A new antiangiogenic C24 oxylipin from the soft coral Sinularia numerosa

Article abstract of DOI:10.1016/j.bmc.2008.10.083

A new oxylipin, 15-hydroxy-tetracosa-6,9,12,16,18-pentaenoic acid (15-HTPE; 1) was isolated as an inhibitor of tube-formation from the soft coral Sinularia numerosa. Its structure was elucidated by means of spectral analysis and chemical degradation. 15-H

Full text of DOI:10.1016/j.bmc.2008.10.083

Bioorganic & Medicinal Chemistry 17 (2009) 2181–2184
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
A new antiangiogenic C24 oxylipin from the soft coral Sinularia numerosa
a
c
Takahiro Yamashita a,z, Yoichi Nakao a,b, , Shigeki Matsunaga , Tsutomu Oikawa ,
*
Yukimitsu Imahara d, Nobuhiro Fusetani a,e,
*
a Graduate School of Agricultural and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan
b School of Advanced Science and Engineering, Waseda University, 3-4-1 Okubo, Shinjuku-ku, Tokyo 169-8555, Japan
c Kanagawa University of Human Services, 1-10-1 Heisei-cho, Yokosuka, Kanagawa 238-8522, Japan
d Wakayama Prefectural Museum of Natural History, 370-1, Funao, Kainan, Wakayama 642-0001, Japan
e Faculty of Fisheries Sciences, Hokkaido University, 3-1-1 Minato-cho, Hakodate 041-8611, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
A new oxylipin, 15-hydroxy-tetracosa-6,9,12,16,18-pentaenoic acid (15-HTPE; 1) was isolated as an
inhibitor of tube-formation from the soft coral Sinularia numerosa. Its structure was elucidated by means
of spectral analysis and chemical degradation. 15-HTPE inhibited tube formation of EA.hy926 cells at the
Received 7 March 2008
Revised 20 September 2008
Accepted 31 October 2008
Available online 5 November 2008
concentration of 20–40 lM.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Angiogenesis inhibitor
Tube formation
Marine natural products
Oxylipin
1. Introduction
taenoic acid (15-HTPE; 1), along with the known 11-hydroxy-eico-
sa-5,8,12,14-tetraenoic acid (11-HETE; 2), and 9-hydroxy-
octadeca-6,10,12-trienoic acid (9-HOTE; 3). This paper deals with
the isolation, structural elucidation, and biological activity of 1.
Angiogenesis is a promising target for cancer chemotherapy1,2
;
the first antiangiogenic agent AvastinÒ was approved in 2004 by
FDA for treatment of colorectal and non-small cell lung cancers3
and numbers of antiangiogenic agents are under clinical trials.4
To discover antiangiogenic compounds, several assay systems have
been developed. Among them, the tube-formation system using
endothelial cells and collagen gel5 is one of the most popular
in vitro models, representing many features of in vivo angiogene-
sis; activation of endothelial cells by the growth factors, degrada-
tion of basement matrices, cell migration, and formation of tube-
like structure of the primary blood vessels.
OH
OR
1: R = H 4: R = Me
O
OH
OH
O
2: R = H 5: R = Me
3: R = H 6: R = Me
OR
OR
In the course of our search for new antiangiogenic agents from
marine invertebrates, we tested anti-tube-formation activity of
150 Japanese marine invertebrates by an assay system of the mod-
ified in vitro tube-formation model6 using EA.hy 926 cells, a per-
manent human endothelial cell line.7,8 Out of 10 active samples,
the soft coral Sinularia numerosa collected in southern Japan
showed remarkable inhibitory activity. Bioassay-guided isolation
provided a new oxylipin, 15-hydroxy-tetracosa-6,9,12,16,18-pen-
O
2. Results and discussion
The combined MeOH and CHCl3 extracts of the frozen sample
(1.0 kg, wet weight) were partitioned between H2O and CHCl3;
the latter layer was further separated by the modified Kupchan
procedure.9 The CHCl3 layer inhibiting tube formation was fur-
ther separated by solvent partitioning with n-hexane/CH2Cl2/
MeCN (10:3:7). The active lower layer was subjected to centrif-
ugal partition chromatography (CPC) with the solvent system of
n-heptane/EtOAc/MeCN/MeOH/H2O (6:6:1:4:3; mobile phase:
upper layer, stationary phase: lower layer) to afford three active
* Corresponding authors. Tel./fax: +81 3 5286 2568 (Y.N.), tel./fax: +81 138 40
8884 (N.F.).
E-mail addresses: ayocha@waseda.jp (Y. Nakao), anobu@fish.hokudai.ac.jp (N.
Fusetani).
z
Deceased author.
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.10.083

2182
T. Yamashita et al. / Bioorg. Med. Chem. 17 (2009) 2181–2184
fractions. The half portions of these fractions were separately
purified by ODS HPLC with MeCN/H2O (75:25) to yield a new
oxylipin 1 (0.7 mg), along with the known 2 (0.5 mg) and 3
(0.4 mg). The remaining portions were treated with CH2N2, fol-
lowed by ODS HPLC purification afforded the corresponding
responding to 2S and 2R derivatives, thus indicating the racemic
nature of 1 (see scheme 1).
The new oxylipin 15-HTPE (1) exhibited the anti-tube-forma-
tion activity in a dose-dependent manner (20–40 lM) in the
in vitro tube-formation model using EA.hy 926 cells. Although de-
tailed evaluation was not possible due to the limited amounts of
samples, similar activity was observed for both 11-HETE (2) and
9-HOTE (3).12 It is interesting that 12(R)-hydroxy-5,8,14-eicosatri-
enoic acid (12-HETE) is known to induce in vitro tube formation of
rabbit limbal microvascular endothelial (RLME) cells via expression
of vascular endothelial growth factor (VEGF).13 The opposite effects
of these closely related compounds may indicate their different
targets or opposite (agonistic/antagonistic) effects on the same
targets.
methyl esters,
respectively.
4
(2.2 mg),
5
(2.8 mg), and
6
(2.0 mg),
The new oxylipin 1 had a molecular formula of C24H38O3 as
established by HRFABMS and NMR data, indicating six degrees of
unsaturation. The 1H NMR spectrum showed three olefinic protons
resonating at d 5.64 (H-16, dd, J = 6.6, 15.2 Hz), d 6.50 (H-17, dd,
J = 10.9, 15.2 Hz), and d 5.96 (H-18, t, J = 10.9 Hz), the last of which
showed a COSY cross-peak to one of seven overlapping olefinic
protons resonating at d 5.40 (H-19), thus establishing connectivi-
ties from C-16 to C-19. The geometry of D16,17 and D18,19 olefins
was deduced to be E and Z, respectively, on the basis of the cou-
pling constants (15.2 and 10.9 Hz, respectively).
Interpretation of the COSY and HOHAHA spectra led to the con-
nectivities from C-19 to C-21 (Fig. 1). The linkage between C-19
and C-20 was also supported by HMBC cross-peaks, H-18/C-20,
H-20/C-18, and H-20/C-19. The terminal propyl unit was con-
nected to C-21 on the basis of HMBC network shown in Figure 1.
A COSY cross-peak between H-16 and the oxymethine proton at
d 4.13 (H-15, dt, J = 6.6, 6.9 Hz), which was connected to non-
equivalent allyl methylene protons at d 2.31 and 2.34 (H2-14), con-
structed the partial structure a (Fig. 1).
The remaining part of the molecule contained the terminal car-
boxyl group which was connected to four consecutive methylenes
based on COSY and HMBC correlations (H-2/C-1 and H-3/C-1). The
remaining three disubstituted olefins and two methylenes flanked
by these olefins, which is a common feature of polyunsaturated
fatty acids, were placed between C-14 and C-5, thus establishing
the partial structure b (Fig. 1). Although the coupling constants
could not be obtained due to the overlapping olefinic proton sig-
nals, chemical shift values for the bis-allyl methylenes (dC 26.7
and 26.9, dH 2.83 and 2.82, respectively) indicated all E-geome-
try.10,11 HMBC cross-peaks, H-15/C-13, H2-14/C-12, and H2-14/C-
13, connected partial structures a and b through C-12 and C-13
to complete the planar structure of 1, that is 15-hydroxy-tetraco-
sa-6,9,12,16,18-pentaenoic acid.
Determination of the absolute stereochemistry at C-15 was not
straightforward due to the difficulty in preparing MTPA esters. Pre-
sumably, the formed MTPA ester was easily eliminated to form a
conjugated tetraene unit during work-up. Instead, ozonolysis of
the methyl ester of 1 followed by the reductive treatment yielded
triol 7 which was converted to the tris-p-bromobenzoyl ester 8.
Chiral HPLC analysis of 8 afforded two peaks in a ratio of 1:1, cor-
PPARc ligands such as 15-deoxy-D
12,14-prostaglandin J2 (15d-
PGJ2), BRL49653, or ciglitazone were reported to suppress dose-
dependently the proliferation and the differentiation into tube-like
structures in three dimensional collagen gels of human umbilical
vein endothelial cells (HUVEC). It was also revealed that 15d-
PDJ2 decreased mRNA levels of both VEGFR-1 (Flt-1) and -2 (Flk-
1/KDR) in HUVEC. However, these effects were not observed for
the PPAR
a
ligand, Wy14643.14 In contrast to this result, Meissner
ligand, suppressed VEG-
et al. reported that Wy14643, a PPAR
a
FR-2 protein expression and inhibited basal and VEGF-induced for-
mation of capillary-like structures of HUVEC on Matrigel, whereas
ciglitazone, a PPAR
These different findings may be attributable to the different length
of exposure time to the respective PPAR ligands; PPAR ligands
c
ligand, exhibited no discernible suppression.15
c
were presumed to require longer time (48 h) to take effects.
In our in vitro tube-formation system, EA.hy 926 cells were ex-
posed to drugs for longer time (preincubation 12 h + incubation
24 h) than in the system used by Meissner et al. (total 12 h)15
allowing drugs to take effects, while cells were exposed to gels
or growth factors for shorter time (24 h) than in the system of
Xin et al. (48 h)14 resulting in the reduced stimulation by growth
factors or gels. Therefore, our system represents the intermediate
assay condition of those by Xin et al. and Meissner et al. Evaluation
of inhibitory effects on the tube formation by PPARs ligands such
as ciglitazone (PPAR
(LTB4; PPAR ) indicated that all these ligands inhibited the tube
formation in our system. This result shows that our system is capa-
ble to detect the anti-tube-formation activity of both PPAR and
c), Wy14643 (PPARa), and leukotriene B4
a
a
c
ligands, and therefore suitable for the broader search of PPARs li-
gands with antiangiogenic activity.
EA.hy 926 cells used in our system are known to express PPAR
a,
d, and . The results mentioned above suggest that PPARs are the
c
possible targets of the obtained oxylipins 1–3. It is also known that
OH
20
24
OH
11
15
1
8
5
a
O
b
COSY
HMBC
Figure 1. Key COSY and HMBC correlations and partial structures a and b.
BzBr-
p
O
OH
O3, CH2Cl2, 0 o
LiAlH4, THF, 0 o
C
C
Chiral HPLC
p
-BrBzCl, DMAP
O
BzBr-p
racemic
HO
1
p-BrBz
O
CH2Cl2, 80 oC, 24h
OH
7
8
Scheme 1. Analysis of the stereochemistry at C-15 of 1.

T. Yamashita et al. / Bioorg. Med. Chem. 17 (2009) 2181–2184
2183
Figure 2. Anti-tube-formation activities of HTPE (1) and known PPARs ligands. (a) Control, (b) ciclitazone (7) 10
lM, (c) WY14643 (8) 10
lM, (d) LTB4 (9) 5
lM, (e) LTB4 (9)
10 M, (f) LTB4 (9) 20 M, (g) HTPE (1) 10 M, (h) HTPE (1) 20 M, and (i) HTPE (1) 40
l
l
l
l
lM. Magnification: (a–f) 1ꢁ, (g–i) 1.6ꢁ.
IL-4 and -13 markedly increased mRNA levels of VCAM-1 in vascu-
lar endothelial cells, and administration of anti-VCAM-1 Ab
blocked in vitro tubular morphogenesis of human microvascular
known PPARs ligands, such as leukotriene B4, ciglitazone, and
Wy14643 in this in vitro tube-formation system, PPARs or
VCAM-1 expression were suggested as the possible targets of 1–
3. However, the targets and the modes of action of these com-
pounds are yet to be elucidated.
endothelial cells induced by IL-4 and -13.16 PPAR
a and c ligands
were reported to inhibit cytokine-induced VCAM-1 expression.17
Therefore, to confirm the target of 1–3, expression levels of VEGF,
VEGFRs, or VCAM-1, and their direct effects on each type of PPARs,
need to be evaluated (see Fig. 2).
4. Experimental
4.1. General procedures
O
H
N
O
Ultraviolet spectra were recorded on SHIMADZU BioSpec-1600
spectrometer. FAB mass spectra were measured with a JEOL SX-
102 mass spectrometer. 1H and 13C NMR spectra were recorded
on JEOL JNM-A600 NMR spectrometer (recorded in CD3OD).
N
S
O
OH
S
NH
N
Cl
O
ciglitazone (7)
WY14643 (8)
4.2. Biological materials
The soft coral sample was collected on 20 July 1998 at Bouno-
tsu, southwest coast of Kagoshima Prefecture, at the depth of
15 m. Latitude: 31°18031.400N; longitude: 130°1303400E. The sample
was frozen as soon as collected, and kept frozen at ꢀ20 °C until
extraction.
OH
OH
O
OH
LTB4 (9)
4.3. Tube-formation assay
Tube-formation assay was carried out basically following the
procedure by Oikawa et al.6 in which EA.hy 926 cells were used in-
stead of Human Dermal Microvascular Endothelial Cells (HDMECs).
EA.hy 926 cells were preincubated at 37 °C under 5% CO2 in Dul-
becco’s modified Eagle’s medium (DMEM) with the indicated con-
centrations of samples. Matrigel (Basement Membrane Matrix, BD
Biosciences) coated plates were prepared by adding Matrigel onto
3. Conclusion
A new oxylipin, 15-hydroxy-tetracosa-6,9,12,16,18-pentaenoic
acid (15-HTPE; 1) having anti-tube-formation activity was isolated
from the soft coral S. numerosa along with known 11-HETE (2) and
9-HOTE (3). 15-HTPE (1) exhibited the anti-tube-formation activity
in a dose-dependent manner within the concentration range of 10–
24-multiwell dish in the amount of 200 lL/well at 4 °C and allowed
40
lM. From the results of anti-tube-formation assay with the

2184
T. Yamashita et al. / Bioorg. Med. Chem. 17 (2009) 2181–2184
to polymerize at 37 °C for 30 min. The trypsinized cells were
seeded (under two cell concentrations of 1.2 and 1.4 ꢁ 105 cells/
mL) onto the Matrigel coated plates in 1 mL of growth medium
(DMEM with 1% FBS) containing the indicated concentration of
samples. After 24 h incubation, formation of tube-like structure
was observed under a phase-contrast microscope. EA.hy 926 cells
with the same sample concentration were incubated on the dishes
without Matrigel coating for the evaluation of cytotoxicity of the
samples at the same concentration.
the reduction by adding water, the reaction mixture was subjected
to the SiO2 column chromatography with CHCl3/MeOH [(10:0),
(19:1), (9:1), and (8:2)]. The fractions eluted with CHCl3/MeOH
(9:1) and (8:2) were shown to contain butanetriol (7) on the basis
of TLC analysis. These fractions were dissolved in CH2Cl2 and com-
bined, and then p-bromobezoylchloride (2.8 mg) and dimethylami-
nopyridine (5.6 mg) were introduced to this solution. Then, the
reaction mixture was stirred overnight in the sealed tube at
80 °C. The reaction mixture was separated by SiO2 column chroma-
tography yielding 1,2,4-trisbromobenzoate in the n-hexane/EtOAc
(9:1) eluting fraction, which was purified by ODS HPLC [COSMOSIL
5C18-ARII (Nacalai), u 10 ꢁ 250 mm, a linear gradient from MeOH/
H2O (9:1) to (10:0), flow rate: 2.0 mL/min]. The Chiral HPLC [Chi-
ralcel OJ, (Daicel Chemical Industries, Ltd), u 4.6 ꢁ 250 mm, n-hex-
ane/i-PrOH (8:2), flow rate: 1.0 mL/min] profile of the obtained
1,2,4-tris-p-bromobenzoate (8) gave two peaks of 1:1 ratio at the
retention time of 12 and 14 min corresponding to those of standard
samples prepared from (S)- and (R)-butanetriols, respectively.
4.4. Isolation
The soft coral S. numerosa (900 g, wet weight) was extracted
with MeOH and CHCl3. The combined extracts was evaporated
and partitioned between CHCl3 and H2O. The CHCl3 layer was fur-
ther separated with hexane and MeOH/H2O (9:1). The aqueous
MeOH fraction was diluted with H2O to make MeOH/H2O (6:4)
which was extracted with CHCl3. The active CHCl3 layer was sub-
jected to solvent partitioning system using n-hexane/CH2Cl2/MeCN
(10:3:7). The active lower layer was further separated by centrifu-
gal partition chromatography (CPC) using n-heptane/EtOAc/MeCN/
MeOH/H2O (6:6:1:4:3) as a solvent system to obtain nine fractions
(fr. 1–9), three of which (fr. 3–5) showed anti-tube-formation
activities. Half of the active fr. 3 obtained was separated on ODS
column [Inertsil (GL Science Inc.), u 10 ꢁ 250 mm, MeCN/H2O
(75:25), flow rate: 2.0 mL/min] to yield pure compound (1;
0.7 mg). In the same way, each half of fr. 4 and fr. 5 was purified
to yield two active compounds (2; 0.5 mg and 3; 0.4 mg, respec-
tively). The remaining halves of the fractions were treated with
CH2N2 in ether at 0 °C prior to the final purification by ODS HPLC
with MeCN/H2O (75:25). The methyl esterification improved the
yields and gave the active substances as methyl esters (4; 2.2 mg,
5; 2.8 mg, and 6; 2.0 mg, respectively).
Acknowledgments
We are indebted to the crew of R/V Toyoshio-maru of Hiro-
shima University for assistance in collection of the samples. This
work was partly supported by Waseda University Grant for Special
Research Projects, Nissui Research foundation, and a Grant-in-Aid
for Scientific Research from the Ministry of Education, Culture,
Sports, Science, and Technology of Japan (#19310138).
References and notes
1. Folkman, J.; Klagsbrun, M. Science 1987, 234, 444.
2. Weidner, N.; Semple, J. P.; Weigh, W. R. N. Engl. J. Med. 1991, 324, 1.
3. Keeding, V. L.; Sandler, A. B. Cancer Sci. 2007, 98, 1825.
4. Hofmeister, V.; Schrama, D.; Becker, J. C. Cancer Immunol. Immunother. 2008, 57,
1.
5. Folkman, J.; Langer, R.; Linhardt, R. J.; Haudenschild, C.; Taylor, S. Science 1983,
221, 719.
4.5. 15-Hydroxy-tetracosa-6,9,12,16,18-pentaenoic acid (15-
HTPE; 1)
6. Oikawa, T.; Sasaki, T.; Nakamura, M.; Shimamura, M.; Tanahashi, N.; Omura, S.;
Tanaka, K. Biochem. Biophys. Res. Commun. 1998, 246, 243.
7. EA.hy 926 cell line was established by fusing HUVECs (human umbilical vein
endothelial cells) with the permanent cell line A 549 (derived from a human
lung carcinoma). It was kindly provided by Dr. Cora Jean S. Edgell (University of
North Carolina, Department of Pathology, 7525 Chapel Hill, NC 27599, USA).
8. Edgell, C. J. S.; McDonald, C. C.; Graham, J. B. Proc. Natl. Acad. Sci. U.S.A. 1983, 80,
3734.
9. Kupchan, S. M.; Britton, R. W.; Ziegler, M. F.; Sigel, C. W. J. Org. Chem. 1973, 38,
178.
10. Breitmaier, E.; Voelter, W. Carbon-13 NMR spectroscopy, 3rd ed.; VCH: New
York, 1989. pp 192–194.
Colorless solid. aD27 0° (c, 0.1, MeOH); UV (MeOH) kmax
236.0 nm; FABMS: [M+Na]+ m/z 397, [MꢀOH]+ m/z 357, [MꢀH]ꢀ
m/z 373, HRFABMS: m/z 373.2736: [MꢀH]ꢀ (calcd for C24H37O3,
373.2743); 1H NMR (600 MHz, CD3OD); d 6.50 (dd, J = 15.2,
10.9 Hz, H-17), 5.96 (t, J = 10.9 Hz, H-18), 5.64 (dd, J = 15.2,
6.6 Hz, H-16), 5.43 (m, H-12, 13), 5.40 (dt, J = 10.9, 7.3 Hz, H-19),
5.37 (m, H-6), 5.36 (m, H-7), 5.34 (m, H-9, 10), 4.13 (dt, J = 6.6,
6.9 Hz, H-15), 2.86 (2H, m, H-11), 2.82 (2H, m, H-8), 2.34 (m, H-
14a), 2.31 (m, H-14b), 2.27 (2H, t, J = 7.5 Hz, H-2), 2.18 (2H, dq,
J = 7.6, 7.3 Hz, H-20), 2.10 (2H, q, J = 7.3 Hz, H-5), 1.62 (2H, quint.,
J = 7.5 Hz, H-3), 1.40 (2H, quint., J = 7.5 Hz, H-4), 1.39 (2H, quint.,
J = 7.6 Hz, H-21), 1.32 (4H, m, H-22, 23), 0.90 (t, J = 6.9 Hz, H-22);
13C NMR (150 MHz, CD3OD); 178.2 (C-1), 136.6 (C-16), 133.2 (C-
19), 130.8 (C-12), 130.6 (C-6), 129.3 (C-18), 129.3 (C-9 or 10),
129.1 (C-7), 129.0 (C-9 or 10), 126.6 (C-17), 126.5 (C-13), 73.2
(C-15), 36.5 (C-14), 35.3 (C-2), 32.6 (C-22), 30.5 (C-21), 30.3 (C-
4), 28.7 (C-20), 28.0 (C-5), 26.9 (C-11), 26.7 (C-8), 26.0 (C-3), 23.6
(C-23), 14.4 (C-24).
11. Ortega, M. J.; Zubia, E.; Sanchez, M. C.; Salva, J.; Carballo, J. L. Tetrahedron 2004,
60, 2517.
12. Tube-formation assay was performed at six concentrations (1–40
lg/mL) of
11-HETE (2) and inhibition was observed at 20 and 40 g/mL. As for 9-
l
hydroxy-octadeca-6,10,12-trienoic acid (9-HOTE; 3), detailed evaluation of
inhibitory activity was not possible because of the scarce amount of the
purified compound. Inhibitory activity at 20 lg/mL was observed for the
fraction mainly containing compound 3. Commercially available related 9-
hydroxy-octadeca-6,10,12-dienoic acid (9-HODE), as well as docosahexaenoic
acid (DHA), eicosapentaenoic acid (EPA), and arachidonic acid was evaluated
for inhibitory activity against tube formation. Inhibition was observed for 9-
HODE at 25
arachidonic acid at 50
l
M, while no remarkable inhibition was observed for DHA, EPA, or
M.
l
13. Mezentsev, A.; Seta, F.; Dunn, M. W.; Ono, N.; Falck, J. R.; Laniado-
Schwartzman, M. J. Biol. Chem. 2002, 277, 18670.
14. Xin, X.; Yang, S.; Kowalski, J.; Gerritsen, M. E. J. Biol. Chem. 1999, 274, 9116.
15. Meissner, M.; Stein, M.; Urbich, C.; Reisinger, K.; Suske, G.; Staels, B.;
Kaufmann, R.; Gille, J. Circ. Res. 2004, 94, 324.
4.6. Stereochemistry of the hydroxy group
16. Fukushi, J.; Ono, M.; Morikawa, W.; Iwamoto, Y.; Kuwano, M. J. Immunol. 2000,
165, 2818.
17. Rival, Y.; Benéteau, N.; Taillander, T.; Pezet, M.; Dupont-Passelaigue, E.;
Patoiseau, J.-F.; Junquéro, D.; Colpaert, F. C.; Delhon, A. Eur. J. Pharmacol. 2002,
435, 143.
A portion of 1 (0.2 mg), dissolved in CH2Cl2, was exposed to
ozone for 15 min at 0 °C. After remaining ozone was removed by
N2 bubbling, solvent was evaporated. The residue was dissolved
in THF, and to this solution LiAlH4 was added. After quenching