T. Matsumoto et al.
Bioorganic & Medicinal Chemistry Letters 45 (2021) 128161
negative optical rotation ([
α
]25
D
ꢀ 19.6 in MeOH). In the EI-MS of 2, a
+
molecular ion peak [M] was observed at m/z 422 and the molecular
formula C26
30 5
H O was determined by HRMS measurement of the mo-
1
13
lecular ion peak. From the H and C NMR spectra of 2, the overall
1
4
structure was determined to be the same as (-)-linderol A, except for
the relative configurations of the monoterpene moiety. NOESY corre-
′
′
′
′
′
′
lations were observed between H-2 /H-3 , H-3 /H-8 , and H-7 /H-2
′
′
′
′
indicating that H-2 , H-3 , CH
Therefore, the relative configuration of 2 was determined to be
1S*,2R*,3S*,4R*). The absolute configurations of 2 were defined as
3
-7 , and CH-8 are all on the same side.
(
1
S,2R,3S,4R via comparison of the experimental and calculated ECD
1
1–13
data using the TDDFT method (Fig. 2).
This optically pure com-
Fig. 1. Chemical structures of isolated compounds from L. umbellata.
pound was first isolated as a natural product in this study, but racemic it
was previously reported as a synthetic intermediate by Dr. Yamashita et.
al. (data not shown).14 Mercifully, Dr. Yamashita provided us H and
1
13
C
Table 1
NMR spectra of epi-linderol A and they were corresponded with that of 2
recorded in this study. Based on all this evidence, the chemical structure
of 2 was determined to be epi-(-)-linderol A (Figure 1).
1
3
1
C NMR (150 MHz) and H NMR (600 MHz) Spectroscopic Data (CDCl
and 2.
3
) for 1
Position
1
2
The inhibitory effects of 1 and 2 against TCF/β-catenin transcrip-
tional activity (TOP activity) were evaluated using STF/293 cells. The
cells were 293 human embryonic kidney cells stably transfected with
modified M50 Super 8 × TOPFlash [luciferase reporter plasmid con-
taining downstream of the TCF-binding site] with hygromycin resistant
gene obtained from pGL4.32 vector as reported.15 In this assay, the
δ
C
δ
H
(J in Hz)
δ
C
H
δ (J in Hz)
1
2
3
4
5
7
8
9
1
1
1
1
1
1
1
2
156.6
112.3
179.8
100.4
165.1
119.1
135.2
134.6
127.3
128.9
129.4
103.4
161.0
113.6
161.7
93.2
6.12 (s)
6.10 (m)
6.63 (d, J = 15.8)
7.38 (d, J = 15.8)
166.6
191.2
143.5
125.7
135.2
129.1
128.4
130.4
128.4
129.1
70.7
inhibitory effects of the test samples were assessed by observing the
decrease in luciferase activity.1
6,17
It was observed that 1 exerted a
7.84 (d, J = 15.8)
7.06 (d, J = 15.8)
significant inhibitory effect without cytotoxicity. The chemical structure
of 1 was completely different from previous reported Wnt signal in-
hibitors such as ICG-001 and IWR-1. Interestingly, 2 with chalcone and
monoterpene moieties did not show an inhibitory effect (Fig. 3).
Therefore, it is suggested that the pyrone moiety is necessary for inhi-
bition of TOP activity.
0 and 14
7.50 (d, J = 7.6)
7.39 (t, J = 7.6)
7.38 (t, J = 7.6)
1 and 13
2
3
4
7.61 (d, J = 6.8)
7.42 (t, J = 6.8)
7.40 (t, J = 6.8)
7.42 (t, J = 6.8)
7.61 (d, J = 6.8)
5
′
′
82.7
37.2
Next, to evaluate the inhibitory effects on TOP activity and the
mechanism of 1 and its related compounds, we synthesized 1 from
α
1.77 (m)
β 1.82 (m)
3.54 (br-s)
1.26 (m)
′
′
′
18
3
4
5
27.4
49.2
22.3
92.8
41.0
46.5
4.37 (d, J = 5.5)
piperitone (1a) according to a previous report (Scheme 1). Briefly,
3.02 (dd, J = 5.5, 11.0)
catalytic reduction of 1a yielded piperitol (1b). Through acid-promoted
α
1.74 (m)
β1.11 (m)
1.62 (m)
1.17 (m)
N
S 1 substitution and intramolecular etherification, 1c (minor product)
′
′
and 1d (major product) were obtained from 1b. Finally, 1c was treated
6
7
39.1
27.8
α
20.4
36.1
α
1.62 (m)
β 1.16 (m)
1.79 (m)
◦
β2.10 (m)
with n-BuLi at ꢀ 78 C and benzaldehyde to obtain a reaction interme-
1.49 (s)
α
2
diate that was treated with Ac O/TEA followed by the addition of DBU
β 1.85 (m)
1.35 (s)
obtained (± )-1 (mixture of 1 and its enantiomer). HPLC separation using
a chiral column successfully produced optically pure 1 and its enan-
tiomer (ent-1). Using the same reaction, 1da (racemic mixture) was
synthesized from 1d. In addition, 4-methoxy-6-methylpyrone (1e) and
′
′
8
9
1
30.0
20.6
22.5
1.25 (m)
25.1
27.0
15.3
0.78 (d, J = 6.2)
1.20 (d, J = 6.2)
1.87 (m)
′
0
0.88 (d, J = 6.8)
5
,6-degydrokawain (1f) was easily synthesized for structure–activity
′
′
′
3
.54 (br-s, H-3 ), 1.26 (m, H-4 ), and 1.25 (m, H-8 )], and a quaternary
19,20
relationship stud-ies.
′
carbon bearing oxygen function [δ
each moiety of 1 was confirmed based on DQF COSY and HMBC spec-
C
82.7 (C-1 )]}. The connection of
The inhibitory effects against TOP activity of the piperitone (1a),
synthetic intermediates (1c and 1d), and related compounds (ent-1, 1d,
troscopy (Fig. 2). Namely, HMBC correlations were observed between H-
1
e, and 1f) were evaluated. The enantiomer (ent-1) and diastereomer
′ ′ ′ ′ ′ ′ ′ ′ ′ ′
2
/C-1,4,7, H-8/C-9,10, H-3 /C-4,5,1 ,5 , H-7 /C-1 ,2 ,6 , H-9 /C-4 ,8 ,
(
1da) showed significant inhibitory effects, and their effects were
′
′
′
and H-10 /C-4 ,8 . The relative configuration of 1 was determined as
equivalent to those of natural product 1 (Fig. 3). Monoterpene (1a) and
pyrones (1e and 1f) did not display inhibitory effects. In contrast, syn-
thetic intermediates 1c and 1d having both moieties showed weak ef-
fects (Supporting Information). According to these data, some structure-
activity relationships were suggested. Namely, the stereochemistry of
the monoterpene moiety is not limited as a naturally occurring com-
pound for inhibitory effects against TOP activity. Both the monoterpene
and pyrone moiety are necessary, and the phenyl group at C-8 may
contribute to the enhancement of these effects.
′ ′ ′
(
1 S*,3 S*,4 S*) via analysis of their nuclear Overhauser effect spec-
′
′
troscopy (NOESY) spectra (Fig. 2). The NOESY cross-peaks of H-3 /H-4
′
′
indicated that H-3 and H-4 are on one side. In addition, in the process of
chemical synthesis of 1, the intermediate 1c that have same relative
configuration as 1 was obtained less amount than diastereomer 1b
(
Scheme 1). This result should be due to steric crowding and suggested
′
′
′
′
that the H-3 and H-4 are on one side. CH -7 and H-3 are on one side
3
because another configuration is impossible to exist based on its mo-
lecular strain. The absolute configurations of 1 were defined via com-
parison of the experimental and calculated electronic circular dichroism
The inhibitory effects against TOP activity of 1, ent-1, and 1da were
confirmed by the cell viability on HT-29 human colon cancer cell. Wnt
(
ECD) data using the time-dependent density functional theory (TDDFT)
21
signaling pathway contributes to HT-29 cell proliferation.
The
1
1–13
′ ′ ′
method (Fig. 2).
Thus, the (1 S,3 S,4 S) absolute configurations of 1
viability of HT-29 cells was significantly decreased by 1, ± 1, and 1da
treatment for 72 hr (Fig. 4) and 24 hr (Supporting Information). The
IC50 values of 1 (IC50: 32.4 ± 2.3) and ± 1 (IC50: 19.9 ± 7.6) were not
stronger than those of positive control ICG-001 (IC50: 10.8 ± 2.0) and
was defined. Based on all this evidence, the chemical structure of 1 was
determined as the unique pyrone derivative (Fig. 1).
Epi-(-)-linderol A (2, Fig. 1) was isolated as a yellowish gum with
2