4
566
H. Kataoka et al. / Bioorg. Med. Chem. Lett. 26 (2016) 4565–4567
In the case of a primary amine such as an amino acid ester, (E,E)-
azomethine ylide (W-shape) is generated in situ by the reaction
with an aldehyde, which gives a cis isomer. In contrast, in the case
of a secondary amine such as 5, (E,Z) or (Z,E) azomethine ylide,
which produces a trans isomer, would be generated dominantly
due to the steric stability. According to the synthetic route indi-
cated in Scheme 2, the dominancy between cis/trans conformation
in the 1,3-dipolar cycloaddition was completely inversed, thus we
obtained trans-6 as the major product. Subsequent co-deprotection
of tert-butyl esters and p-methoxybenzyl group with the acid
afforded trans-1a as expected.
Figure 1. Structure of fullerene derivatives 1a–e.
in this reaction were separated by silica gel column chromatogra-
1
The compounds cis-1a–e and trans-1a were identified by 1
phy. To identify cis/trans isomerism,
H
NMR analysis was
H
1
performed. In the NOESY experiments for cis-4d and cis-4e, a
cross-peak between the methine proton on C-5 position of the
pyrrolidine ring and the proton on the substituent (R) introduced
to C-2 position was observed. Conversely, in the case of trans-4d
and trans-4e, the cross-peak between the corresponding protons
in cis isomer was not observed. Furthermore, the observation that
the methine proton on C-5 of the pyrrolidine ring in the trans iso-
mer was downfield shifted compared to the cis isomer, which was
NMR and HRMS. The purity was verified by H NMR. All the prod-
ucts except cis-1a were synthesized as racemic mixtures.
NS5B inhibitory activities of the synthesized derivatives were
examined in a manner similar to our previous Letter with several
8
modifications. The examined fullerene derivatives 1a–e signifi-
cantly inhibited NS5B with IC50 values in the submicromolar range
(Table 1). The introduction of various substituents on the pyrro-
lidine ring of 1a, or cis/trans isomerism of 1a appears to have no
impact on NS5B inhibitory activity, although further investigation
is required to determine the structure–activity relationships in
1
3
consistent with the report Salvatore et al. previously described.
This downfield shift led us to identify the cis/trans isomerism of
the other precursors more easily. In these reactions, the cis isomers
of 4a–e were generated more dominantly than the trans isomers,
which would be due to the stability of azomethine ylides generated
in situ. The synthesized precursors 4a–e were subsequently trea-
ted with the strong Brønsted acid, trifluoromethanesulfonic acid,
for the deprotection of tert-butyl esters to afford the target pro-
line-type derivatives 1a–e.
In our previous Letter, we evaluated various biological activities
of 1a as diastereomeric mixtures.9 In the present study, we sepa-
rately prepared cis-4a and trans-4a to compare the inhibition activ-
ities between both diastereomers. According to the synthetic route
indicated in Scheme 1, the major isomer, cis-4a was purified by
silica gel column chromatography; however, the minor isomer,
trans-4a was obtained as an inseparable mixture with cis-4a.
Therefore, as shown in Scheme 2, trans-1a was synthesized alter-
detail. In our previous Letter, IC50 for the diastereomeric mixture
9
of 1a was determined to be 2.0
l
M. The lower IC50 values
observed for each isomer of 1a in the present study may be due
to different experimental conditions.
In addition, the NS3/4A inhibition assay was performed in a
manner reported by Sudo et al. previously with some modifica-
1
5
tions. All fullerene derivatives we examined inhibited NS3/4A
significantly in the submicromolar range (Table 1). Similar to the
results of NS5B inhibition assay, there is no remarkable difference
in the inhibitory activities of 1a–e, while cis-1a has slightly more
potent than other derivatives.
In conclusion, the present study demonstrates, for the first time,
that the fullerene derivatives inhibit HCV NS3/4A protease in addi-
tion to NS5B polymerase. The inhibition activities against the
enzymes were weaker than a known selective NS5B inhibitor,
1
4
natively, using N-(p-methoxybenzyl)glycine tert-butyl ester (5).
VX-222, and a clinically used NS3/4A inhibitor, telaprevir,
Scheme 1. Synthesis of novel proline-type fullerene derivatives.
Kataoka, Hiroki
