Synthesis of the reported structure of homocereulide and its vacuolation assay

Article abstract of DOI:10.1016/j.bmcl.2019.01.007

Homocereulide, isolated from marine bacterium Bacillus cereus, is an analog of emetic toxin cereulide. There is no report on its structure determination and involvement in B. cereus-associated food poisoning. Homocereulide is a cyclic dodecadepsipeptide composed of L-O-Val-L-Val-D-O-Leu-D-Ala and L-O-allo-Ile-D-Val-D-O-Leu-D-Ala. Here, we synthesized homocereulide using liquid phase fragment condensation. The NMR spectrum of synthesized homocereulide confirmed the intended structure and LC-MS results were consistent with natural products. Morphological evaluation using HEp-2 cells showed higher toxicity with homocereulide (1.39 nM) than cereulide (3.95 nM). Though cereulide is the main component in broth culture, homocereulide is also likely involved in B. cereus-associated food poisoning.

Full text of DOI:10.1016/j.bmcl.2019.01.007

Bioorganic&MedicinalChemistryLetters29(2019)734–739
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis of the reported structure of homocereulide and its vacuolation
assay
⁎
Toshihito Nakaa,c, , Yoshihide Hattorib, Hiroshi Takenakab, Yoichiro Ohtab, Mitsunori Kirihatab,
Shinji Tanimoric
a Toyo College of Food Technology, 4-23-2 Minami-hanayashiki, Kawanishi-shi, Hyogo 666-0026, Japan
b Research Organization for the 21st Century, Osaka Prefecture University, 1-1 Gakuen-cho, Nakaku, Sakai, Osaka 599-8531, Japan
c Graduate School of Life and Environmental Sciences, Osaka Prefecture University, 1-1 Gakuen-cho, Nakaku, Sakai, Osaka 599-8531, Japan
A R T I C L E I N F O
A B S T R A C T
Keywords:
Homocereulide, isolated from marine bacterium Bacillus cereus, is an analog of emetic toxin cereulide. There is
no report on its structure determination and involvement in B. cereus-associated food poisoning. Homocereulide
is a cyclic dodecadepsipeptide composed of L-O-Val-L-Val-D-O-Leu-D-Ala and L-O-allo-Ile-D-Val-D-O-Leu-D-Ala.
Here, we synthesized homocereulide using liquid phase fragment condensation. The NMR spectrum of synthe-
sized homocereulide confirmed the intended structure and LC-MS results were consistent with natural products.
Morphological evaluation using HEp-2 cells showed higher toxicity with homocereulide (1.39 nM) than cereu-
lide (3.95 nM). Though cereulide is the main component in broth culture, homocereulide is also likely involved
in B. cereus-associated food poisoning.
Homocereulide
Total synthesis
Bacillus cereus
Emetic toxin
Cereulide analogs
Bacillus cereus is a Gram-positive bacterium found in various en-
vironments and often causes food poisoning, particularly the emetic or
diarrheal type. There are many cases of the diarrheal type in North
America and Northern Europe, while the emetic type mainly occurs in
Japan and the United Kingdom.1 In such cases, vomiting is usually mild,
but rare cases of death have occurred.2 Cereulide, an emetic toxin in B.
cereus-associated food poisoning, was isolated from culture broth and
its chemical structure was determined by Isobe and co-workers.3,4
Cereulide shows heat resistance even at 126 °C and does not decompose
during food processing.5–7
poisoning, of homocereulide have not yet been reported. Herein, we
described the total synthesis of the proposed structure of homocereulide
and the measurement of vacuole activity in HEp-2 cells for its toxicity
evaluation.
Homocereulide was synthesized by the liquid phase fragment con-
densation method using commercially available amino acids. The
starting hydroxy acids were prepared through sterically retaining hy-
droxylation of the corresponding amino acids.11 Amino acids and hy-
droxy acids were coupled to afford the corresponding three tetra-
depsipeptides, which were sequentially condensed to lead to
a
Cereulide is a 36-membered cyclic depsipeptide containing 12 ste-
reogenic centers and a cyclo [L-O-Val-L-Val-D-O-Leu-D-Ala]3 depsipep-
tide sequence (Fig. 1). Cereulide shows ionophore activity with specific
affinity for K+ and inhibits respiration in mitochondria.4,8,9 Its food
toxicity is assessed by vacuolation activity in HEp-2 cells, a human
laryngeal carcinoma cell line.6,8 Homocereulide, one of the analogs of
cereulide, was isolated from marine bacterium B. cereus.10 The chemical
structure of homocereulide is proposed to be cyclo (L-O-allo-Ile-D-Val-D-
O-Leu-D-Ala [L-O-Val-L-Val-D-O-Leu-D-Ala]2) depsipeptide, as one O-Val
of cereulide is replaced by O-allo-Ile.10 Although strong cytotoxic ac-
tivity is reported, chemical and biological properties, such as food
precursor dodecadepsipeptide. Finally, homocereulide was synthesized
by a head-to-tail intramolecular coupling reaction.
Dipeptide 5 was obtained in good yield by coupling D-O-Leu 3 and D-
alanine p-toluene sulfonate
4 using N-ethyl-N′-(3-dimethylamino-
propyl)carbodiimide (EDCI) and N-hydroxybenzotriazole (HOBt) in the
presence of N,N-diisopropylethylamine (DIPEA). Dipeptide 8 was pre-
pared using the same method from L-O-allo-Ile 6 and
ester hydrochloride 7. The hydroxy group in 8 was prDo-tveacltiende absentezryt-l
butyldimethylsilyl ether to provide dipeptide 9. Subsequently, the
benzyl group in 9 was cleaved to obtain acid 10. Dipeptides 5 and 10
were coupled with p-toluoyl chloride in the presence of triethylamine
⁎ Corresponding author at: Toyo College of Food Technology, 4-23-2 Minami-hanayashiki, Kawanishi-shi, Hyogo 666-0026, Japan.
E-mail address: toshihito_naka@toshoku.ac.jp (T. Naka).
https://doi.org/10.1016/j.bmcl.2019.01.007
Received 28 October 2018; Received in revised form 21 December 2018; Accepted 8 January 2019
Availableonline10January2019
0960-894X/©2019PublishedbyElsevierLtd.

T. Naka et al.
Bioorganic&MedicinalChemistryLetters29(2019)734–739
cleavage of the tert-butyldimethylsilyl group gave alcohol 22 (Scheme
2).
Tetradepsipeptidic acid 12 and octadepsipeptidic alcohol 22 were
coupled using the same method used for the preparation of 11 to obtain
23 in 74% yield. Removal of the tert-butyldimethylsilyl group in 23
with hydrofluoric acid/pyridine gave dodecadepsipeptidic alcohol 24.
Subsequent cleavage of the benzyl group gave dodecadepsipeptide
precursor 25. Macrolactonization of dodecadepsipeptide 25 was carried
out under high-dilution conditions (1.5 mM) using m-nitrobenzoic an-
hydride (MNBAN)12 to give the reported structure of homocereulide (2)
in 74% yield, [α]D +10.8° (c 0.40, CH3OH, lit. +10.5°, c 0.12, CH3OH)
(Scheme 3). The synthesis of homocereulide was achieved in 11%
overall yield in 18 steps.
Fig. 1. Chemical structure of cereulide and homocereulide.10
The methanol solution of synthetic homocereulide was analyzed by
LC-MS using C18 reversed-phase chromatography. The retention time
of synthetic homocereulide and natural product in trypticase soy broth
(TSB) were both 16.1 min (Fig. 2). In the analytical chromatogram of
these mixed solutions, one peak formed without separation. The mass
spectrum of synthetic homocereulide was [M+H]+ m/z 1167.6961 and
that of natural homocereulide was [M+H]+ m/z 1167.6926, which
were consistent (Fig. 3). Therefore, synthetic homocereulide is con-
sidered to be the same as the natural product.
and 4-(N,N-dimethylamino)pyridine (DMAP) to afford tetradepsipep-
tide 11. In the course of this reaction, a small amount of partially
epimerized by-product was observed by TLC. The minor by-product was
easily removed by silica gel chromatography. The compound 11 was
then hydrogenated to afford acid 12 (Scheme 1).
Dipeptide 15 was prepared from L-O-Val 13 and
ester hydrochloride 14 using the same method for theLp-vraelpinareatbioennzoyfl
dipeptide 5. The hydroxy group in 15 was protected as tert-butyldi-
methylsilyl ether to afford dipeptide 16. Subsequently, the benzyl
group in 16 was cleaved to obtain acid 17. Dipeptides 5 and 17 were
coupled using the same method used for the preparation of 11 to obtain
tetradepsipeptide 18. Hydrogenation of 18 gave tetradepsipeptidic acid
19 in quantitative yield. Removal of the tert-butyldimethylsilyl group in
18 with hydrofluoric acid/pyridine gave tetradepsipeptidic alcohol 20
in 85% yield. The coupling reaction between 19 and 20 with p-toluoyl
chloride afforded octadepsipeptide 21 in 61% yield. Subsequent
Subsequently, 1H, 13C NMR, and HSQC spectra in CDCl3 of synthetic
homocereulide were obtained. These results supported the synthesized
structure. In comparison with the literature,10 our obrain spectra were
highly coincident with the reported ones. However, the multiplet signal
of O-allo-Ile (measured δ = 1.42 ppm, Lit.10 1.55 ppm) was confirmed
to be lower than multiplet signals of Ala (measured δ = 1.43–1.46 ppm,
Lit. 1.45 ppm) in 1H NMR, which was opposite to that in the literature
(Table 1). In 13C NMR, the side chain [2 × CH3] of O-allo-Ile had a
different measured value (11.8, 14.1 ppm) with respect to the literature
Scheme 1. Synthesis of tetradepsipeptidic acid 12. Reagents and conditions: (a) EDCI, HOBt, DIPEA, CH3CN, r.t.; (b) tert-butyldimethyl-chlorosilane (TBDMSCl),
imidazole, N,N-dimethylformamide (DMF), 50 °C; (c) H2, Pd/C, CH3OH, r.t.; (d) p-toluoyl chloride, Et3N, DMAP, toluene, r.t.
735

T. Naka et al.
Bioorganic&MedicinalChemistryLetters29(2019)734–739
O
Scheme 2. Synthesis of octadepsipeptidic alcohol 22. Reagents and conditions: (a) EDCI, HOBt, DIPEA, CH3CN, 0 °C to r.t.; (b) TBDMSCl, imidazole, DMF, 50 °C; (c)
H2, Pd/C, CH3OH, r.t. (d) p-toluoyl chloride, Et3N, DMAP, toluene, r.t.; (e) 70% HF/pyridine, tetrahydrofuran (THF), −10 °C to r.t.
Scheme 3. Synthesis of homocereulide (2). Reagents and conditions: (a) p-toluoyl chloride, Et3N, DMAP, toluene, r.t.; (b) 70% HF/pyridine, THF, −10 °C to r.t.; (c)
H2, Pd/C, CH3OH, r.t.; (d) MNBAN, DMAP, CH2Cl2 (c 1.5 mM), r.t.
value (15.1, 23.3 ppm). The reason for these disagreements is currently
under investigation.
observed for vacuole responses in HEp-2 cells (Fig. 4). The concentra-
tion of the highest dilution of the test sample producing vacuole for-
mation in more than 30% of HEp-2 cells at 10 vacuoles/cell was de-
termined (Table 2). Cereulide showed vacuolation activity at 3.95 nM,
whereas homocereulide showed activity at 1.39 nM, indicating that
homocereulide has stronger toxicity than cereulide. The LC-MS
Morphological changes in HEp-2 cells were examined using the
method described by Sakurai et al.8 An aliquot of 2-fold serially diluted
sample solution was mixed with a suspension of HEp-2 cells in each
well of a 96-well tissue culture plate, incubated at 37 °C for 24 h, and
736

T. Naka et al.
Bioorganic&MedicinalChemistryLetters29(2019)734–739
Intens.
x10
homocereulide_1-5_01_2001.d: EIC 1149.9992-1220.0001 +
5
2
(a)
6
4
2
0
Intens.
24H 5mL_1-1_01_1997.d: EIC 1150.0000-1220.0000 +
4
4
x10
1
(b)
3
2
1
2
14.8
15.0
15.2
15.4
15.6
15.8
16.0
16.2
16.4
16.6
Time [min]
Fig. 2. LC-MS chromatogram of homocereulide. (a) Synthetic homocereulide, (b) Culture broth of B. cereus cultivated in 5 mL TSB for 24 h at 35 °C and 180 rpm. Peak
No. 1 indicates cereulide and peak No. 2 indicates homocereulide.
Intens.
homocereulide_1-5_01_2001.d: +MS, 16.1min #963
1167.6967
(a)
(b)
1168.7019
4000
2000
0
1169.7040
1170.7075
1162
1164
1166
1168
1170
m/z
Intens.
800
600
400
200
0
24H 5mL_1-1_01_1997.d: +MS, 16.1min #960
1167.6925
1168.6898
1170.6971
1162
1164
1166
1168
1170
m/z
Fig. 3. Mass spectra of homocereulide. (a) Synthetic homocereulide. (b) Culture broth of B. cereus cultivated in 5 mL TSB for 24 h at 35 °C and 180 rpm.
quantitative results of the culture broth cultivated for 24 h (Fig. 2, TSB
medium) showed that the concentration of cereulide was 65.4 nM and
that of homocereulide was 9.3 nM. Obviously, homocereulide is a minor
component in culture but has strong toxicity. Thus, it is considered to
be involved in B. cereus-associated food poisoning. The three synthetic
intermediates (23–25) exhibited vacuole change at 1000 nM or more.
These results suggest that the cyclic structure is important for food
poisoning induced by depsipeptide compounds.
In conclusion, homocereulide was prepared efficiently and con-
veniently using liquid phase condensation from readily available che-
micals. The structure of synthesized homocereulide was confirmed by
NMR and LC-MS. In this study, we demonstrated the emetic toxicity of
homocereulide by vacuolating assay using HEp-2 cells. Homocereulide
showed higher toxicity than cereulide. These data suggest the need for
examination of minor components in emetic type food poisoning caused
by B. cereus.
737

T. Naka et al.
Bioorganic&MedicinalChemistryLetters29(2019)734–739
Table 1
1H and 13C NMR spectroscopic data for homocereulide (2).
Synthetic homocereulide (CDCl3, 500 MHz)
Lit. (CDCl3, 400 MHz)10
aδH
δC
δH
δC
NH
7.80 (1H, d, aJ = 7.0)
7.71–7.74 (4H, m)
7.76 (6H, m)
171.5 (s)
48.9 (d)
15.7 (q)
7.68 (1H, d, J = 6.5)
Ala
4.28–4.40 (3H, m)
1.43–1.46 (9H, m)
171.2, 171.3 (2C)
48.7, 48.8, 48.9
15.7, 15.8 (2C)
4.35 (3H, m)
1.451 (3H, d, J = 7.0)
1.442 (3H, d, J = 7.0)
1.438 (3H, d, J = 7.0)
O-Leu
5.26 (2H, dd, J = 4.0, 8.5)
5.28 (1H, dd, J = 4.0, 11.0)
1.63–1.83 (6H, m)
171.7 (2C), 171.8
72.8 (2C), 72.9
40.5 (3C)
5.30 (3H, dd, J = 8.1, 4.7)
171.9
1.76 (6H, m)
72.8 (d)
40.6 (t)
24.4 (d)
23.3 (q)
21.3 (q)
1.68 (3H, m)
1.63–1.83 (3H, m)
24.4 (3C)
0.92 (9H, d, J = 6.2)
0.89 (9H, d, J = 6.2)
0.89–0.93 (18H, m)
23.3 (3C)
21.2 (3C)
Val
4.09–4.19 (3H, m)
2.25–2.35 (3H, m)
1.05 (9H, d, J = 6.0)
0.96–0.97 (9H, m)
170.3, 170.4 (2C)
4.10 (3H, m)
170.4 (s)
59.3 (d)
28.7 (d)
19.3 (q)
19.3 (q)
59.3 (3C)
2.31 (3H, m)
28.7 (3C)
1.05 (9H, d, J = 6.6)
0.95 (9H, d, J = 6.6)
19.3 (3C)
19.2 (3C)
O-Val
O-allo-Ile
4.99 (2H, d, J = 3.5)
2.25–2.35 (2H, m)
0.97–0.99 (12H, m)
170.9 (2C)
78.8 (2C)
30.6 (2C)
18.5 (2C)
16.9 (2C)
171.6
77.4
37.1
25.8
14.1
11.8
4.99 (1H, d, J = 3.3)
4.98 (1H, d, J = 3.3)
2.31 (2H, m)
171.0 (s)
78.8 (d)
30.5 (d)
18.6 (q)
16.9 (q)
170.9 (s)
78.3 (d)
37.2 (d)
24.5 (t)
23.3 (q)
15.1 (q)
0.97 (12H, d, J = 7.3)
5.10 (1H, d, J = 3.0)
2.05–2.10 (1H, m)
1.42–1.46 (1H, m)
1.21–1.29 (1H, m)
0.93–0.95 (6H, m)
5.05 (1H, d, J = 4.6)
2.01 (1H, m)
1.55 (1H, ddq)
1.30 (1H, ddq)
0.94 (3H, d, J = 7.0)
0.91 (3H, d, J = 7.0)
a
Chemical shifts (δ) are expressed in ppm, and J values are presented in Hz.
A
B
Fig. 4. Micrographs of HEp-2 cells. (A) Control cells with no treatment. (B) Vacuolar cells incubated with 2.0 nM homocereulide.
Acknowledgments
Table 2
Concentration of indicated compounds in vacuole formation and B. cereus
culture broth.
This work was funded by The Promotion and Mutual Aid
Corporation for Private Schools of Japan the Science (Research
Promotion Fund 2016). We thank Dr. Fujihiko Matsunaga and Dr.
Sakiko Inatsu (Toyo College of Food Technology) for technical assis-
tance in bacterial culture, and Mr. Toru Takahashi and Mr. Ryoichi
Izuchi (Toyo Institute of Food Technology) for technical assistance in
recording mass spectra and cell culture.
Compound No.
Vacuole formation
(nM)a
Concentration in B. cereus culture
broth (nM)b
Cereulide (1)c
3.95
1.39
1845
1592
1124
1.22
0.43
1,081
723
65.4
9.3
–
15.0
1.2
Homocereulide (2)
25
24
23
–
348
–
a
b
Data are presented as mean concentration
standard deviation (N = 6).
Appendix A. Supplementary data
Quantitative results of LC-MS analysis of B. cereus culture cultivated in
5 mL TSB for 24 h at 35 °C and 180 rpm. Data are presented as mean con-
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bmcl.2019.01.007. These data include MOL files and
InChiKeys of the most important compounds described in this article.
centration
standard deviation (N = 3).
c
A previously reported synthesized cereulide was used.13
738

T. Naka et al.
Bioorganic&MedicinalChemistryLetters29(2019)734–739
References
7. Rajkovic A, Uyttendaele M, Debevere J, et al. Lett Appl Microbiol. 2008;46:536–541.
8. Sakurai N, Koike KA, Hayashi H, et al. Microbiol Immunol. 1994;38:337–343.
9. Agata N, Ohta M, Isobe M, et al. FEMS Microbiol Lett. 1995;129:17–20.
10. Wang GYS, Kuramoto M, Uemura D, et al. Chem Lett. 1995:791–792.
11. Kallmerten J, Gould JT. J Org Chem. 1986;51:1152–1155.
1. Arnesen SLP, Fagerlund A, Granum PE. FEMS Microbiol Rev. 2008;32:579–606.
2. Naranjo M, Denayer S, Dierick K, et al. J Clin Microbiol. 2011;49:4379–4381.
3. Agata N, Mori M, Isobe M, et al. FEMS Microbiol Lett. 1994;121:31–34.
4. Suwan S, Isobe M, Ohta M, et al. J Chem Soc Perkin Trans. 1995;1:765–775.
5. Mikami T, Horikawa T, Masuko S, et al. FEMS Microbiol Lett. 1994;119:53–58.
6. Shinagawa K, Konuma H, Sugii S, et al. FEMS Microbiol Lett. 1995;130:87–90.
12. Makarasen A, Nishikawa T, Isobe M. Synthesis. 2009;13:2184–2204.
13. Murakami H, Naka T, Kirihata M, et al. Pept Sci. 2012:225–228.
739