1854 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 7
Brief Articles
of at least 1 order of magnitude with respect to their
nucleus: the longer the side chains, the lower the
potency. Among the 5-chloro-4′-phenyl-unsubstituted
derivatives 22-24, the highest potency was exhibited
by 22 (Ki ) 2.8 nM), which bears two n-propyl groups
on the amide nitrogen. Insertion of a chlorine in both
the 5 and 4′ positions improved potency with respect to
the unsubstituted derivatives when the amide chain
featured n-propyl or n-butyl groups (compare 25 and 26
vs 4 and 5); otherwise, it was unfavorable (compare 27
vs 7, both bearing n-hexyl groups). In other words, the
effects on affinity of a chlorine in the 5 and 4′ positions
are not additive but depend on the nature of the N,N-
dialkyl chains. It is tempting to speculate that the
length of the 5,4′-dichloro-2-phenylindole scaffold within
the binding site restricts the mobility of the whole ligand
within the binding site. Under these steric constraints,
the L3 and L4 pockets are optimally filled by the smaller
n-propyl or n-butyl, compared with the bulkier n-hexyl
chains. To sum up, the lack of additive effects of
substituents NR1R2, R3, and R4 is somehow related to a
simultaneous increase in their dimensions, which makes
each single interaction at the three lipophilic sites less
effective.
The most potent indolylglyoxylamides (4-7, 9, 13-
27) were tested for their ability to increase pregnenolone
concentration in rat C6 glioma cells in comparison with
PK 11195 and Ro5-4864 (Table 1). Products 19, 26, and
27 proved to be highly effective in increasing preg-
nenolone production by more than 145% vs controls.
Several compounds (13, 14, 18, 20, 21, and 25) increased
pregnenolone accumulation with a potency similar to
or slightly higher than that of PK 11195 and Ro5-4864.
As a general trend, unsubstituted indolylglyoxylamides
(4-7, 9) did not produce a great effect on the preg-
nenolone level, whereas a better performance was
displayed by the 4′-substituted derivatives 13-21 and
25-27.
2-phenylindole-3-acetamide counterparts II.5
SARs will be discussed in light of our pharmacophore/
topological model of the PBR binding site8 made up of
three lipophilic pockets (L1, L3, and L4) and a H-bond
donor group (H1) (see Figure 1).
Among the unsubstituted indolylglyoxylamides 1-12
(R3 ) R4 ) H), the least effective were the N-monosub-
stituted derivatives 1 and 2 probably because they
cannot occupy both the L3 and L4 lipophilic pockets.
Thus, we did not further examine this type of substitu-
tion.
The N,N-disubstituted derivatives 3-9 exhibited
nanomolar affinity because they fill both the L3 and L4
sites. Compound 7, bearing two n-hexyl groups, was the
most potent one, with a Ki of 1.4 nM. A progressive
enhancement of affinity was generally observed by
increasing the length of the linear N-alkyl groups (3-
7), due to the progressively better filling of both the L3
and L4 lipophilic pockets, with 6 being the only excep-
tion. Branching of the N,N-dialkyl groups (8 and 9) gave
mixed results. Compound 8 (Ki ) 103 nM) showed a
surprising drop in affinity if compared with its un-
branched counterpart 3 (Ki ) 43 nM), while product 9
maintained the affinity with respect to 4. This affinity
difference between 8 and 9 does not find any easy
rational explanation either from the point of view of the
lipophilicity or from that of the steric hindrance. In any
case, the affinity data of 3-9 suggest that interactions
at the L3 and L4 sites are governed primarily by
lipophilicity and that steric factors may also play some
role.
Also, in the subseries of the cyclic amides 10-12, the
affinity was enhanced with increasing dimensions of the
cycle. However, 10 and 11 showed a considerable drop
in affinity with respect to the linear derivatives 3 and
4 probably because of their incapability of optimally
filling both the L3 and L4 sites, which are separated in
the binding site, while the more lipophilic and flexible
12 displayed an appreciable affinity (Ki ) 33 nM), again
confirming the role of lipophilicity in the interaction of
these ligands with the receptor site.
The three most potent derivatives 4, 5, and 7 were
selected as leads for further affinity optimization efforts.
Insertion of the electron-donating lipophilic methyl
group in the 4′-position of the 2-phenyl ring (R3) left the
affinity nearly unchanged because 19-21 were equi-
potent with the parent compounds 4, 5, and 7. By in-
troducing an electron-withdrawing substituent such as
chlorine or fluorine into the 4′-position, we obtained 13-
15 or 16-18, respectively, which exhibited a gain in
affinity of 2.5-fold to 7.5-fold. Among them, the N,N-di-
n-hexyl derivative 18 stood out as the most potent of
all the ligands, with a Ki of 0.37 nM, similar to that of
alpidem. These results suggest that the phenyl ring of
2-phenylindoles might be involved in a π-stacking
interaction with an electron-rich aromatic ring within
the L1 pocket and that this interaction is reinforced by
4′ electron-withdrawing substituents such as the halo-
gens.
Exp er im en ta l Section
Ch em istr y. General directions are in the Supporting
Information.
Gen er a l P r oced u r e for th e Syn th esis of N,N-Dia lk yl-
[5-su bstitu ted -2-(4-su bstitu ted p h en yl)in d ol-3-yl]glyoxyl-
a m id e Der iva tives 1-27. A solution of the appropriate amine
(2.75 mmol) in 50 mL of dry toluene was added dropwise to a
stirred suspension, at 0 °C, of chloride 38-43 (2.5 mmol) and
triethylamine (3.0 mmol) in 50 mL of the same solvent. The
reaction mixture was left to warm to room temperature, stirred
for 2-24 h (TLC analysis), and then filtered. In the case of
products 1 and 2, the precipitate was triturated with a
saturated NaHCO3 aqueous solution and washed with water
to give the first portion of crude product. The toluene solution
was washed with diluted HCl and then with a saturated
NaHCO3 aqueous solution and water, dried (MgSO4), and
evaporated to dryness to yield crude 3-27.
All products 1-27 were purified by recrystallization from
the appropriate solvent or by flash chromatography (eluting
system: petroleum benzine (60-80 °C) and ethyl acetate in
varying ratios). Yields, recrystallization solvents, melting
points, and spectral data are listed in Tables 1 and 2 in the
Supporting Information.
Su p p or tin g In for m a tion Ava ila ble: General chemistry
directions, synthesis and physical properties of compounds 33
and 35-43, Tables 1 and 2 containing yields, recrystallization
In the 4′-phenyl-substituted indoles, potency cor-
related with the N-alkyl length: n-hexyl > n-butyl >
n-propyl. This affinity trend was reversed when a
chlorine was introduced in the 5-position of the indole
1
solvents, melting points, the IR and H NMR spectral data of
1-27, and biological methods. This material is available free