Synthesis and stereochemical confirmation of the HI/JK ring system of prymnesins, potent hemolytic and ichthyotoxic glycoside toxins isolated from the red tide alga

Article abstract of DOI:10.1016/S0040-4039(01)01067-X

Stereocontrolled synthesis of the HI/JK ring model of the prymnesins, glyosidic toxins isolated from the red tide phytoflagellate Prymnesium parvum, is described. Comparison of its ¹H and ¹³C NMR data with those of the natural toxins established the earlier stereochemical assignments.

Full text of DOI:10.1016/S0040-4039(01)01067-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 5725–5728
Synthesis and stereochemical confirmation of the HI/JK ring
system of prymnesins, potent hemolytic and ichthyotoxic
glycoside toxins isolated from the red tide alga
Makoto Sasaki,a,b,* Takeshi Shidaa and Kazuo Tachibanaa
aDepartment of Chemistry, Graduate School of Science, The University of Tokyo, and CREST,
Japan Science and Technology Cooperation (JST), Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
bGraduate School of Life Science, Tohoku University, Aoba-ku, Sendai 981-8555, Japan
Received 28 May 2001; revised 14 June 2001; accepted 15 June 2001
Abstract—Stereocontrolled synthesis of the HI/JK ring model of the prymnesins, glyosidic toxins isolated from the red tide
phytoflagellate Prymnesium parvum, is described. Comparison of its 1H and 13C NMR data with those of the natural toxins
established the earlier stereochemical assignments. © 2001 Elsevier Science Ltd. All rights reserved.
Two glycosidic toxins, prymnesin-1 (PRM1, 1) and
prymnesin-2 (PRM2, 2), were isolated from cultured
cells of the red tide phytoflagellate Prymnesium
parvum.1 These toxins possess extremely potent
hemolytic activity, which is about 5,000-fold greater
than that of Merk saponin on a molar basis and also
exhibit potent ichthyotoxicity.2 Their gross structures
and partial stereochemical assignments have been dis-
closed by Igarashi and Yasumoto.3,4 Prymnesins pos-
sess unique structural features: an unbranched single
chain of 90 carbons except for a single methyl group, a
fused polyether ring system (A–E ring), four distinct
1,6-dioxadecalin units, conjugated double and triple
bonds, chlorine and nitrogen atoms, and an uncommon
assigned stereochemistry. We have already demon-
strated the utility of a synthetic approach for the
configurational assignment of acyclic portions of maito-
toxin, the most toxic and largest non-biopolymer.6 As
part of our studies toward complete stereochemical
assignment of prymnesins, we describe herein the
stereoselective synthesis of the HI/JK ring model 3,
which culminated in the confirmation of the proposed
stereochemical assignment.
Synthesis of the H ring alkyne 9 started with tri-O-ace-
tyl- -glucal (4), which was converted to alcohol 5 in
D
three steps (Scheme 1). An oxidation–reduction
sequence allowed inversion of the hydroxyl group to
give 6 in 78% yield. Benzylation followed by removal of
the acetonide group provided diol 7, which upon
regioselective activation and protection by a one-pot
procedure7 gave triflate 8. Alkylation of 8 with lithium
(trimethylsilyl)acetylide was followed by desilylation to
give the desired alkyne 9 in 64% overall yield.
L
-xylose. The relative stereochemistry of the fused A–E
polyether ring domain and four 1,6–dioxadecalin units
(rings FG, HI, JK, and LM) was determined by exten-
sive NMR analysis. The diastereomeric relationship
among these polyethers was assigned by the extensive
3
NOE and JH,H analysis, indicating that all ring link-
ages, except for that of rings I/J, took the CꢀC anti
conformation with respect to the CꢀC linking bond.
For rings I/J, an exceptional CꢀC gauche rotamer was
proposed on the basis of NOE analysis; however, force
field calculations suggested a CꢀC anti system as the
lowest energy rotamer.5 Accordingly, we decided to
synthesize the HI/JK ring model to confirm the
Construction of the JK ring lactone 15 is shown in
Scheme 2. Routine protective group manipulation
allowed conversion of the known alcohol 108 into alco-
hol 11. Inversion of the secondary alcohol followed by
deprotection of the primary alcohol and benzylation
gave bis-benzyl ether 12 in 98% overall yield. Oxidative
cleavage of the double bond provided the correspond-
ing aldehyde, which upon chelation-controlled allyla-
* Corresponding author. Tel.: +81-3-5841-4357; fax: +81-3-5841-8380;
e-mail: msasaki@chem.s.u-tokyo.ac.jp
9
tion with allyltributyltin and MgBr2·OEt2 yielded
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01067-X

5726
M. Sasaki et al. / Tetrahedron Letters 42 (2001) 5725–5728
homoallylic alcohol 13 in 90% yield along with the
undesired diastereomer 14 (7% yield). Protection as its
p-methoxybenzyl (MPM) ether, desilylation, oxidative
cleavage of the terminal olefin and oxidation of the
derived hemiacetal with TPAP/NMO10 furnished the
desired lactone 15 in 84% overall yield.
(Scheme 3).12 Lindlar reduction followed by oxidative
removal of the MPM group with DDQ produced
alcohol 18. Stereoselective installation of the axial
chlorine atom at C56 was achieved by treatment with
CCl4 and Ph3P.13 Dihydroxylation of the cis-double
bond of 19 with OsO4 and NMO gave a mixture of
diols, which was oxidized under Swern conditions to
give diketone 20. Removal of the TBS group with
HF·pyr led to the exclusive formation of six-mem-
bered hemiketal 21 in 53% yield for the two steps. In
this transformation, the corresponding five-membered
hemiketal was not produced.14 DIBALH reduction
proceeded stereoselectively to yield alcohol 22, which
was converted to the corresponding methyl acetal and
then subjected to Et3SiH/BF3·OEt2 to yield tetracyclic
ether 23. Finally, removal of the benzyl groups fur-
nished the targeted HI/JK ring model 3 in 65% over-
all yield for the four steps. The stereochemistry of 3
Lithium acetylide generated from 9 was reacted with
lactone 15 in THF–HMPA to produce hemiketal 16.
Axial hydride reduction of the C5411 hemiketal
(Et3SiH, BF3·OEt2, CH2Cl2, −30°C) afforded alkyne
17 in 68% overall yield as a single stereoisomer
OAc
OBn
X
Y
OAc
OAc
O
OH
OH
f,g
a-c
O
3
O
O
O
H
was unambiguously established by NOE and JH,H
H
H
data.15
4
7
5: X = OH, Y = H
6: X = H, Y = OH
d,e
1
Examination of the chemical shifts in the H and 13C
OBn
OBn
NMR spectra of 3 thus prepared showed that the
observed NMR data matched well with those of N-
acetylprymnesin-1 (NAPRM1) and N-acetylprym-
OTBS
OTf
OTBS
i,j
h
H
O
1
nesin-2 (NAPRM2). Especially, the H and 13C NMR
O
H
H
signals of the C51-C56 portion correspond extremely
well to those of NAPRM1 and 2 (Table 1). Further-
more, the coupling constants, J53,54=2.5 Hz, and
NOE data of 3 agreed well with those of NAPRM2
(Fig. 1). Thus, the formerly assigned configuration of
the HI/JK ring portion was confirmed to be repre-
sented by structure 3. Also, the present results
showed that the I/J-ring juncture adopts an excep-
tional CꢀC gauche conformer, although the precise
reason is not yet clear.16
8
9
Scheme 1. Reagents and conditions: (a) H2, Pd/C, MeOH,
rt; (b) NaOMe, MeOH, rt; (c) Me2C(OMe)2, PPTS, DMF,
rt, quant. (three steps); (d) (COCl)2, DMSO, Et3N, CH2Cl2,
−78“0°C; (e)
L-selectride, THF, −78°C, 78% (two steps);
(f) NaH, BnBr, DMF, 0°C“rt; (g) p-TsOH, MeOH, rt,
83% (two steps); (h) Tf2O, 2,6-lutidine, CH2Cl2, −78°C,
then TBSOTf; (i) trimethylsilylacetylene, n-BuLi, THF–
HMPA, −78°C, 77% (three steps); (j) K2CO3, MeOH, rt,
quant.

M. Sasaki et al. / Tetrahedron Letters 42 (2001) 5725–5728
5727
H
H
H
H
O
O
d-g
k-o
O
OMMTr
OH
a-c
HO
O
Ph
TIPSO
10
11
HO
H
H
H
H
O
O
56
OBn
OBn
OBn
OBn
h-j
TIPSO
TIPSO
12
13
HO
H
H
MPMO
J
O
H
H
OBn
OBn
O
K
+
OBn
OBn
TIPSO
O
O
14
H
15
Scheme 2. Reagents and conditions: (a) TIPSOTf, 2,6-lutidine, CH2Cl2, 0°C; (b) CSA, MeOH–CH2Cl2, rt, 84% (two steps); (c)
,
MMTrCl, Et3N, DMAP, CH2Cl2, rt, quant.; (d) TPAP, NMO, 4 A molecular sieves, CH2Cl2, rt; (e) L-selectride, THF, −78°C;
(f) CSA, MeOH–CH2Cl2, rt, 98% (three steps); (g) NaH, BnBr, DMF, 0°C“rt; (h) OsO4, NMO, acetone–H2O, rt, 93% (two
steps); (i) NaIO4, THF–H2O, rt; (j) allyltributyltin, MgBr2·OEt2, CH2Cl2, −78°C“rt, 90% (two steps) [+14, 7% (two steps)]; (k)
t-BuOK, MPMCl, n-Bu4NI, THF, 0°C“rt; (l) n-Bu4NF, THF, rt, 91% (two steps); (m) OsO4, NMO, THF–H2O, rt; (n) NaIO4,
,
THF–H2O, rt, 93% (two steps); (o) TPAP, NMO, 4 A molecular sieves, CH2Cl2, rt, quant.
Cl
O
OBn
H
H
H
O
OTBS
MPMO
TBS
O
BnO
O
OBn
OBn
X
H
H
O
H
Y
H
H
OBn
H
O
K
O
OBn
56
O
OBn
OBn
J
OTBS
b
57
54
f,g
c,d
H
O
H
a
9
O
R
OBn
O
H
H
H
O
H
16: R = OH
17: R = H
18: X = OH, Y = H
19: X = H, Y = Cl (J56,57 = 3 Hz)
20
e
Cl
Cl
Cl
H
H
H
H
H
H
H
O
O
O
BnO
OBn
BnO
O
OBn
OBn
HO
OH
OH
54 J
O
K
H
OH
H
H
H
H
O
O
O
I
j,k
l
h
O
OBn
O
53
52
53
H
O
H
R
H
H
H
H
O
i
OH
OH
H
H
H
21: R = O
22: R = β-OH, α-H
23 (J52,53 = 10 Hz)
3
Scheme 3. Reagents and conditions: (a) 9, n-BuLi, THF–HMPA, −78°C, then 15; (b) Et3SiH, BF3·OEt2, CH2Cl2, −30°C, 68% (two
steps); (c) H2, Lindlar catalyst, MeOH, rt; (d) DDQ, CH2Cl2 pH 7.0 phosphate buffer, rt; (e) PPh3, CCl4, reflux; (f) OsO4, NMO,
THF–H2O, rt, 39% (four steps); (g) (COCl)2, DMSO, Et3N, CH2Cl2, −78“0°C; (h) HF·pyr, THF, rt, 53% (two steps); (i)
DIBALH, CH2Cl2, −78°C; (j) p-TsOH, MeOH, 60°C; (k) Et3SiH, BF3·OEt2, CH2Cl2, 0°C; (l) H2, Pd(OH)2/C, MeOH, rt, 65%
(four steps).
1
Table 1. Selected H and 13C NMR data of 3, NAPRM1, and NAPRM2a
3
NAPRM1
NAPRM2
lH (pattern)
No
lH (pattern)
lC
lH (pattern)
lC
lC
51
51
52
53
54
55
55
56
1.57 (ddd, 11.3, 11.3, 11.3)
2.40 (ddd, 11.3, 4.5, 4.5)
3.59 (ddd, 11.3, 9.7, 4.5)
3.48 (dd, 9.7, 2.5)
4.47 (ddd, 11.4, 2.5, 2.2)
2.48 (ddd, 14.7, 11.4, 3.1)
2.03 (ddd, 14.7, 3.1, 2.2)
4.51 (ddd, 3.2, 3.1, 3.1)
41.5
1.64
2.45
41.3
1.60
2.42
3.59
3.46
4.43
2.47
2.03
4.48
41.4
68.0
85.7
74.2
35.7
3.63 (dt, 10, 5)
3.48 (dd, 10, 2.6)
4.47
2.46
2.02
4.46
67.5
85.2
73.8
35.3
67.8
85.6
74.2
35.8
60.5
60.3
60.3
a The spectra were all measured in CD3OD–C5D5N (1:1).

5728
M. Sasaki et al. / Tetrahedron Letters 42 (2001) 5725–5728
work from Kishi’s group, see: Zheng, W.; DeMattei, J.
A.; Wu, J.-P.; Duan, J. J.-W.; Cook, L. R.; Oinuma, H.;
Kishi, Y. J. Am. Chem. Soc. 1996, 118, 7946–7968.
7. Mori, Y.; Yaegashi, K.; Furukawa, H. J. Am. Chem. Soc.
1996, 118, 8158–8159.
8. Fuwa, H.; Sasaki, M.; Tachibana, K. Tetrahedron 2001,
57, 3019–3033.
9. Charette, A. B.; Mellon, C.; Pouillard, L.; Malenfant, E.
Synlett 1993, 81–82.
10. Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P.
Synthesis 1994, 639–666.
11. The numbering of carbon atoms of all compounds in this
paper corresponds to that of prymnesins.
12. Lewis, M. D.; Cha, J. K.; Kishi, Y. J. Am. Chem. Soc.
1982, 104, 4976–4978.
Figure 1.
13. (a) Hooz, J.; Gilani, S. S. H. Can. J. Chem. 1968, 46,
86–87; (b) Appel, R. Angew. Chem., Int. Ed. Engl. 1975,
14, 801–811.
Acknowledgements
14. Preference formation of the six-membered acetal over the
five-membered has been reported, see: (a) Fujiwara, K.;
Morishita, H.; Saka, K.; Murai, A. Tetrahedron Lett.
2000, 41, 507–508; (b) Matsuo, G.; Hinou, H.; Koshino,
H.; Suenaga, T.; Nakata, T. Tetrahedron Lett. 2000, 41,
903–906; (c) Mori, Y.; Mitsuoka, S.; Furukawa, H. Tet-
rahedron Lett. 2000, 41, 4161–4164.
We are grateful for Dr. T. Igarashi, Japan Food
Research Laboratories, and Professor M. Satake,
Tohoku University, for valuable discussions. This work
was supported by a Grant-in-Aid from the Ministry of
Education, Culture, Sports, Science and Technology of
Japan.
15. Selected data for compound 3: 1H NMR (500 MHz,
CD3OD/C5D5N=1:1): l 4.51 (1H, ddd, J=3.2, 3.1, 3.1
Hz, 56-H), 4.47 (1H, ddd, J=11.4, 2.5, 2.2 Hz, 54-H),
4.29 (1H, ddd, J=9.4, 9.3, 4.6 Hz, 58-H), 4.14 (1H, m,
48-H), 4.08 (1H, m, 60-H), 3.95 (1H, dd, J=11.4, 6.7 Hz,
62-H), 3.86 (1H, dd, J=11.4, 5.0 Hz, 62-H), 3.81 (1H,
ddd, J=11.9, 11.9, 2.4 Hz, 46-H), 3.53–3.67 (4H, 46-H,
50-H, 52-H, 61-H), 3.48 (1H, dd, J=9.7, 2.5 Hz, 53-H),
3.26 (1H, dd, J=9.3, 3.2 Hz, 57-H), 2.98 (1H, dd, J=9.5,
2.7 Hz, 49-H), 2.48 (1H, ddd, J=14.7, 11.4, 3.1 Hz,
55-H), 2.40 (1H, ddd, J=11.3, 4.5, 4.5 Hz, 51-H), 2.27
(1H, ddd, J=12.7, 4.6, 3.3 Hz, 59-H), 2.03 (1H, ddd,
J=14.7, 3.1, 2.2 Hz, 55-H), 1.77 (1H, m, 47-H), 1.72 (1H,
m, 47-H), 1.67 (1H, m, 59-H), 1.57 (1H, ddd, J=11.3,
11.3, 11.3 Hz, 51-H); 13C NMR (125 MHz, CD3OD/
C5D5N=1:1) l 85.7 (C53), 83.4 (C61), 82.3 (C49), 80.7
(C57), 74.2 (C54), 71.0 (C50), 69.6 (C58), 68.2 (C60), 68.0
(C52), 66.6 (C48), 64.2 (C46), 63.8 (C62), 60.5 (C56), 41.5
(C51), 39.4 (C59), 35.7 (C55), 35.0 (C47); HRMS (FAB)
calcd for C17H27ClNaO8 [(M+Na)+] m/z 417.1292, found
417.1280.
References
1. Igarashi, T.; Satake, M.; Yasumoto, T. J. Am. Chem.
Soc. 1996, 118, 479–480.
2. Igarashi, T.; Aritake, S.; Yasumoto, T. Nat. Toxins 1998,
6, 35–41.
3. Igarashi, T.; Satake, M.; Yasumoto, T. J. Am. Chem.
Soc. 1999, 121, 8499–8511.
4. Only recently, the absolute configurations at C14 and
C85 were determined to be S, respectively, see: Moro-
hashi, A.; Satake, M.; Igarashi, T.; Oshima, Y.;
Yasumoto, T. Chirality, in press. However, the relative
configuration of the C74–C85 acyclic part remains
unknown.
5. Glendenning, L.; Igarashi, T.; Yasumoto, T. Bull. Chem.
Soc. Jpn. 1996, 69, 2253–2263.
6. (a) Sasaki, M.; Nonomura, T.; Murata, M.; Tachibana,
K. Tetrahedron Lett. 1994, 35, 5023–5026; (b) Sasaki, M.;
Nonomura, T.; Murata, M.; Tachibana, K. Tetrahedron
Lett. 1995, 36, 9007–9010; (c) Sasaki, M.; Matsumori, N.;
Murata, M.; Tachibana, K.; Yasumoto, T. Tetrahedron
Lett. 1995, 36, 9011–9014; (d) Sasaki, M.; Matsumori,
N.; Maruyama, T.; Nonomura, T.; Murata, M.;
Tachibana, K.; Yasumoto, T. Angew. Chem. 1996, 108,
1782–1785; (e) Nonomura, T.; Sasaki, M.; Matsumori,
N.; Murata, M.; Tachibana, K.; Yasumoto, T. Angew.
Chem., Int. Ed. Engl. 1996, 108, 1786–1789; (f) For the
16. In CDCl3 solution, compound 3 preferentially adopted
the C–C anti conformation around the C53–C54 linkage
(J53,54=8.5 Hz), presumably due to intramolecular
hydrogen bond stabilization between the C52 hydroxyl
and the J ring oxygen. Similar conformational properties
of the V/W-ring juncture of maitotoxin have been
reported, see: Cook, L. R.; Oinuma, H.; Semones, M. A.;
Kishi, Y. J. Am. Chem. Soc. 1997, 119, 7928–7937.
.