Communications to the Editor
J ournal of Medicinal Chemistry, 2000, Vol. 43, No. 23 4357
this with the more optimal isopropylsulfonamide. Early
in our SAR studies we discovered that the platform in
which the distal phenyl of, e.g. 5e, was replaced with a
tert-butyl group also provided potent AMPA potentia-
tors. We hypothesized that the distal aromatic ring
might have a lipophilic interaction with the receptor
protein, and therefore tert-butyl could be a viable
replacement for this group. Our results with compounds
such as 19a appear to confirm our suspicions. The ready
availability of methyl (4-tert-butylphenyl)acetate as a
substrate for alkylation facilitated this aspect of the
SAR. We prepared compounds 19a -e which possess
respectively a methyl, ethyl, n-propyl, benzyl, and
phenylethyl substituent. Analogues 19a ,b, having either
a methyl or ethyl, were comparably active; the n-propyl
compound 19c was less active; and the two aromatic
substituted compounds 19d ,e were significantly less
active.
We prepared analogues of compounds 5e and 19b ,
which were identical except that they lacked the methyl
substituent adjacent to the aromatic ring (11 and 19f).
We found 11 and 19f were less active than 5e and 19b,
respectively, indicating the relative importance of the
2-arylpropylsulfonamide substructure found in our lead
compound 5a .
We examined replacement of the distal phenyl group
with the well-documented isosteric 3-thienyl group. To
our delight, we found that the activity of 13a or 13b
was significantly greater than their counterparts 5a or
5e, respectively, with 13b being nearly 7-fold more
potent than 5e and almost twice as potent as 22a .
F igu r e 1. Effects of AMPA receptor modulators on inward
currents evoked by glutamate (100 µM) in acutely isolated rat
cerebellar Purkinje neurons under whole-cell voltage clamp
recording conditions (Vh ) -70 mV). Currents were evoked
every 30 s by 10-s applications of glutamate. Compound was
applied, and 10-s applications of glutamate continued every
30 s until steady-state potentiation of the evoked currents was
reached. Data at each concentration of compound is expressed
relative to the potentiation observed with 2 (cyclothiazide; 100
µM). For each data point values are from at least 3 separate
cells.
5i > 5a . We also observed that these compounds were
significantly more potent than 2 and 4 in these native
AMPA receptors. The thienyl-substituted compound 13b
was 100-fold more potent than 2 and at least 1000-fold
more potent than 4 (when comparing the concentration
of compound required to produce a similar percentage
of the response elicited by cyclothiazide). Thus, these
are the most potent AMPA potentiators described to
date.
Herein we disclosed a novel series of biarylpropyl-
sulfonamides that are very potent potentiators of
responses mediated through AMPA receptors. SAR
studies demonstrated significant changes in potency
when the methylsulfonamide was changed to an iso-
propylsulfonamide and when the o-fluorophenyl group
of the lead 5a was changed to a tert-butyl, a 3-thienyl,
or a 4-cyanophenyl group. Further studies with this
series of compounds will be reported soon.
Finally, we directed our attention to substitution on
the distal aromatic ring of the biphenyl group, focusing
on substitution in the 4′-position. We first prepared the
unsubstituted biphenyl analogue 22a ; its activity was
about 4-fold better than that of 5e and 20-fold better
than that of the lead 5a . We explored a range of
electron-withdrawing and electron-donating substitu-
ents, including methyl (22b), trifluoromethyl (22c),
chloro (22d ), formyl (22e), carboxy (22f), cyano (22g),
and amino (22h ). While the compounds with a chloro-
or trifluoromethyl group were less active than the
unsubstituted derivative 22a , the carboxy analogue
was about equal in activity to 22a . Even greater potency
was observed for the methyl-, formyl-, cyano-, and
amino-substituted compounds, with 22b ,e,g about 4-
fold more potent than 22a , and 22h about 8-fold more
than 22a . All told, we realized a 150-fold increase in
AMPA potentiator potency versus our lead compound
5a .
A select group of compounds from this SAR (5a ,i and
13b) were evaluated using whole-cell voltage clamp
recordings on acutely isolated cerebellar Purkinje cells
to determine their ability to potentiate AMPA responses
on a native rat brain receptor population. Acutely
isolated cerebellar Purkinje neurons were isolated ac-
cording to methods previously described.17,18
Figure 1 shows a comparison of activities of com-
pounds 5a ,i and 13b, along with 2 (cyclothiazide) and
4 (CX516) with responses expressed as a percentage of
those evoked by 100 µM 2. These biarylpropylsulfon-
amide AMPA potentiators showed the same rank order
of potency in native AMPA receptors that we observed
in iGluR4 flip receptors in HEK-293 cells, with 13b >
Ack n ow led gm en t. The authors thank the Physical
Chemistry Department of Lilly Research Laboratories
for spectral data and elemental analyses.
Refer en ces
(1) Hollmann, M.; Heinemann, S. Cloned glutamate receptors.
Annu. Rev. Neurosci. 1994, 17, 31-108.
(2) Bleakman, D.; Lodge, D. Neuropharmacology of AMPA and
kainate receptors. Neuropharmacology 1998, 37, 1187-1204.
(3) Sommer, B.; Keina¨nen, K.; Verdoorn, T. A.; Wisden, W.; Burna-
shev, N.; Herb, A.; Ko¨hler, M.; Takagi, T.; Sakmann, B.; Seeburg,
P. H. Flip and flop: a cell-specific functional switch in glutamate-
operated channels of the CNS. Science 1990, 249, 1580-1585.
(4) Ito, I.; Tanabe, S.; Kohada, A.; Sugiyama, H. Allosteric modula-
tion of quisqualate receptors by a nootropic drug aniracetam. J .
Physiol. 1990, 424, 533-543. Copani, A.; Genazzani, A. A.;
Aleppo, G.; Casabono, G.; Canonico, P. L.; Scapagnini, U.;
Nicoletti, F. Nootropic drugs positively modulate R-amino-3-