Letter
3
5
enantioselective have thus far been unsuccessful; however,
the enantiomers of both compounds (±)-6 and (±)-2 can be
resolved using preparative SFC with a chiral stationary phase.
X-ray crystallography confirmed the structure and absolute
stereochemistry of (+)-2.
In a preliminary screen, five pyrroloindoline compounds,
including (±)-2, were tested for modulation of eight
pentameric ligand-gated ion channels (pLGICs): muscle type
nAChR, α4β2 nAChR, α7 nAChR, 5-HT3A receptor, α1β2γ2
GABAA receptor, α1β2 GABAA receptor, GluR2, and the
Figure 1. Chemical structures of selected pyrroloindolines.
screened a representative collection of these structures and
found that compounds bearing aryl substitution at C8a can act
Although no GABA receptor activity has been
A
A
27,28
27,28
as PAMs of GABAA receptors.
Here, we report the
synthesis of pyrroloindoline (+)-2 (Figure 1) and modification
of this scaffold by substitution at N1, C3a, C5, and C8a,
previously reported for physostigmine, compound (±)-2
appears to selectively potentiate α1β2γ2 GABA
receptors
A
yielding a novel series of GABA receptor ligands. All of the
over other Cys-loop receptors.
Based on the selective PAM profile of (±)-2, we decided to
further characterize this ligand. We set out to determine if both
A
compounds were tested for agonism and allosteric modulation
properties at the human α1β2γ2 GABA receptor, the most
A
abundant GABA subtype in the adult brain, expressed in
enantiomers are active at the α1β2γ2 GABA receptor.
A
A
Xenopus laevis oocytes via two-electrode voltage clamp
electrophysiology. Additionally, we performed mutagenesis
experiments to identify the binding site of these ligands.
Enantiomer-specific effects would imply a specific drug−
receptor interaction, rather than some more generic effect such
as altering membrane properties. To assess functional effects,
we used a similar two-electrode voltage clamp protocol to one
previously described by Marotta et al. Briefly, the current
responses of three identical EC50 doses of GABA were
recorded, followed by a dose of the test-ligand at 40 μM.
After a 30 s incubation, a dose was applied containing both
GABA at its EC50 and the test-ligand at 40 μM. Finally, two
28
RESULTS AND DISCUSSION
The synthesis of the pyrroloindoline framework commenced
with protection of tryptamine (3) to provide carbamate 4
■
(Scheme 1). Pd-catalyzed C2 arylation with iodobenzene
doses of GABA EC were applied. The first three GABA doses
Scheme 1. Synthesis of the Pyrroloindoline Scaffold
50
establish a baseline of the GABA response at that
concentration, and the purpose of the last two GABA doses
is to verify proper functioning of the receptor post modulation
and control for independent rise in current amplitude. Of the
two (±)-2 enantiomers, only (+)-2 showed a meaningful
potentiation of the EC50 GABA dose, with a mean of 16 ±
4
.1%, as shown in Figure 2B.
To determine activity at the α1β2 subtype and consequent
involvement of the γ2 subunit in potentiation, we performed
the same experiment for this subtype. For (±)-2 a mean
potentiation of 28 ± 5.2% was observed (Figure 2). Similar to
the observations for the α1β2γ2 subtype, (+)-2 showed
2
.6% and 9.2 ± 1.1%, respectively (Figure 2B and Table S2).
These results demonstrate that the γ2 subunit is not required
for potentiation of the α1β2γ2 receptor by (±)-2.
The amplitude of potentiation is dependent on several
factors, among which are both the PAM concentration and the
GABA concentration at which we tested the modulation. Next,
we determined the effect of 40 μΜ (±)-2 on the GABA EC
5
0
(
ΔEC ((±)-2)) at the α1β2γ2 receptor. The observed (±)-2-
5
0
5
0
under microwave conditions gave 2-phenyl tryptamine 5 in
and Table S3. This shift is comparable to the induced shift
2
9
7
7% yield. Various approaches were investigated for effecting
seen for this subtype by the benzodiazepine Triazolam, 16−50
30
36
oxidative cyclization of 5. Although there are many examples
μΜ. Moreover, we wanted to determine the potency of the
of related cyclizations of tryptamine and tryptophan deriva-
pure enantiomer (+)-2. Well-studied modulators, such as
flurazepam and zolpidem, have EC s in the nanomolar range
2
4,31−35
tives,
we found that many of these conditions were
5
0
unsuitable for tryptamine 5, presumably due to the phenyl
substituent at C2. After extensive experimentation, it was
found that oxidative cyclization of 5 by treatment with N-
chlorosuccinimide followed by water afforded C3a-hydroxy
pyrroloindoline (±)-6 in 85% yield. Reduction of carbamate
when coapplied with GABA EC2−5, being 270 nM and 340
37
nM, respectively. The PAM tested here, (+)-2, appears to be
GABA EC , as shown in Figure 3B and Table S3.
5
Having established that pyrroloindoline (+)-2 acts as a PAM
(
±)-6 with Red-Al provided the N1-methyl pyrroloindoline
on the α1β2γ2 GABA receptor, further potentiation experi-
A
3
1
(
±)-2. Attempts to render the cyclization of 5 to 6
ments used the GABA EC10−15 instead of EC . Using the
50
B
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX