T. Nguyen et al. / Bioorg. Med. Chem. Lett. 24 (2014) 1830–1838
1831
n
N
R
R
R
HN
R
HN
NH(CH2)nNR2
OCH3
OCH3
OCH3
OCH3
N
Cl
N
Cl
N
Cl
N
Cl
N
Group 4
Group 3
Group 5
Group 2
Group 1
Substituted Acridines
R
N
R
Cl
Cl
N
Substituted Quinolines
Substituted Tetrahydroacridines
Figure 1. Overview of scaffolds investigated for neuroprotective properties.
Group 6, the 3-chloro-5,6,7,8-tetrahydroacridin-9-amino ring was
differentially substituted at the 9-amino function and the same ap-
plied to Group 7 which comprised functionalized 7-chloroquino-
lin-4-amines. Structures of Groups 1–7 are detailed in Table 1.
Several of these compounds have been previously reported.24,25
Those that have not (18, 20, 21, 26–31, 39, 40, 44, 49, 50, 52, 53,
55, 56, 58 and 59) were prepared by reaction sequences which
are briefly described in the following paragraphs.
To assess the protective effects of the test compound, viability
of HT22 cells were determined after exposure to glutamate
(5 mM) and varying concentrations of test compound for 24 h. Glu-
tamate (5 mM) reduced viability of HT22 cells to 10% or less after
this time. The concentration of test compound required to retain
viability in 50% of cells co-treated with glutamate (protective
EC50) was derived from dose response curves (Supplementary
data). As cell viability would be affected by the intrinsic cytotoxic-
ity of the test compound, the experiment was repeated under sim-
ilar conditions in the absence of glutamate. In this way, the
cytotoxic EC50 of test compound was obtained. Desirable com-
pounds were those that had protective EC50 which were lower than
Compounds 18, 20, 21 and 26–31 were obtained by reacting 6,9-
dichloro-2-methoxyacridine with the relevant aniline in ethanol.
Catalytic amounts of hydrochloric acid were added to drive
the nucleophilic displacement of the 9-chloro functionality by
the reacting amine. 3-[4-(4-Aminophenyl)-piperazin-1-yl)-pro-
pan-1-ol, [4-(4-aminophenyl)-piperazin-1-yl]-cyclohexylmetha-
none, and [4-(4-aminophenyl)-piperazin-1-yl]-phenylmethanone
which were the anilines required for the syntheses of 18, 20 and
21, respectively, were obtained by Hartwig–Buchwald amination
of iodobenzene in a palladium coupling reaction, followed by
catalytic reduction of the aromatic nitro group to give the desired
aniline (Scheme 1). For 26–31, the anilines were obtained from
the catalytic reduction of commercially available nitrobenzenes.
1-Chloro-4-(2-chloroethyl)-benzene was obtained by reacting
4-chlorophenylethanol with thionyl chloride. It was then
reacted with tert-butylpiperidin-4-ylcarbamate to give tert-butyl
(1-(4-chlorophenenyl)piperidin-4-yl)carbamate. The carbamate
moiety was removed by hydrolysis with trifluoroacetic acid to give
1-(4-chlorophenethyl)piperidin-4-amine which was reacted with
6,7-dichloro-2-methoxyacridine in melted phenol to give 39
(Scheme 2). The same reaction sequence was followed for 40 except
that 1-methoxy-4-(2-bromoethyl)-benzene was the reactant in
step (a). When 6,9-dichloro-2-methoxyacridine was stirred in
melted phenol in the absence of an amine, 6-chloro-2-methoxy-
9-phenoxyacridine (44) was isolated as an intermediate in good
yield.
cytotoxic EC50
.
As seen from Table 1, only compounds in Group 2 and selected
members of Groups 5–7 had protective EC50 values. These ‘actives’
were distinguished by the presence of an aryl-NH-aryl residue that
was conspicuously absent among the inactive members of the
other groups. Of the two aryl rings present in the actives, one
was phenyl while the other was a heteroaromatic ring (acridine,
tetrahydroacridine, quinoline). Exceptions were the tetrahydroac-
ridines 49 and 52: 49 without the Aryl1-NH-Aryl2 motif retained
protective activity (EC50 2.22 lM) whereas 52 which had the motif
lacked protective activity. Of the heteroaromatic rings present in
the actives, the 6-chloro-2-methoxyacridinyl ring of Group 2 ap-
peared to be optimal. Notably, for the same substituent on the N-
phenyl ring, Group 2 compounds fared better than their counter-
parts in Group 5 (unsubstituted acridine), Group 6 (tetrahydroacri-
dine) and Group 7 (quinoline). Thus for 9, 48, 53 and 59 which have
the same 9-N-(4-diethylaminophenyl) substituent, potency was of
the order 9 (Group 2) > 48 (Group 5) > 53 (Group 6) > 59 (Group 7).
The same was true for 16, 54 and 60 which had the 9-N-4-(4-meth-
ylpiperazin-1-yl)-phenyl side chain. The potency of 16 (Group 2)
exceeded that of 54 (Group 6) and 59 (Group 7). The more potent
protective properties of the 6-chloro-2-methoxyacridine-9-amines
9 and 16 may be linked in part to their greater lipophilicities, as
seen from their estimated logP values (Table 1). The neuroprotec-
tive potencies of flavonoids and tyrphostins have been correlated
to their lipophilicities.15,20 The relationship may be traced to the
ability of lipophilic compounds to penetrate lipid bilayers, thus
gaining greater access to the cellular organelles involved in
oxytosis.
6-Chloro-1,2,3,4-tetrahydro-acridin-9-ylamine (49) was syn-
thesized by reacting 6,9-dichloro-1,2,3,4-tetrahydroacridine25 with
hydrazine, followed by reduction of the hydrazide with nickel chlo-
ride and sodium borohydride (Scheme 3). The same method when
applied to 4,7-dichloroquinoline yielded 55.
The tetrahydroacridine-9-amines 50, 52, 53 and the quinoline-
4-amine 56 were synthesized by reacting 6,9-dichloro-1,2,3,4-tet-
rahydroacridine with the relevant commercially available amines
in melted phenol or ethanol as shown in Scheme 2.
In contrast to the 6-chloro-2-methoxyacridinyl ring which ap-
pears to be optimal for activity, considerable flexibility was per-