54 Journal of Medicinal Chemistry, 2008, Vol. 51, No. 1
Yanagita et al.
(2H, m, H-21), 2.44 (2H, m, H-9b and H-16), 2.71 (1H, ddd, J )
15.1, 8.7, 2.0 Hz, H-8a), 2.85 (3H, s, H-19), 2.93 (1H, m, H-8b),
3.10 (1H, d, J ) 10.8 Hz, H-13), 3.28 (1H, m, H-15a), 3.49 (1H,
m, H-15b), 4.03 (2H, m, H-20), 4.30 (1H, m, H-10), 5.27 (1H, d,
J ) 9.6 Hz, H-13), 6.90 (1H, s, H-2), 6.97 (1H, dd, J ) 7.0, 1.2
Hz, H-5), 7.11 (1H, dd, J ) 8.2, 7.0 Hz, H-6), 7.14 (1H, dd, J )
8.2, 1.2 Hz, H-7); for the r-trans-fold-like form, 0.88 (3H, d, J )
6.4 Hz, H-18), 1.37 (3H, d, J ) 5.3 Hz, H-17), 2.06 (1H, m, H-16),
3.06 (1H, m, H-10), 3.19 (3H, s, H-19), 3.23 (1H, d, J ) 10.7 Hz,
H-13), 5.53 (1H, d, J ) 7.0 Hz, H-11), 6.94 (1H, s, H-2), 7.08
(1H, dd, J ) 7.5, 1.3 Hz, H-5), 7.20 (1H, dd, J ) 7.8, 1.3 Hz,
H-7); for the twistlike form, 0.63 (3H, d, J ) 6.8 Hz, H-18), 1.00
(3H, d, J ) 6.3 Hz, H-17), 2.50 (1H, m, H-16), 2.94 (3H, s, H-19),
4.69 (1H, d, J ) 9.1 Hz, H-13), 4.85 (1H, m, H-10), 6.48 (1H, d,
J ) 7.6 Hz, H-5), 6.88 (1H, d, J ) 8.0 Hz, H-7), 6.91 (1H, s,
H-2). Other peaks in the r-trans-fold-like and twistlike forms had
weak intensities and overlapped those of the sofalike form. 13C
NMR δ (125 MHz, 300 K, CDCl3, 0.112 M): for the sofa form,
14.01 (C-25), 19.97 (C-18), 20.12 (C-8), 20.17 (C-17), 22.54 (C-
24), 25.45 (C-16), 26.71 (C-23), 20.03 (C-21), 31.41 (C-22), 35.59
(C-19), 35.69 (C-9), 46.47 (C-20), 50.17 (C-10), 65.85 (C-15), 75.32
(C-13), 107.26 (C-6), 117.27 (C-3), 120.61 (C-7), 121.89 (C-5),
124.52 (C-4′), 127.57 (C-2), 138.18 (C-7′), 146.35 (C-4), 173.27
(C-12). HR-EI-MS m/z: 399.2885 (M+, calcd for C24H37N3O2,
399.2886).
Synthesis of 5-Chloro-1-hexylindolactam-V10 (14). To a
solution of 5 (17.4 mg, 0.0436 mmol) in CH2Cl2 (1.0 mL) was
added N-chlorosuccinimide (5.8 mg, 0.0434 mmol) at 0 °C, and
the mixture was stirred for 15 min at 0 °C. The solvent was removed
in vacuo, and the residue was purified by HPLC using 85% MeOH
to give 14 (8.2 mg, 0.0189 mmol, 43%). Compound 14: [R]D -103°
(c 0.34, MeOH, 24.7 °C). UV λmax (MeOH) nm (ꢀ): 316 (6640),
239 (28 200). 1H NMR δ (400 MHz, 297 K, CDCl3, 0.012 M, sofa/
r-trans-fold ) 39.9:1): for the sofalike-form, 0.88 (3H, t, J ) 6.8
Hz, H-25), 0.98 (3H, d, J ) 6.4 Hz, H-18), 1.22 (1H, dd, J ) 11.2,
6.7 Hz, OH), 1.30 (6H, m, H-22–24), 1.36 (3H, d, J ) 6.9 Hz,
H-17), 1.42 (1H, m, H-9a), 1.77 (2H, m, H-21), 2.58 (2H, m, H-9b
and H-16), 2.80 (2H, m, H-8), 2.92 (3H, s, H-19), 3.06 (1H, m,
H-15a), 3.15 (1H, d, J ) 10.5 Hz, H-13), 3.39 (1H, m, H-15b),
4.01 (2H, t, J ) 7.2 Hz, H-20), 4.13 (1H, m, H-10), 5.32 (1H, br
d, J ) 7.8 Hz, H-10), 6.93 (1H, s, H-2), 7.08 (1H, d, J ) 8.7 Hz,
H-7), 7.21 (1H, d, J ) 8.7 Hz, H-6); for the r-trans-fold-like form,
0.78 (3H, s, H-18), 3.22 (3H, s, H-19), 3.30 (1H, d, J ) 9.1 Hz,
H-13). Other peaks in the r-trans-fold-like had weak intensities
and overlapped those of the sofalike form. 13C NMR δ (CDCl3,
0.038 M, 500 MHz, 300 K): for the sofalike form, 13.99 (C-25),
20.28 (C-8), 20.62 (C-17), 20.88 (C-18), 22.53 (C-24), 26.64 (C-
23), 27.37 (C-16), 30.10 (C-21), 31.37 (C-22), 33.74 (C-9), 35.70
(C-19), 46.53 (C-20), 49.35 (C-10), 66.03 (C-15), 70.92 (C-13),
108.62 (C-7), 116.92 (C-3), 123.59 (C-6), 126.48 (C-5), 126.68
(C-4′), 127.92 (C-2), 136.86 (C-7′), 140.31 (C-4), 172.79 (C-12).
HR-EI-MS m/z: 433.2493 (M+, calcd for C24H36N3O2Cl, 433.2496).
Synthesis of 2-Bromo-1-hexylindolactam-V10 (15). To a
solution of 5 (23.9 mg, 0.0599 mmol) in CH2Cl2 (1.0 mL) was
added N-bromosuccinimide (11.7 mg, 0.0657 mmol) at 0 °C, and
the mixture was stirred for 15 min at 0 °C. The solvent was removed
in vacuo, and the residue was purified by HPLC using 85% MeOH
to give 15 (13.1 mg, 0.0273 mmol, 46%). Compound 15: [R]D
-98.2° (c 0.59, MeOH, 26.7 °C). UV λmax (MeOH) nm (ꢀ): 297
(8400), 229 (30 300). 1H NMR δ (500 MHz, 300 K, CDCl3, 0.055
M, sofa/r-trans-fold ) 4.7:1): for the sofalike form, 0.88 (3H, t, J
) 6.8 Hz, H-25), 0.94 (3H, d, J ) 6.5 Hz, H-18), 1.27 (1H, d, J )
6.7 Hz, H-17), 1.30–1.34 (6H, m, H-22–24), 1.42 (1H, m, H-9a),
1.73 (2H, m, H-21), 2.37 (1H, m, H-9b), 2.46 (1H, m, H-16), 2.73
(1H, ddd, J ) 15.3, 7.7, 2.3 Hz, H-8a), 2.81 (3H, s, H-19), 2.91
(1H, ddd, J ) 15.3, 9.2, 2.3 Hz, H-8b) 3.09 (1H, d, J ) 10.8 Hz,
H-13), 3.31 (1H, dd, J ) 10.5, 5.7 Hz, H-15a), 3.54 (1H, br d, J )
10.8 Hz, H-15b), 4.15 (1H, t, J ) 7.6 Hz, H-20), 4.34 (1H, m,
H-10), 5.11 (1H, br d, J ) 9.7 Hz, H-11), 6.98 (1H, dd, J ) 6.3,
2.2 Hz, H-5), 7.12 (2H, m, H-6,7); for the r-trans-fold-like form,
0.88 (3H, t, J ) 6.8 Hz, H-25), 0.89 (3H, d, J ) 5.9 Hz, H-18),
1.32–1.34 (6H, m, H-22–24), 1.37 (3H, d, J ) 6.4 Hz, H-17), 1.73
(2H, m, H-21), 2.13 (1H, m, H-16), 2.13 (1H, m, H-16), 2.79 (1H,
m, H-8a), 3.05 (1H, m, H-10), 3.16 (3H, s, H-19), 3.23 (1H, d, J
) 10.7 Hz, H-13), 3.37 (1H, m, H-8b), 3.54 (1H, m, H-15a), 3.61
(1H, m, H-15b), 4.13 (2H, t, J ) 7.6 Hz, H-20), 5.43 (1H, br d, J
) 6.9 Hz, H-11), 7.12 (2H, m, H-5,6), 7.18 (1H, d, J ) 7.5 Hz,
H-7). 13C NMR δ (CDCl3, 0.055 M, 500 MHz, 300 K): for the
sofalike form, 14.00 (C-25), 19.93 (C-18), 20.23 (C-17), 20.37 (C-
9), 22.54 (C-24), 25.40 (C-16), 26.53 (C-23), 29.68 (C-21), 31.46
(C-22), 32.50 (C-8), 35.67 (C-19), 45.38 (C-20), 50.44 (C-10), 65.67
(C-15), 75.40 (C-13), 107.35 (C-6), 115.12 (C-2), 116.05 (C-3),
121.43 (C-5), 122.29 (C-7), 124.06 (C-4′), 138.03 (C-7′), 145.40
(C-4), 172.85 (C-12). HR-EI-MS m/z: 477.1993 (M+, calcd for
C24H36N3O2Br, 477.1991).
Conformational Analysis of 1-Hexylindolactam-V10 (5). A
conformational search was carried out using the random search
function in Sybyl 7.1 (Tripos, Inc.). The hexyl group of 5 was
replaced with a methyl group to simplify the calculation. The
following conditions and parameters were used: all rotatable bonds
were perturbed; the energy cutoff was set to 15 kcal/mol; the
maximum number of search iterations and the maximum number
of hits were both set to 100 000; the rms threshold was set to 0.2
Å. The conformations produced by the random conformational
search were fully optimized using the MMFF94s force field. The
dielectric constant (ꢀ) was set to 5.0 (CHCl3). Powell’s method
was used for energy minimization until the gradient value was
smaller than 0.001 kcal mol-1 Å-1. As a result, 62 conformations
with energy values within 10 kcal/mol from the global minimum
were obtained and classified into eight groups (sofa, r-trans-fold,
trans-fold, r-sofa, r-trans-fold (two), fold, cis-sofa, and twist) on
the basis of Itai’s nomenclature (Supporting Information).32
Inhibition of Specific [3H]PDBu Binding to PKC Isozyme
C1 Peptides. The binding of [3H]PDBu to the PKC C1 peptides
was evaluated using the procedure of Sharkey and Blumberg41 with
modifications as reported previously37 with 50 mM Tris-maleate
buffer (pH 7.4 at 4 °C), 10–40 nM PKC C1 peptide, 20 nM
[3H]PDBu (16.3 Ci/mmol), 50 µg/mL 1,2-dioleoyl-sn-glycero-3-
phospho-L-serine, 3 mg/mL bovine γ-globulin, and various con-
centrations of inhibitor. Binding affinity was evaluated on the basis
of concentration required to cause 50% inhibition of the specific
binding of [3H]PDBu, IC50, which was calculated with PriProbit
1.63 software.59 The inhibition constant, Ki, was calculated using
the method of Sharkey and Blumberg.41 Although we used each
PKC C1 peptide in the range 10–40 nM, the concentration of the
properly folded peptide was estimated to be about 3 nM based on
the Bmax values of the Scatchard analyses reported previously.37
Therefore, the concentration of free PDBu will not markedly vary
over the dose response curve.
Synthesis of γ-C1A(P11dfP). γ-C1A(P11dfP) was prepared by
solid-phase Fmoc synthesis as reported previously.25 The Kd value
of [3H]PDBu for γ-C1A(P11dfP) in the presence of 1,2-dioleoyl-
sn-glycero-3-phospho-L-serine was 5.1 nM. The identity and purity
of the peptide were confirmed by MALDI-TOF-MS [average
molecular mass, 6133.10 (MH+, calcd 6132.61)] and HPLC (>98%
pure), respectively.
Docking Study. The crystal structure of the PKCδ C1B domain60
(PDB code 1ptq) was obtained from the Protein Data Bank.
Hydrogen atoms were added to the protein, and all of the hydrogens
were energetically minimized, keeping all heavy atoms fixed. The
hexyl groups of the ligands were replaced with methyl groups to
simplify the calculation. Each ligand was docked using GOLD53
(version 3.2) in 50 independent genetic algorithm (GA) runs, and
for each of these a maximum number of 100 000 GA operations
were performed on a single population of 50 individuals. Operator
weights for crossover, mutation, and migration in the entry box
were used as default parameters (95, 95, and 10, respectively),
as well as the hydrogen bonding (4.0 Å) and van der Waals (2.5
Å) parameters, and the ChemScore scoring function was selected.
The active site was created around ꢀ-NH of Gln-27, and the radius
was set to 10 Å, with the automatic active-site detection on. The
top-ranked solutions were geometrically optimized at the protein