Total syntheses and endoplasmic reticulum stress suppressive activities of hericenes A?C and their derivatives

Article abstract of DOI:10.1016/j.tetlet.2018.03.065

We report the syntheses and neuroprotective activities of hericenes and their derivatives against endoplasmic reticulum (ER) stress-dependent cell death. Four natural products, including hericenes A?C and hericenol A, and five synthetic derivatives were synthesized and their protective activities were evaluated. In designing the synthetic derivatives, we focused on the binding position of the fatty chain. Hericenes B and C showed moderate protective activity against thapsigargin-induced ER stress-dependent cell death. In contrast, their regioisomers (with respect to the position of the fatty chain) exhibited higher protective activity against tunicamycin-induced ER stress. This study clearly shows that the number and the binding position of the fatty chain are critical for protective activity against ER stress-dependent cell death.

Full text of DOI:10.1016/j.tetlet.2018.03.065

Accepted Manuscript
Total syntheses and endoplasmic reticulum stress suppressive activities of her-
icenes A−C and their derivatives
Shoji Kobayashi, Yoshiki Hamada, Takeshi Yasumoto, Yuuki Hashino, Araki
Masuyama, Kaoru Nagai
PII:
S0040-4039(18)30390-3
https://doi.org/10.1016/j.tetlet.2018.03.065
TETL 49834
DOI:
Reference:
Tetrahedron Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
13 February 2018
14 March 2018
22 March 2018
Please cite this article as: Kobayashi, S., Hamada, Y., Yasumoto, T., Hashino, Y., Masuyama, A., Nagai, K., Total
syntheses and endoplasmic reticulum stress suppressive activities of hericenes A−C and their derivatives,
Tetrahedron Letters (2018), doi: https://doi.org/10.1016/j.tetlet.2018.03.065
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Total syntheses and endoplasmic reticulum stress suppressive activities of hericenes A−C and
their derivatives
Shoji Kobayashia,*,† Yoshiki Hamadaa, Takeshi Yasumotoa, Yuuki Hashinoa, Araki Masuyamaa,
Kaoru Nagaib,*,†
a Department of Applied Chemistry, Faculty of Engineering, Osaka Institute of Technology
5-16-1 Ohmiya, Asahi-ku, Osaka 535-8585, Japan
b
Department of Nutrition, Koshien University, 10-1, Momijigaoka, Takarazuka, Hyogo,
665-0006, Japan
* Corresponding authors. Tel.: +81 6 6954 4081 (S.K.), +81 797 87 8086 (K.N.)
E-mail addresses: shoji.kobayashi@oit.ac.jp (S. Kobayashi), k-nagai@koshien.ac.jp (K. Nagai)
† These authors contributed equally to this work.
ABSTRACT
We report the syntheses and neuroprotective activities of hericenes and their derivatives against
endoplasmic reticulum (ER) stress-dependent cell death. Four natural products, including
hericenes A−C and hericenol A, and five synthetic derivatives were synthesized and their
protective activities were evaluated. In designing the synthetic derivatives, we focused on the
binding position of the fatty chain. Hericenes B and C showed moderate protective activity
against thapsigargin-induced ER stress-dependent cell death. In contrast, their regioisomers (with
respect to the position of the fatty chain) exhibited higher protective activity against
tunicamycin-induced ER stress. This study clearly shows that the number and the binding
position of the fatty chain are critical for protective activity against ER stress-dependent cell
1

death.
Keywords:
Geranyl resorcylate; Total synthesis; Structure-activity relationship; Endoplasmic reticulum
stress; Neuroprotective agent
Hericium erinaceum, an edible mushroom named yamabushitake in Japan, houtouku in China,
and lion’s mane mushroom in Europe, is known as a medicinal food that is effective for the
prevention of various diseases, including cancer, inflammation, diabetes, immune disorders, and
dementia.1 Its unique and broader medicinal effects have stimulated many scientists to elucidate
the structures and compositions of its bioactive components and their modes of actions.2 Among
its reported medicinal virtues, the observed anti-dementia effect is specific to this mushroom.1d–f
Kawagishi et al. showed in a pioneering study that hericenones C−E isolated from the fruiting
bodies of H. erinaceum stimulate mouse astrocytes to produce nerve growth factor (NGF), a
neurotrophic factor responsible for the growth, maintenance, and survival of neurons.3 In 2014,
the Sabaratnam group independently demonstrated that hericenones C−E increased NGF levels in
culture medium and potentiated neurite outgrowth in PC12 cells when induced with a low
concentration of NGF.4
Kawagishi and one of the authors of this paper reported the neuroprotective effect of
3-hydroxyhericenone F (1a) against endoplasmic reticulum (ER) stress-dependent cell death (Fig.
1).5 The ER is the major organelle responsible for Ca2+ storage and release, as well as protein
synthesis, folding, and export. Accumulation of unfolded or misfolded proteins and alterations in
calcium homeostasis lead to a disturbance in ER function, which is collectively called ER stress.6
While various signal transduction pathways, including the unfolded protein response, are
activated in response to ER stress, intense and prolonged ER stress ultimately leads to apoptosis.7
A number of pathological analyses suggest that ER stress is involved in neurodegenerative
2

disorders.8 Therefore, ER stress suppressive compounds that enhance neuronal cell viability have
attracted attention for the prevention and treatment of neurodegenerative disorders such as
Alzheimer’s, Parkinson’s, and prion diseases.9
Structurally, 3-hydroxyhericenone F (1a) consists of a poly-substituted chromane core and a
palmitate chain (Fig. 1). It can be assumed that hericenone C (i.e., 5′-oxohericene A, 1b) is a
biosynthetic precursor of 1a and the tetrahydropyran ring is constructed by epoxidation of the
C2′-C3′ double bond, followed by 6-endo cyclization. Another characteristic of 1a is the presence
of a hydroxyl group at C3, which is not found in other naturally-occurring geranyl
resorcylates.1b,2 Of the fungal-derived geranyl resorcylates isolated to date, only 1a has been
shown to possess cytoprotective activity against ER stress-dependent cell death.5 It remains
unknown if other geranyl resorcylates possess similar protective activities. 10 Hence, we
undertook a synthesis and structure-activity relationship (SAR) study to clarify the fundamental
structure critical for cytoprotective activity, and to discover novel drug lead compounds that can
suppress ER stress-dependent cell death. Herein, we report the synthesis and neuroprotective
activity of hericenes A−C (5′-deoxohericenones) (2a–c) and their derivatives. The results show
that the number and the binding position of the fatty chain are important for protective activity.
3

Fig. 1. Structures of 3-hydroxyhericenone F (1a), hericenone C (1b) and hericenes A−C (2a–c)
In 2011, we reported the first total syntheses of hericenone J and hericene A
(5′-deoxohericenone C, 2a), using CuBr2-mediated one-pot phthalide formation and Stille
coupling as key reactions.11,12 In the present study, the common intermediate 311 was prepared
using our established method (Scheme 1). Esterification of 3 with various acid chlorides afforded
monoesters 4a–c (38−47%), along with their regioisomers 5a–c (6−24%) and diesters 6a–c
(15−55%). Importantly, these reactions were initiated at low temperature to suppress
diesterification. In addition, LiCl was added with the expectation that one of the hydroxyl groups
would be deactivated by lithium-mediated chelation with a MOM-oxygen. Although the effect
was subtle, the natural-type compounds 4a–c prevailed over their regioisomers 5a–c. 13
Oxidation14,15 of 4a–c followed by deprotection11,16 afforded hericenes A–C (2a–c) in 58–62%
yield over two steps. Similarly, two selected regioisomers (7a and 7c) and diester 8a were
4

synthesized from 5a, 5c, and 6a, respectively. It is noted that in the synthesis of 2a–c,
deprotection was facilitated by the presence of the ortho formyl group, which could coordinate
with the Lewis acid, whereas this interaction was absent in the case of the regioisomers (7a and c)
and diester 8a. The low solubility of 6a in CH2Cl2 at low temperature may also account for the
unsatisfactory formation of 8a.
Scheme 1. Syntheses of hericenes A–C (2a–c) and their derivatives (7a, 7c and 8a). Reagents and
conditions: (a) RCOCl, pyridine, LiCl, THF, −78 or −30 °C to −30 ~ 0 °C; (b)
2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPO), PhI(OAc)2, CH2Cl2, rt; (c) 2-azaadamantane
N-oxyl (AZADO), PhI(OAc)2, CH2Cl2, rt; (d) Me2BBr, amylene, CH2Cl2, −78 °C; (e) TiCl4,
2,6-di-tert-butyl-4-methylpyridine, CH2Cl2, −78 to −30 °C.
To circumvent this deprotection issue, the regioisomer of hericene B (7b) and its diester
derivative (8b) were prepared by changing the phenolic protective group (Scheme 2). Thus, the
phenolic hydroxyl group of hericenone J (9), synthesized previously,11 was protected as its TBS
ether and the lactone moiety was reduced with LiAlH4. Unexpectedly, the phenolic TBS group
migrated to the primary alcohol and the regioisomeric alcohol 11 predominated over the expected
product 10. Furthermore, the fully deprotected triol 12 (i.e., hericenol A17) was isolated in 21%
5

yield, which represents the second total synthesis of this molecule.18 The alcohol 10 was then
esterified with oleoyl chloride and converted into hericene B (2b), the regioisomer 7b, and diester
8b.19
Scheme 2. Syntheses of hericene B (2b), the derivatives (7b and 8b) and hericenol A (12).
Reagents and conditions: (a) TBSOTf, 2,6-lutidine, rt; (b) LiAlH4, THF, 0 °C to rt; (c) oleoyl
chloride, pyridine, DMAP, CH2Cl2, −60 °C to rt; (d) AZADO, PhI(OAc)2, CH2Cl2, rt; (e) TBAF,
THF, −78 °C.
With these natural and synthetic hericenes in hand, a cell protection assay was carried out
using murine neuroblastoma cell line (Neuro2a).9e Tunicamycin (TM), an inhibitor of N-linked
glycosylation in the ER, and thapsigargin (TG), an inhibitor of sarcoplasmic/endoplasmic
reticulum Ca2+-ATPase, were used as inducers of ER stress.6,20 In a typical experiment, Neuro2a
cells were incubated with various concentrations of compounds (2a–c, 7a–c, 8a,b and 12) in the
absence or presence of TM (final concentration: 0.5 μg/mL) or TG (final concentration: 10 nM)
for 24 h (Fig. 2).
The results show that the natural hericenes, specifically hericenes B (2b) and C (2c), exhibited
moderate protective activity against TG-induced ER stress at concentrations above 30 μM. Their
protective activities were less than that of 1a, where significant protection against TG toxicity
6

was observed above 1 μg/mL (1.7 μM).5 For 2a–c, no significant protection was observed against
TM-induced ER stress, in contrast to 1a.5 It is noteworthy that a significant increase in cell
viability was observed when TM-induced ER stressed Neuro2a cells were incubated in the
presence of the regioisomers of hericene B (7b) and C (7c). An approximately 20% increase in
cell viability was observed at 10 μM, comparable to the activity of 1a where an approximately
10% increase in cell viability was observed at 10 μg/mL (17 μM).5 The repeated protection assay
with the regioisomers (7a-c) indicated that they also possessed moderate protective activity
21
against TG-induced cell death under the slightly different incubation conditions.
(Supplementary Fig. S1-S3). In contrast, the diesters (8a and 8b) showed no protective activity,
presumably due to the high hydrophobicity [calculated log P (clog P) = 19.7 for 8a and 20.9 for
8b, respectively]22 that was too large to dissolve in aqueous media and enter the cytoplasm.
Hericenol A (12) (clog P = 3.2), which lacks the ester moiety, showed significant toxicity rather
than protective activity, demonstrating the importance of one fatty ester component for protective
activity.23
7

**
A
D
G
hericene A (2a)
B
E
hericene B (2b)
C
F
I
hericene C (2c)
*
120
100
80
120
100
80
140
120
100
80
*
60
60
60
40
40
40
20
20
20
0
0
0
0
10 30 100
0
10 30 100
0
10 30 100
Sample
M)
0
10 30 100
0
10 30 100
0
10 30 100
Sample
M)
0
10 30 100
0
10 30 100
0
10 30 100
Sample
M)
(
(
(
+Tunicamycin +Thapsigargin
+Tunicamycin +Thapsigargin
+Tunicamycin +Thapsigargin
7a
7b
7c
*
140
140
120
100
80
140
120
100
80
*
120
100
80
60
40
20
0
*
60
60
40
40
20
20
0
0
Sample
M)
0
3
10 30
0
3
10 30
0
3
10 30
Sample
M)
0
3
10 30
0
3
10 30
0
3
10 30
Sample
M)
0
3
10 30
0
3
10 30
0
3
10 30
(
(
(
+Tunicamycin +Thapsigargin
+Tunicamycin +Thapsigargin
+Tunicamycin +Thapsigargin
hericenol A (12)
8a
H
8b
**
120
140
120
100
80
120
**
100
80
60
40
20
0
100
80
60
40
20
*
60
40
20
0
0
0
10 30 100
0
10 30 100
0
10 30 100
0
3
10 30
0
3
10 30
0
3
10 30
Sample
M)
Sample
M)
0
10 30 100
0
10 30 100
0
10 30 100
Sample
M)
(
(
(
+Tunicamycin +Thapsigargin
+Tunicamycin +Thapsigargin
+Tunicamycin +Thapsigargin
Fig. 2. Protective effects of (A) hericene A (2a), (B) hericene B (2b), (C) hericene C (2c), (D) the
regioisomer of hericene A (7a), (E) the regioisomer of hericene B (7b), (F) the regioisomer of
hericene C (7c), (G) the diester derivative of hericene A (8a), (H) the diester derivative of
hericene B (8b), and (I) hericenol A, against ER stress-dependent cell death. The cell viabilities
were analyzed by MTT assay, and the values were represented as means ± SEM of the relative
percentage of surviving cells compared with the untreated cells (n=6 for A, B and D−I, and n=12
for C). * p < 0.05, ** p < 0.01, Tukey−Kramer multiple comparison tests.
In conclusion, we have shown the neuroprotective activities of hericenes and their synthetic
derivatives for the first time. Hericenes B (2b) and C (2c) exhibited moderate protective activity
against TG-induced ER stress-dependent cell death, whereas their regioisomers (7b and 7c)
exhibited significant protective activity against TM-induced ER stress at a concentration of 10
μM. Although details regarding the protective mechanisms presently remain unknown, these
8

preliminary SAR data clearly show the importance of the fatty chain: namely, the binding
position of the fatty ester chain significantly affects protective activity. It is worth noting that the
synthetic isomers exhibited stronger activity than the natural products. This finding will
encourage synthetic chemists to design new ER stress-suppressive molecules aimed at the
prevention and treatment of neurodegenerative disorders.
Acknowledgements
This work was supported by JSPS KAKENHI Grant Number JP17K07779. We thank Ms.
Sayaka Kado and Ms. Makiko Fujinami, Center for Analytical Instrumentation, Chiba University,
for measurement of mass spectra.
A. Supplementary data
Supplementary data (synthetic procedures, spectral data for all synthetic compounds, and cell
protection assay) associated with this article can be found, in the online version, at @@@.
9

References
1
(a) Sokół S, Golak-Siwulska I, Sobieralski K, Siwulski M, Górka K. Acta Mycol. 2016;50:1-18;
(b) Thongbai B, Rapior S, Hyde KD, Wittstein K, Stadler M. Mycol Progress. 2015;14:325–23;
(c) Jiang S, Wang S, Sun Y, Zhang Q. Appl Microbiol Biotechnol. 2014;98:7661-7670; (d) Mori
K, Obara Y, Moriya T, Inatomi S, Nakahata N. Biomedical Research. 2011;32:67-72; (e)
Hazekawa M, Kataoka A, Hayakawa K, et al. J Health Sci. 2010;56:296-303; (f) Mori K, Inatomi
S, Ouchi K, Azumi Y, Tuchida T. Phytother Res. 2009;23:367-372.
2
For reviews: (a) Friedman M. J Agric Food Chem. 2015;63:7108-7123; (b) Ma B-J, Shen J-W,
Yu H-Y, Ruan Y, Wu T-T, Zhao X. Mycology. 2010;1:92-98.
3
Kawagishi H, Ando M, Sakamoto H, et al. Tetrahedron Lett. 1991;32:4561-4564.
4
Phan CW, Lee GS, Hong SL, et al. Food Funct. 2014;5:3160-3169.
5
Ueda K, Tsujimori M, Kodani S, et al. Bioorg Med Chem Lett. 2008;16:9467-9470.
6
Pahl HL. Physiol Rev. 1999;79:683-701.
7
Zhang K, Kaufman RJ. Identification and characterization of endoplasmic reticulum
stress‐induced apoptosis in vivo. In: Khosravi-Far R, Zakeri Z, Lockshin RA, Piacentini M. eds.
Programmed Cell Death, General Principles for Studying Cell Death, Part A, Methods in
Enzymology. Elsevier; 2008;442:395–419.
8
Lindholm D, Wootz H, Korhonen L. Cell Death Differ. 2006;13:385-392.
9
For representative ER stress-suppressive compounds: (a) Choi J-H, Suzuki T, Okumura H, et al.
J Nat Prod. 2014;77:1729-1733; (b) Wu J, Tokuyama S, Nagai K, et al. Angew Chem Int Ed.
2012;51:10820-10822; (c) Wu J, Fushimi K, Tokuyama S, et al. Biosci Biotechnol Biochem.
2011;75:1631-1634; (d) Ueda K, Kodani S, Kubo M, et al. Biosci Biotechnol Biochem.
2009;73:1908-1910; (e) Nagai K, Chiba A, Nishino T, Kubota T, Kawagishi H. J Nutr Biochem.
2006;17:525-530.
10
It was reported that hericenones I and J, the natural products isolated with 1, did not show
cytoprotective activity (ref 5).
11
Kobayashi S, Ando A, Kuroda H, Ejima S, Masuyama A, Ryu I. Tetrahedron
2011;67:9087-9092.
12
For other total syntheses of geranyl resorcylates from our laboratory: (a) Kobayashi S,
Tamanoi H, Hasegawa Y, Segawa Y, Masuyama A. J Org Chem. 2014;79:5227-5238; (b)
Kobayashi S, Inoue T, Ando A, Tamanoi H, Ryu I, Masuyama A. J Org Chem.
2012;77:5819-5822.
13
Probably the differences of selectivity are not due to the type of acylating agents but to the
minor differences of reaction conditions. Experimental procedures are presented in
Supplementary data.
10

14
(a) De Mico A, Margarita R, Parlanti L, Vescovi A, Piancatelli G. J Org Chem.
1997;62:6974-6977; (b) Shibuya M, Tomizawa M, Suzuki I, Iwabuchi Y. J Am Chem Soc.
2006;128:8412-8413.
15
After some trials, oxidation with AZADO turned out to be better than other oxidation
conditions such as Swern, Dess-Martin, and TEMPO/PhI(OAc)2 oxidations regardless of the type
of fatty esters.
16
Guindon Y, Yoakim C, Morton HE. J Org Chem. 1984;49:3912-3920.
17
Omolo JO, Anke H, Sterner O. Phytochemistry. 2002;60:431-435.
18
Cordes J, Calo F, Anderson K, et al. J Org Chem. 2012;77:652-657.
19
Low yields are probably attributed to the esterification step, during which migration of the
TBS group occurred to some extent. We also observed that diol 10 tended to degrade during
extraction and purification. Further optimization of the synthetic route or the modification of the
protective group was not investigated, as our initial purpose was to obtain preliminary SAR data.
20
Hitomi J, Katayama T, Taniguchi M, Honda A, Imaizumi K, Tohyama M. Neurosci Lett.
2004;357:127-130.
21
We found that cell reactivity varied to some extent depending on the incubation time, cell
passage number, cell density in medium, and other trivial factors. We repeatedly carried out
experiments with modified conditions until significant and reliable results were observed, in
consideration of inherent variability of cell reactivity. Figure 2 represents the results that were
carried out under certain fixed optimal conditions. Variability of blank cell viability may attribute
to the time when the data was collected. In practice, cell viability analyses were carried out in
parallel to the syntheses in order to gain preliminary information on the structure-activity
relationship. On the basis of the preliminary SAR data, we set up further experiments. As can be
seen from synthetic schemes (Schemes 1 and 2), unexpectedly it took long time to synthesize the
molecules. Therefore, the time when each data was collected is naturally different. The evidence
that cell reactivity changes with cell passage number is reported in the following reference: Witek
P, Korga A, Burdan F, Ostrowska M, Nosowska B, Iwan M, et al. Cytotechnology.
2016;68:2407–2415.
22
The calculated log P (clog P) was obtained by a ChemDraw Professional 17.0 program.
23
One reviewer pointed out the relevance between the activity and the clog P value. The clog P
values of 2a, 2b, and 2c are 12.3, 12.9, and 13.4, respectively, which are slightly higher than
those of their regioisomers (11.9 for 7a, 12.5 for 7b, and 13.0 for 7c, respectively). The clog P of
3-hydroxyhericenone F (1a) is 9.6 that is lower than those of 7a-c, but the activity against
TM-induced ER stress is in the same level as 7b and 7c. Given these results, it might be
reasonable to consider that the protective activity is somewhat dependent on the hydrophobicity
but not parallel to the clog P value.
11


Structure-activity relationship of geranyl resorcylates derived from Hericium erinaceum was
investigated.


Four natural products and five derivatives were synthesized.
Neuroprotective activities against endoplasmic reticulum (ER) stress-dependent cell death
were evaluated.

Synthetic derivatives exhibited higher neuroprotective activities than natural products.
12

OH
OH
O
CHO
O
O
C17H33
C17H33
MeO
MeO
CHO
O
regioisomer of hericene B (synthetic derivative)
potent
hericene B (natural product)
less potent
Neuroprotective effect against ER stress-dependent cell death