Natural-product hybrids: Design, synthesis, and biological evaluation of quinone - Annonaceous acetogenins

Article abstract of DOI:10.1002/1521-3773(20000616)39:12<2099::AID-ANIE2099>3.0.CO;2-Y

Replacement of the butyrolactone unit of the natural products with a quinone subunit has resulted in the synthesis of natural-product hybrids of quinone-mucocin (1). These compounds, which are combined from partial structures of annonaceous acetogenins an

Full text of DOI:10.1002/1521-3773(20000616)39:12<2099::AID-ANIE2099>3.0.CO;2-Y

COMMUNICATIONS
[7] a) A. Varki, Glycobiology 1993, 3, 97 ± 130; b) A. Varki, Proc. Natl.
Acad. Sci. USA 1994, 91, 7390 ± 7397.
this slice had corresponding protons that, in common, gave
trNOEs to H1Fuc. In addition, scalar coupling partners were
easily identified since the w1/w2 slice reflected TOCSY
pathways. The scalar coupling pattern was attributed to
protons proR-H6GlcNAc, proS-H6GlcNAc, and H5GlcNAc, as indi-
cated in Figure 2. It should be emphasized that the intra-
glycosidic trNOEs between H2Fuc and H3Fuc were also present
in this spectrum and overlapped with the proR-H6GlcNAc/proS-
H6GlcNAc pattern. However, the scalar coupling of the latter
two protons to H5GlcNAc at d ˆ 3.51 allowed a clear discrim-
ination. Inspection of Scheme 2 shows that only a (1 !6)
linkage between the two sugar residues explains the observed
cross peak pattern.
In summary, a 3D-TOCSY-trNOESY experiment allowed
unambiguous identification of the bioactive component of a
carbohydrate library without prior knowledge of the individ-
ual components of the library. Carbohydrates certainly
represent an important class of biological macromolecules[7, 8]
but, nonetheless, the experimental protocol described here is
not limited to carbohydrate derivatives.
Â
[8] J. Jimenez-Barbero, J.-L. Asensio, F.-J. CanÄada, A. Poveda, Curr. Opin.
Struct. Biol. 1999, 9, 549 ± 555.
[9] A. L. Davis, G. Estcourt, J. Keeler, E. D. Laue, J. J. Titman, J. Magn.
Reson. A 1993, 105, 167 ± 183; b) T. Weimar, Magn. Reson. Chem., in
press.
Natural-Product Hybrids: Design, Synthesis,
and Biological Evaluation of
Quinone ±Annonaceous Acetogenins**
Sabine Hoppen, Ulrich Emde, Thorsten Friedrich,
Lutz Grubert, and UlrichKoert*
The acetogenins of Annonaceae are a class of natural
products withinteresting antitumor, immunosuppressive,
pesticide, and antimicrobial activities.[1] Their main mode of
action is the inhibition of the mitochondrial complex I
(NADH-ubiquinone oxidoreductase).[2] Mucocin[3] and squa-
mocin D[4] are two representative members of the Annona-
ceae acetogenins, which show their characteristic structural
features: An ether core consisting of tetrahydrofuran (THF)
and tetrahydropyran (THP) rings flanked with a left and a
right side chain. At the end of the right side chain is located a
butenolide unit. The long alkyl chains place at least parts of
the compound in the lipophilic interior of the mitochondrial
membrane.[5]
Experimental Section
All spectra were acquired on a Bruker DRX 500 MHz spectrometer witha
5 mm TXI probehead at 306 K. TOCSY and STD-TOCSY were measured
with512 increments and 16 transients using a 50 ms MLEV-17 spinlock
field of 7.5 kHz. Saturation transfer was achieved by using 40 selective 2708
Gaussian pulses of duration 50 ms and spacing 10 ms. The protein envelope
was irradiated at 2.8 ppm (on-resonance) and 40 ppm (off-resonance).
Subsequent subtraction of on- and off-resonance spectra was achieved via
phase cycling.
For the 3D-TOCSY-trNOESY experiment, 124, 256, and 1014 data points
in F1, F2, and F3, respectively, were acquired with eight transients each.
The acquisition time was 127 ms, which results in a total relaxation delay of
1.6 s. The total measurment time was approximately five days. The mixing
time for the NOESY step was 130 ms. Suppression of zero-quantum
coherence[9] was achieved by parallel application of a spinlock field (2.5 ms,
7.5 kHz) and a gradient pulse (2.5 ms, approximately 5 GcmÀ1) prior to the
NOE mixing time. Signals from HDO were suppressed by presaturation
witha weak radio-frequency field.
It has been proposed that the annonaceous acetogenins act
at the terminal electron-transfer step of complex I between
[2]
À
the Fe S cluster N2 and the ubiquinone pool. The buteno-
lide subunit may bind at the quinone binding site of
complex I. The reduction potential of an a-alkyl-a,b-unsatu-
rated butyrolactone (Epc ˆ À2.69 V (irrev.), CH3CN, versus
the saturated calomel electrode (SCE)) is much more
negative than the reduction potentials of the quinone group
(EcpI ˆ À0.75 V, EcpII ˆ À1.48 V, CH3CN, versus SCE).[6] There-
Received: February 2, 2000 [Z14634]
À
fore an electron transfer from the Fe S cluster N2 to the
butenolide is unlikely to occur. With the goal to further
elucidate the mechanism of action of the acetogenins and to
provide molecular probes for complex I studies we designed
quinone ± mucocin 1 and quinone ± squamocin D 2. In these
compounds the butenolide part of mucocin and squamocin D
is exchanged against the quinone part of ubiquinone. Here we
[1] a) J. M. Moore, Curr. Opin. Biotechnol. 1999, 10, 54 ± 58; b) P. Keifer,
Curr. Opin. Biotechnol. 1999, 10, 43 ± 41; c) B. J. Stockman, Prog. Nucl.
Magn. Reson. Spectrosc. 1998, 33, 109 ± 151.
[2] a) P. J. Hajduk, T. Gerfin, J.-M. Boehlen, M. Haeberli, D. Marek, S. W.
Fesik, J. Med. Chem. 1999, 42, 2315 ± 2317; b) S. B. Shuker, P. J. Hajduk,
R. P. Meadows, S. W. Fesik, Science 1996, 274, 1531 ± 1534.
[3] a) P. J. Hajduk, E. T. Olejniczak, S. W. Fesik, J. Am. Chem. Soc. 1997,
119, 12257 ± 12261; b) A. Chen, M. J. Shapiro, J. Am. Chem. Soc. 1998,
120, 10258 ± 19259; c) M. Lin, M. J. Shapiro, J. R. Wareing, J. Org.
Chem. 1997, 62, 8930 ± 8931.
[4] a) M. Mayer, B. Meyer, Angew. Chem. 1999, 111, 1902 ± 1906; Angew.
Chem. Int. Ed. 1999, 38, 1784 ± 1788; b) J. Klein, R. Meinicke, M.
Mayer, B. Meyer, J. Am. Chem. Soc. 1999, 121, 5336 ± 5337.
[5] a) B. Meyer, T. Weimar, T. Peters, Eur. J. Biochem. 1997, 246, 705 ± 709;
b) T. Peters, B. Meyer, WO-B 19649359, 1996; c) D. Henrichsen, B.
Ernst, J. L. Magnani, W.-T. Wang, B. Meyer, T. Peters, Angew. Chem.
1999, 111, 106 ± 110; Angew. Chem. Int. Ed. 1999, 38, 98 ± 102.
[6] a) C. Griesinger, O. W. Sùrensen, R. R. Ernst, J. Magn. Reson. 1989, 84,
14 ± 63; b) H. Oschkinat, C. Cieslar, C. Griesinger, J. Magn. Reson.
1990, 86, 453 ± 469.
[*] Prof. Dr. U. Koert, Dipl.-Chem. S. Hoppen, Dipl.-Chem. U. Emde,
Dr. L. Grubert
Institut für Chemie der Humboldt-Universität zu Berlin
Hessische Strasse 1-2, 10115 Berlin (Germany)
Fax : (‡49)30-2093-7266
E-mail: koert@chemie.hu-berlin.de
Dr. T. Friedrich
Institut für Biochemie
Heinrich-Heine Universität Düsseldorf (Germany)
[**] This work was supported by the Deutsche Forschungs-gemeinschaft,
Fonds der Chemische Industrie, and ASTA-Medica AG.
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/angewandte/ or from the author.
Angew. Chem. Int. Ed. 2000, 39, No. 12
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3912-2099 $ 17.50+.50/0
2099

COMMUNICATIONS
OMe
OMe
OMe
a
MeO
O
OMe
OMe
b
MeO
OMe
OMe
O
OMe
O
OH
O
8
3
4
ubiquinone
OMe
MeO
OMe
OMe
H
O
PPh3I
O
H
TBDPSO
H
HO
6
R
O
O
5
OH
H
HO
H
H
H
c
OMe
MeO
OMe
O
O
HO
OMe
mucocin: R =
O
TBDPSO
H
H
7
OMe
O
OMe
d
OMe
1:
R =
RO
O
MeO
OMe
OMe
H
I
O
OR
H
H
O
O
H
H
C8H17
9 R = TBDMS
8
e
R
O
O
HO
OH
O
HO
H
H H
H
OMe
O
MeO
OMe
OMe
squamocin D: R =
HO
OMe
O
O
H
H
OH OH H
H
OMe
O
O
C8H17
f
2:
R =
10
1
Scheme 1. a) nBuLi, TMEDA, n-hexane, 08C, 30 min, succinic anhydride,
THF, 2 h, 63%; b) 1. LiAlH4, THF, 3 h, 72%; 2. Et3SiH, BF3 ´ OEt2,
CH2Cl2, 12 h, 88%; 3. (COCl)2, DMSO, NEt3, CH2Cl2, À408C, 2.5 h,
78%; c) 6, NaHMDS, THF, 08C, 20 min; 5, 08C, 1 h, 87%; d) 1. Pt/C, H2,
EtOAc, 6 h, 97%; 2. nBu4NF, THF, 2.5 h, 95%; 3. (COCl)2, DMSO, NEt3,
CH2Cl2, À408C, 1.5 h, 89%; e) 1. 9 (1.3 equiv), tBuLi (2.4 equiv), Et2O,
À1058C, 4 min, MgBr2 ´ OEt2 (2.8 equiv), À100 ! À 408C (2 h )!À788C, 8
(1 equiv) !À108C, 2 h, 47%, diastereoselectivity 7:1, chromatographic
separation of the epimers; 2. HF, CH3CN/CH2Cl2, 1 h, 90%; f) 2,6-
pyridinedicarboxylic acid, CAN, CH3CN/H2O, 0 8C, 4 h, 68%. TBDPS ˆ
tert-butyldiphenylsilyl, TMEDA ˆ tetramethylethylenediamine, NaHMDSˆ
sodium hexamethyldisilazide, CAN ˆ Cer(IV) ammonium nitrate.
report on the first synthesis of this novel type of hybrid
compounds as well as on their interaction with mitochondrial
complex I.
The syntheses of 1 and 2 build upon the modular strategy
developed for our total syntheses of mucocin[7] and squamo-
cin D.[8] ortho-Lithiation of 2,3,4,5-tetramethoxytoluene (3)[9]
and reaction with succinic anhydride gave the oxocarboxylic
acid 4 (Scheme 1). Reduction with LiAlH4 provided a diol,
whose benzyl alcohol function was removed by reduction with
Et3SiH/BF3 ´ OEt2.[10] A subsequent Swern oxidation to the
aldehyde 5 followed by a Wittig reaction with the phospho-
nium salt 6[11] delivered the olefin 7, which was transformed
into the aldehyde 8 by a standard sequence. A chelation-
controlled addition of an organomagnesium compound pre-
pared from the iodide 9[7a,b] by iodine lithium exchange/
transmetalation to the aldehyde 8 gave, after removal of the
tert-butyldimethylsilyl (TBDMS) protecting groups, the triol
10. Oxidation of the tetramethoxytolyl subunit with ceric
ammonium nitrate (CAN)[12] yielded the target compound
quinone ± mucocin 1 (Table 1).
and removal of the TBDMS groups the triol 15. A CAN
oxidation of 15 yielded the target compound quinone ± squa-
mocin D (2, Table 1).
The synthetic compounds 1, 2, 10, and 15 were tested as
inhibitors of complex I (bovine-heart mitochondrial; Ta-
ble 2).[14±16] The known strong inhibitor rotenon (IC50 ˆ
1.0 nm) was used as a reference sample.[2] It was found that
quinone ± mucocin (1, IC50 ˆ 3.6 nm) was ten times more
active than mucocin itself (IC50 ˆ 33 nm). In contrast the
hydroquinone ± mucocin dimethyl ether (10) was a weaker
inhibitor than the natural product. Quinone ± squamocin D
(2) and its hydroquinone-dimethyl ether (15) also displayed
activities in the nanomolar range (IC50 ˆ 1.7 and IC50 ˆ
4.7 nm, respectively). These results show that natural-product
hybrids 1 and 2 are very potent inhibitors of mitochondrial
The starting point for the synthesis of quinone ± squamo-
cin D (2) was the aldehyde 11 (Scheme 2). Its reaction with
lithiated compound 3 gave 12, which was converted into the
iodide 13. Iodine ± metal exchange and addition to the
aldehyde 14[13] gave after oxidation/l-selectride reduction
2100
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3912-2100 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 12

COMMUNICATIONS
Table 1. Selected analytical and spectroscopical data of compounds 1 and
2.
Table 2. Inhibition of mitochondrial complex I by natural acetogenins and
synthetic analogues.
1: [a]2D2 ˆ À20.5 (c ˆ 0.044, CH2Cl2); 1H NMR (300 MHz, CDCl3): d ˆ 0.85
(t, J ˆ 6.4 Hz, 3H, 34-H3), 1.13 ± 1.88 (m, 42H, alkyl), 1.89 ± 2.14 (m, 3H, 13,
14, 22-H2), 1.99 (s, 3H, 35-Me), 2.42 (t, J ˆ 6.8 Hz, 2H, 3-H2), 2.69 (brs, 1H,
OH), 2.82 (brs, 1H, OH), 3.02 (dt, J ˆ 8.8, 2.2 Hz, 1H, 24-H), 3.08 ± 3.18 (m,
1H, 20-H), 3.18 ± 3.32 (m, 1H, 23-H), 3.34 ± 3.52 (m, 2H, 16, 19-H), 3.71 ±
3.90 (m, 2H, 12, 15-H), 3.96 (s, 6H, 37, 38-OMe); 13C NMR (75 MHz,
CDCl3): d ˆ 11.9 (35-Me), 14.1 (C-34), 22.7, 25.5, 26.2, 26.4, 26.9, 28.3, 28.3,
28.7, 28.7, 29.3, 29.4, 29.4, 29.5, 29.6, 29.7, 29.7, 29.8, 31.9, 32.0, 32.4, 32.6,
35.6, (C-3 to C-11, C-13, C-14, C-17, C-18, C-21, C-22, C-25 to C-33), 61.1
(37, 38-OMe), 70.6 (C-23), 73.5 (C-19), 73.8 (C-16), 79.3 (C-12), 80.1 (C-20),
81.9 (C-15), 82.0 (C-24), 138.6, 143.1, 144.3 (C-2, C-35, C-37, C-38), 184.2,
Compound
IC50
IC50
IC50
[nm][a]
[mmol/ mgprotein][a]
[mmol/ mgprotein][b]
mucocin
1
10
squamocin D
squamocin A
2
15
34
3.6
123
±
1.0
1.7
4.7
1
45
4.9
163
±
1.3
2.3
6.2
1.3
33.3
8.7
rotenone
‡
[a] Bovine-heart mitochondria were prepared as described before.[15] The
inhibition of oxygen uptake was measured. The respiratory activities were
analyzed with a Clark-type oxygen electrode (100 mm sodium phosphate
pH ˆ 7.4, 1 mm N,N,N',N'-ethylenediaminetetraacetic acid (EDTA), 1 mm
MgCl2, 0.5 mgmLÀ1 protein).[16] [b] Literature data.[14b]
184.7 (C-1, C-36); HR-MS (EI): calcd for C41H72O9 ([M ‡ 2H]): 708.5176;
found: 708.5166.
2: [a]2D2 ˆ ‡3.1 (c ˆ 0.13, CHCl3); 1H NMR (300 MHz, CDCl3): d ˆ 0.86 (t,
J ˆ 6.6 Hz, 3H, 34-H3), 1.20 ± 1.67 (m, 42H, 4-CH2 to 14-H2, 17-H', 18-H',
21-H', 22-H', 25-H2 to 27-H2, 29-H2 to 33-H2), 1.89 ± 2.10 (m, 4H, 17-H'', 18-
H'', 21-H'', 22-H''), 1.99 (s, 3H, 39-H3), 2.42 (t, J ˆ 6.4 Hz, 2H, 3-H2), 3.37 ±
3.53 (m, 2H, 15-H, 24-H) overlap with3.53 ± 3.62 (m, 1H, 28-H), 3.84 ± 3.98
(m, 10H, 16-H, 19-H, 20-H, 23-H, 2 Â OCH3); 13C NMR (75 MHz, CDCl3):
d ˆ 11.9 (CH3 ± 39), 14.1 (C-34), 21.4 (C-26), 22.6 (C-33), 25.4, 25.7, 26.4,
28.67, 28.74, 29.4, 29.5, 29.6, 29.7, 29.9, 31.9 (C-3 to C-13, C-17, C-18, C-21,
C-22, C-30 to C-32), 32.6 (C-25), 33.1 (C-14), 36.8, 37.7 (C-27, C-29), 61.2
(OCH3), 71.7 (C-28), 74.4, 74.7 (C-15, C-24), 81.7, 81.8, 82.9, 83.0 (C-16,
C-19, C-20, C-23), 138.7, 143.1, 144.2 (C-2, C-35, C-37, C-38), 184.2, 184.7
none binding site of complex I.[2c] Our data support this
hypothesis, although other structural features may also
contribute, since the IC50 of quinone 2 and hydroquinone
dimethyl ether 15 are in the same range. It has been
demonstrated that the quinone-binding site in complex I is
quite large, probably occupied by bothacetogenin subunits,
the butenolide as well as the THF/THP part.[14a]
‡
(C-1, C-36); HR-MS: (EI): calcd: 708.5176 ([M ‡ 2H] ); found: 708.5177.
Natural-product hybrids of type 1 and 2 should be useful
molecular probes for further investigations of the ubiquinone
binding site of complex I. In particular a more detailed look at
the redox properties of the system complex I/quinone anno-
naceous acetogenin seems promising. The new modular
synthetic approach presented here allows the efficient access
to quinone ± annonaceous hybrids with different quinone
moieties and different redox properties.
OMe
MeO
9
OMe
OMe
a
O
+
3
H H
O
TrO
TrO
H
9
OH
11
12
b
OMe
Received: February 14, 2000 [Z14696]
MeO
OMe
OMe
H
O
H
13C6
O
O
I
[1] a) F. Q. Alali, X.-X. Liu, J. L. McLaughlin, J. Nat. Prod. 1999, 62, 504 ±
H
H
TBDMSO TBDMSO
9
Á
540; b) M. C. Zafra-Polo, B. Figadere, T. Gallardo, J. R. Tormo, D.
14
13
Â
Á
Cortes, Phytochemistry 1998, 48, 1087 ± 1117; c) A. Cave, B. Figadere,
A. Laurens, D. Cortes in Progress in the Chemistry of Organic Natural
Products (Eds.: W. Herz, G. W. Kirby, R. E. Moore, W. Steglich, C.
Tamm), Springer, New York, 1997, 70, pp. 81 ± 288.
c
OMe
[2] a) M. Degli Eposti, Biochim. Biophys. Acta 1998, 1364, 222 ± 235;
b) H. Miyoshi, M. Ohshima, H. Shimada, T. Akagi, H. Iwamura, J. L.
McLaughlin, Biochim. Biophys. Acta 1998, 1365, 443 ± 452; c) T.
Friedrich, P. Van Heek, H. Leif, T. Ohnishi, E. Forche, B. Kunze, R.
Jansen, W. Trowitzsch-Kienast, G. Höfle, H. Reichenbach, H. Weiss,
Eur. J. Biochem. 1994, 219, 691 ± 698.
[3] G. Shi, D. Alfonso, M. O. Fatope, L. Zeng, Z.-M. Gu, K. He, J. M.
McDougal, J. L. McLaughlin, J. Am. Chem. Soc. 1995, 117, 10409 ±
10410.
[4] a) M. Sahai, S. Singh, M. Singh, Y. K. Gupta, S. Akashi, R. Yuji, K.
Hirayama, H. Asaki, H. Araya, N. Hara, T. Eguchi, K. Kakinuma, Y.
Fujimoto, Chem. Pharm. Bull. 1994, 42, 1163 ± 1174; b) G.-X. Zhao,
L. R. Miesbauer, D. L. Smith, J. L. McLaughlin, J. Med. Chem. 1994,
37, 1971 ± 1974.
MeO
9
OMe
OMe
H13C6
HO
O
OH
HO
H
H H
H
d
15
2
Scheme 2. a) nBuLi, TMEDA, n-hexane, 08C, 2.5 h, 54%; b) 1. Et3SiH,
BF3 ´ Et2O, CH2Cl2, À78 !208C, 14 h, 81%; 2. I2, PPh3, imidazole, CH2Cl2,
08C, 4 h, 65%; c) 1. 13, tBuLi, MgBr2 ´ Et2O, Et2O, À100 ! À 40 ! À 788C,
14, À78 !08C, 3.5H, 52%; 2. (COCl)2, DMSO, NEt3, CH2Cl2, À60 !
À 408C, 1.5 h, 96% ; 3. l-selectride, THF, À100 ! À 708C, 1.5 h, 99%,
diastereoselectivity 88:12, separation of the epimers by chromatography;
4. 5% HF in CH3CN, THF, 208C, 1.5 h, 77%; d) CAN, 2,6-pyridinedicar-
boxylic acid, 08C, 4 h, 70%. Tr ˆ triphenylmethyl.
[5] H. Shimada, J. B. Grutzner, J. F. Kozlowski, J. L. McLaughlin,
Biochemistry 1998, 37, 854 ± 866.
[6] The cyclic voltammograms of various quinones and butenolide
samples are given in the Supporting Information.
[7] a) S. Bäurle, S. Hoppen, U. Koert, Angew. Chem. 1999, 111, 1341 ±
1344; Angew. Chem. Int. Ed. 1999, 38, 1263 ± 1266; b) S. Hoppen, S.
Bäurle, U. Koert, Chem. Eur. J. 2000, in press; c) S. Takahashi, T.
Nakata, Tetrahedron Lett. 1999, 40, 723 ± 726; d) S. Takahashi, T.
Nakata, Tetrahedron Lett. 1999, 40, 727 ± 730; e) P. Neogi, T. Dound-
doulakis, A. Yazbak, S. C. Sinha, S. C. Sinha, E. Keinan, J. Am. Chem.
complex I. The structural similarity between the butenolide
and the quinone led to the assumption that both groups
interact in a similar way withthe ubiquinone reduction site in
complex I[2b] and previous studies postulated that the annona-
ceous acetogenins are competitive inhibitors at the ubiqui-
Angew. Chem. Int. Ed. 2000, 39, No. 12
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3912-2101 $ 17.50+.50/0
2101

COMMUNICATIONS
Soc. 1998, 120, 11279 ± 11284; f) P. A. Evans, V. S. Murthy, Tetrahe-
dron Lett. 1999, 40, 1253 ± 1256; f) P. A. Evans, V. S. Murthy,
Tetrahedron Lett. 1998, 39, 9627 ± 9628; g) P. A. Evans, J. D. Roseman,
Tetrahedron Lett. 1997, 38, 5249 ± 5252.
resins. In 1972 Asahi Chemical started to produce the acetal
homopolymer utilizing the worldꢁs third type of polyacetal
technology.[3] Asahi Chemical also industrialized the acetal
copolymer in 1985.[4] At present, the annual demand of acetal
resins in the world is about 400000 t per year.
[8] a) U. Emde, U. Koert, Tetrahedron Lett. 1999, 40, 5979 ± 5982; b) U.
Emde, U. Koert, Eur. J. Org. Chem. 2000, in press.
[9] E. Keinan, D. Eren, J. Org. Chem. 1987, 52, 3872 ± 3875.
[10] a) J. L. Fry, M. Orfanopoulos, M. G. Atlington, W. R. Dittmann, S. B.
Silverman, J. Org. Chem. 1978, 43, 374 ± 375; b) I. Fleming, A.
Barbero, D. Walter, Chem. Rev. 1997, 97, 2063 ± 2192.
[11] The phosphonium salt 6 was prepared from (2S,5S)-5-[(tert-butyldi-
phenylsiloxy)methyl]tetrahydrofurane-2-carbaldehyde (U. Koert, M.
Stein, H. Wagner, Liebigs Ann. 1995, 1415 ± 1426) by side-chain
elongation witha Wittig reaction.
The biggest problem in the case of acetal resins is the
energy consumption during its production. The main aspect is
the energy requirement in the monomer process. For example,
it requires a great deal of energy to get the monomeric
trioxane that is needed for the acetal copolymer from aqueous
formaldehyde. In the commercial process, trioxane is obtained
by heating aqueous formaldehyde in the presence of an acid
catalyst like sulfuric acid [Eq. (1)].
[12] L. Syper, K. Kloc, J. Mlochowski, Z. Szula, Synthesis 1979, 521 ± 522.
[13] 14 was prepared by routes described for the synthesis of squamocin D
see reference [8].
[14] a) M. Oshima, H. Miyoshi, K. Sakamoto, K. Takegami, J. Iwata, K.
Kuwabara, H. Iwamura, T. Yagi, Biochemistry 1998, 37, 6436 ± 6445;
b) D. Alfonso, H. A. Johnson, T. Colman-Saaizarbitooria, C. P.
Presley, G. P. McCabe, J. L. McLaughlin, Nat. Toxins 1996, 181 ± 188.
[15] A. L. Smith, Methods Enzymol. 1967, 10, 81 ± 86.
[16] P. C. Hinkle, M. L. Yu, J. Biol. Chem. 1979, 254, 2450 ± 2455.
Even though the equilibrium concentration of trioxane is
low in the reaction mixture, in the commercial production
process trioxane is removed as the distillate from the reaction
mixture in the distillation tower.[5] The vapor± liquid equili-
brium between trioxane and aqueous formaldehyde is such
that when the trioxane concentration in the liquid phase is
low, trioxane shows a very high volatility compared to
formaldehyde and water in the vapor phase. Thus, almost all
the trioxane contained in the vapor phase from the reaction
mixture can be concentrated into the distillate in the
distillation tower under the proper refluxing conditions. Since
the heat of vaporization of the water± formaldehyde mixture
is much higher than that of trioxane,[6] most of the energy for
trioxane synthesis is consumed in the vaporization of water
and formaldehyde.[7] From the viewpoint of the energy
requirement, the key point of the trioxane synthesis is the
highyield and highselectivity in a one-pass vaporization.
At Asahi Chemical, we developed the tert-butyl alcohol
process, that is the selective hydration of isobutene using a
highly concentrated heteropolyacid as the catalyst.[8] We also
developed a new process for producing polyoxytetramethy-
lene glycol witha narrow molecular weight distribution using
a heteropolyacid as a catalyst for the polymerization of
tetrahydrofuran.[9, 10] With this knowledge, we investigated the
catalytic activity of heteropolyacids for the synthesis of
trioxane. We observed some interesting phenomena and a
superior advantage of heteropolyacids over conventional
catalysts like sulfuric acid.
Synthesis of Trioxane Using Heteropolyacids as
Catalyst
Junzo Masamoto,* Katsuhiko Hamanaka,
Kohichi Yoshida, Hajime Nagahara, Kenji Kagawa,
Toshiyuki Iwaisako, and Hajime Komaki
Acetal resin is a term used to describe the high molecular
weight polymers and the copolymers of formaldehyde. First
commercialized as a homopolymer in 1960 by DuPont, acetal
resins are engineering thermoplastics which have found broad
use in areas where traditionally metals were applied.[1] Shortly
thereafter, researchers at Celanese developed an acetal resin
based on the copolymerization of trioxane and cyclic ethers,
[2]
suchas ethylene oxide.
In 1962 a commercial plant began
producing this acetal copolymer. Since then, a rapid expan-
sion of acetal resin production has occurred worldwide.
Up to 1971 DuPont and Celanese (alone or in joint ventures
with other companies) were the sole producers of acetal
[*] Prof. J. Masamoto[‡]
Department of Polymer Science and Engineering
Kyoto Institute of Technology
Sakyo-ku, Kyoto 606-8585 (Japan)
E-mail: masamoto@ccmails.fukui-ut.ac.jp
The results for the reaction at atmospheric pressure and
1008C are shown in Table 1. The best features of using the
heteropolyacid, instead of sulfuric acid, as a catalyst for
trioxane synthesis were the higher conversion and selectivity.
For example, for the same selectivity of 97%, the conversion
by sulfuric acid was 20%, while the conversion by the
heteropolyacid was 27% (drawn from entries 10 and 6
respectively in Table 1). Heteropolyacids provided a 35%
higher yield (yield ˆ conversion  selectivity) than sulfuric
acid. A similar result was also obtained for the hydration of
isobutene.[8] In this case, high selectivity of heteropolyacids
was reported to be related to the big size of their anions.
K. Hamanaka, K. Yoshida, H. Nagahara, K. Kagawa, T. Iwaisako,
H. Komaki
Technical Research Laboratory
Asahi Chemical Industry Co., Ltd.
Fuji-shi 416- 8501 (Japan)
‡
[ ] New address:
Department of Management Science
Fukui University of Technology
Gakuen, Fukui-shi 910-8505 (Japan)
Fax : (‡81)776-29-7891
2102
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3912-2102 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 12