and a core structure of in￾dolepropylamine and N-phenylacetamide, matches the pharmacophore model quite well (Figure S1) and repre￾sents a novel class of D3R ligands. It was identified to bind hD3R with 2161 nM affinity and was thus chosen as the lead compound for further optimization. Rational design and structure-activity relationships The structural analysis and the ligand–receptor interaction elucidated by molecular docking were investigated to guide the structure modification and optimization of com￾pound 11a. Compound 11a is characterized by an indole head, a linear alkyl linker and the N-phenylacetamide tail connected to a piperazine moiety. To rationalize the design of the derivatives, the structural model of the complex D3R-11a was constructed by combining molecular dock￾ing and all available experimental data (Figure 2A). Three important interactions were identified in the D3R-11a model: the conserved salt bridge interaction between the protonated nitrogen atom (N1) of 11a and the carboxylate group of D3.32; the cation-p contact between the proton￾ated nitrogen atom and F6.51; and the hydrogen bond formed by the oxygen atom of carbonyl group in 11a and Y7.35 in D3R. It indicates that the piperazine ring and the carbonyl group are critical to the activity, as these indis￾pensable interactions determined the binding orientation of the head down into the orthosteric binding site (OBS; enclosed by TM-III, -V, -VI, -VII) and the tail up to the sec￾ond binding pocket (SBP; comprised of ECL2 and the extracellular segments of TM-III, -VII) (18). The hollow space was found in the OBS and SBP (Figure 2A), sug￾gesting that 11a could be optimized by appending larger groups in the head and tail or lengthening the linker. Therefore, a series of IBA derivatives were designed, syn￾thesized, and bioassyed for D3R activity with the aim to improve the potency of this series of ligands (Table 2). As our lead compound 11a already carries an aromatic head and a bulky tail, the length of the linker was first con￾sidered to be incremented to fill the hollow space in the active site and 11p was obtained, providing a delightful improvement in the binding affinity (Ki = 636 nM). To verify our predicted binding mode, the effect of the length of the linker (n = 2–4) on affinity was further examined. Indeed, the affinity of 11o with a 4-carbon linker is superior to those of 11m with 2-carbon and/or 11k with 3-carbon linkers. As predicted, it proved that the longer linker could 330 Chem Biol Drug Des 2013; 82: 326–335 Du et al