Crystal structures of four indole derivatives as possible cannabinoid allosteric antagonists

The crystal structures of four indole derivatives with various substituents at the 2-, 3- and 5-positions of the ring system are described. The dominant intermolecular interaction in each case is an N—H⋯O hydrogen bond, which generates either chains or inversion dimers. Weak C—H⋯O, C—H⋯π and π–π interactions occur in these structures but there is no consistent pattern amongst them. Two of these compounds act as modest enhancers of CB1 cannabanoid signalling and two are inactive.


Chemical context
The indole ring system is an important element of many natural and synthetic molecules with important biological activities (Biswal et al., 2012;Kaushik et al., 2013;Sharma et al., 2010). As part of our ongoing studies in this area, a group of indole derivatives with different substituents at the 2, 3 and 5-positions of the ring system were synthesised and tested as possible cannabinoid allosteric antagonists (Kerr, 2013). These compounds are analogues of 3-(2-nitro-1-phenylethyl)-2-phenyl-1H-indole (known as F087; see scheme), a positive allosteric modulator of CB1 (Adam et al., 2007).
The dihedral angle between the indole ring system and the phenyl ring is 81.69 (7) . Fig. 3 shows the molecular structure of (III). The r.m.s. deviation for the atoms making up the indole ring system is 0.013Å , and O3, C9 and C17 deviate from the mean plane by 0.0273 (12), À0.1302 (14), and 0.148 (1)Å , respectively. The dihedral angle between the indole ring plane and the C17-ring is 53.76 (3). This is similar to the equivalent value for (I), but the twist is in the opposite sense, as indicated by the C7-C8-C17-C22 torsion angle of À52. 40 (15) : in this case no intramolecular C-HÁ Á ÁO bond is present. The dihedral angle between the indole ring and the C11 ring is 67.12 (3) . The C8-C7-C9-H9, C8-C7-C9-C11 and C8-C7-C9-C10 torsion angles are À17, 102.46 (11) and À133.20 (10) , respectively, which are almost identical to the corresponding values for (II). These indicate that the C9-H9 bond is twisted away from the indole plane to the same side of the molecule as the nitro group: looking down the C9-C7 bond, C9-H9 is rotated in a clockwise sense with respect to the ring. The disposition of N2 and C11 about the C10-C9 bond is anti [torsion angle = À171.63 (8) ]. The methyl C atom of the methoxy group deviates from the indole plane by À0.1302 (14) Å , i.e. slightly towards the side of the molecule occupied by the C11 phenyl ring.
A view of the molecular structure of (IV) can be seen in Fig. 4. The indole ring system has an r.m.s. deviation of 0.008 Å for its nine non-hydrogen atoms and Cl1, C9 and C17 deviate from the mean plane by 0.009 (1), 0.093 (1) and À0.044 (1)Å . Thus, the displacement of C9 is slightly smaller than in the other three structures presented here. In terms of the orientation of the substituents at the 3-position of the indole ring, the C8-C7-C9-H9, C8-C7-C9-C11 and C8-C7-C9-C10 torsion angles are À17, 102.42 (14) and À133.94 (12) , respectively, which are very similar to the equivalent data for (II) and (III), again indicating that C9-H9 is twisted towards the nitro group. The N2-C10-C9-C11 torsion angle of 179.61 (9) shows that the anti orientation of N2 and C11 exactly mirrors that of the equivalent atoms in (II) and (III).
All-in-all, the conformations of (II), (III) and (IV) are very similar, especially in terms of the orientations of the substituents attached to C9 with respect to the indole ring. (I) differs slightly in that C9-H9 lies almost in the indole ring plane rather than being twisted away from it, which possibly correlates with the intramolecular C-HÁ Á ÁO interaction noted above. Of course, in every case, crystal symmetry generates an equal number of molecules of the opposite chirality (i.e., S configuration of C9), with an anticlockwise twist of C9-H9 with respect to the indole ring system.

Supramolecular features
As might be expected, the dominant supramolecular motif in all these compounds involve N-HÁ Á ÁO hydrogen bonds, although the resulting topologies [chains for (I) and (II) and dimers for (III) and (IV)] are different. Various weak interactions also occur, as described below and listed in Tables 1-4 Table 1. Table 1 Hydrogen-bond geometry (Å , ) for (I).
In (I), the N1-H1Á Á ÁO2 i [(i) = 1 2 À x, y À 1 2 , z] bond links the molecules into [100] chains with a C(8) chain motif (Fig. 5); adjacent molecules are related by b-glide symmetry. A PLATON (Spek, 2009) analysis of the packing in (I) indicated the presence of no fewer than four C-HÁ Á Á interactions, although the C10, C16 and C19 bonds must be very weak based on the long HÁ Á Á separation. Together, these links lead to a three-dimensional network in the crystal. There are no aromaticstacking interactions in (I), as the shortest ring centroid-centroid separation is greater than 4.6 Å .
The molecules of (II) are linked by N1-H1-O2 i [(i) = x, 1 2 À y, z À 1 2 ] hydrogen bonds into [001] chains ( Fig. 6) characterized by a C(8) motif: adjacent molecules are related by c-glide symmetry. Just one C-HÁ Á Á interaction occurs in the crystal of (II) but astacking interaction involving inversion-related pairs of C1-C6 benzene rings is also observed: the centroid-centroid separation is 3.7122 (16) Å and the slippage is 1.69 Å . The weak links connect the chains into a three-dimensional network.

Figure 6
Partial packing diagram for (II), showing the formation of [001] chains linked by N-HÁ Á ÁO hydrogen bonds (double-dashed lines). Symmetry code as in Table 2.

Figure 8
Fragment of an [010] chain in the crystal of (IV) linked by N-HÁ Á ÁO and C-HÁ Á ÁO hydrogen bonds (double-dashed lines). Symmetry codes as in Table 4.

Database survey
There are over 4000 indole derivatives with different substituents (including H) at the 2, 3 and 5 positions of the ring system reported in the Cambridge Structural Database (CSD; Groom & Allen, 2014). Narrowing the survey to indole derivatives with a C atom bonded to the 2-position of the ring and an sp 3 -hybridized C atom with two further C atoms and one H atom bonded to it at the 3-position (as per C9 in the present structures) yielded 72 hits. An analysis of the dihedral angle in these structures corresponding to C8-C7-C9-H9 in the present structures showed a wide spread of values with no obvious overall pattern.  2911,1738,1629,1581,1556,1445,1399,1283,1271,1215,1208,1145,1113,1077,874,852,761  Computer programs: CrystalClear (Rigaku, 2012), SHELXS97 and SHELXL97 (Sheldrick, 2008), ORTEP-3 for Windows (Farrugia, 2012) and publCIF (Westrip, 2010).

Refinement
Crystal data, data collection and structure refinement details are summarized in Table 5. The N-bound H atoms were located in difference maps and their positions freely refined. The C-bound H atoms were geometrically placed (C-H = 0.93-0.98 Å ) and refined as riding atoms. The constraint U iso (H) = 1.2U eq (carrier) or 1.5U eq (methyl carrier) was applied in all cases. The methyl H atoms (if any) were allowed to rotate, but not to tip, to best fit the electron density.

(I) Ethyl 3-(5-chloro-2-phenyl-1H-indol-3-yl)-3-phenylpropanoate
Special details Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > σ(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.

Special details
Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > σ(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.

Hydrogen-bond geometry (Å, º)
Cg2 and Cg4 are the centroids of the C1-C6 ring. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > σ(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å 2 )
x y z U iso */U eq C1 0.19835 (10) 0.50476 (11)  Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > σ(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.  (6)