Highly efficient stereocontrolled total synthesis of the polyfunctional carotenoid peridinin.

Full text of DOI:10.1002/1521-3773(20020315)41:6<1023::AID-ANIE1023>3.0.CO;2-A

COMMUNICATIONS
OH
3'
5'
Highly Efficient Stereocontrolled Total
Synthesis of the Polyfunctional Carotenoid
Peridinin**
15
O
10'
6'
H
•
15'
8
O
6
5
O
Noriyuki Furuichi, Hirokazu Hara, Takashi Osaki,
Hajime Mori, and Shigeo Katsumura*
3
AcO
OH
1
Peridinin (1; Scheme 1) was isolated from the planktonic
algae dinoflagellates,[1] which causes red tides, and is a
carotenoid of an auxiliary light-harvesting pigment for photo-
synthesis in the sea.[2] The structure of this carotenoid was
determined by Lieean-Jensen and co-workers.[3] Peridinin is a
unique, highly oxidized C37 nor-carotenoid that possesses an
allene and a ylidenebutenolide function in the main con-
jugated polyene chain as well as functionalized cyclohexane
rings at both ends of molecule. The synthesis of peridinin was
reported by Ito and co-workers in 1993, 100 years after its
isolation, and the structure of this unique carotenoid was
confirmed.[4] In that synthesis, however, control of stereo-
chemistry and therefore yield were not considered. The
stereocontrolled synthesis of peridinin presents several chal-
lenges: 1) the stereocontrolled construction of the all-trans-
conjugated polyene chain that contains the (Z)-g-ylidenebu-
tenolide moiety and the asymmetric allene function, and
2) the stereocontrolled construction of the oxygen functions
at the terminal cyclohexane ring. We overcame the problems
caused by the instability of the conjugated polyene chain by
effectively utilizing our own and/or new synthetic methods in
a highly efficient total synthesis of peridinin. The stereo-
chemistry of the six asymmetric carbons and the geometry
of six of the seven double bonds in the molecule were
controlled.
OH
H
•
O
OH
HO
O
O
AcO
OH
2
3
CHO
O
TBSO
4
Scheme 1. Polyfunctional carotenoid, peridinin (1), and its precursors 2, 3,
and 4.
numerous trials. Enantiomerically pure allylic alcohol 6[6] was
prepared from known vinyltriflate 5[4] by palladium-catalyzed
methoxycarbonylation[7] followed by LAH reduction
(Scheme 2). Sharpless epoxidation of 6 was realized by the
OTf
OH
O
a, b
TBSO
TBSO
In the retrosynthetic analysis of peridinin (1), we bisected
this molecule at the central C15 C15' double bond (carote-
noid numbering). The molecule was thus divided into an
5
6
OH
d
c
4
allene segment
2 and a ylidenebutenolide segment 3
TBSO
(Scheme 1). An important intermediate for the syntheses of
both segments was the chiral epoxyaldehyde derivative 4.
Complete stereocontrol at the C3, C5, and C6 positions of the
terminal cyclohexane ring had not been achieved in previous
syntheses of carotenoids. The stereochemistry at these stereo-
genic carbon atoms was controlled by finding precise reaction
conditions for the Sharpless asymmetric epoxidation[5] after
7
Scheme 2. Synthesis of 4: a) CO, [Pd(PPh3)4], Et3N, MeOH, DMF, 708C,
15 h (97%); b) LiAlH4, THF, 458C, 20 h (87%); c) (À)-diethyl-d-tartrate,
Ti(OiPr)4, 1.5m TBHP in toluene, molecular sieves (4 ä), CH2Cl2, À208C,
30 min (99%, 92% de); d) (COCl)2, DMSO, CH2Cl2, À788C, 40 min;
Et3N, 10 min (100%). DMF ˆ N,N-dimethylformamide, Tf ˆ trifluoro-
methanesulfonyl, TBHP ˆ tert-butylhydroperoxide.
use of freshly prepared reagents, solvents and exact amounts
of the reagents, ((À)-diethyl-d-tartrate (0.3 equivalents),
Ti(OiPr)4 (0.2equivalents), and tert-butylhydroperoxide
(2.0 equivalents)) at a constant temperature (À208C). Under
these precise conditions, we obtained the desired a-epoxide 7
in 99% yield and with 92% de, even on a 10-g scale. The
diastereoselectivity under usual conditions for the Sharpless
asymmetric epoxidation was 60 70% de. Epoxide 7 was
transformed into epoxyaldehyde 4 by means of the Swern
oxidation. This is the first example in which the stereo-
chemistry between the C3 hydroxy group and the C5, C6
epoxide in the cyclohexane terminal of carotenoids was
satisfactorily controlled.[4, 8]
[*] Prof. Dr. S. Katsumura, N. Furuichi, H. Hara, T. Osaki, Dr. H. Mori
Department of Chemistry, Faculty of Science, Kwansei Gakuin
University
Gakuen, Sanda, 669-1337, Hyogo (Japan)
Fax : (‡81)795-65-9077
E-mail: katsumura@ksc.kwansei.ac.jp
[**] We are grateful to Prof. Dr. M. Ito and Dr. Y. Yamano (Kobe
Pharmaceutical University) for providing samples of natural and
synthetic peridinin. We also thank Dr. K. Saiki (Kobe Pharmaceutical
University), Mr. T. Yamamoto and co-workers (Institute for Life
Science Research, Asahi Chemical Industry Co., Ltd.) for high-
resolution mass spectral measurements. This work was partially
supported by a Grant-in-Aid for Scientific Research from the Ministry
of Education, Science, Sports, and Culture of Japan, and a Sasagawa
Science Research Grant from The Japan Science Society for N.F.
The stereocontrolled synthesis of the allene segment 2 from
4 is shown in Scheme 3. The preparation of the allylic
Supporting information for this article is available on the WWW under
http://www.angewandte.com or from the author.
Angew. Chem. Int. Ed. 2002, 41, No. 6
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4106-1023 $ 17.50+.50/0
1023

COMMUNICATIONS
CO2Me
a, b
c
O
O
4
HO
HO
9
8
H
OH
•
1
e, f, g
d
2
5
3
HO
OH
10
h, i
CO2Me
I
I
OH
11
12
Scheme 3. Synthesis of 2: a) ClCH2P Ph3ClÀ, nBuLi, THF, À308C, 3 h;
b) tBuOK, DMSO, room temperature, 20 min (53% over two steps); c) 12,
[Pd(PPh3)4] , CuI, iPr2NH, room temperature, 1 h (84%); d) DIBAL,
CH2Cl2, 08C, 10 min (80%); e) MnO2, diethyl ether, room temperature,
3 h; f) Ac2O, pyridine, room temperature, 15 h (86% over two steps);
g) NaBH4, MeOH, room temperature 15 min (98%); h) MnO2, diethyl
ether, room temperature, 2h; i) (MeO) 2P(O)CH2CO2Me, NaH, THF,
room temperature, 5 min (74% over two steps). DMSO ˆ dimethyl
sulfoxide, DIBAL ˆ diisobutylaluminum hydride.
‡
Scheme 4. Synthesis of 18: a) 14, nBuLi, diethyl ether, 08C, 3 h; b) O2,
TPP, CH2Cl2, hn, À788C, 30 min (77% over two steps); c) iPr2EtN, DMSO,
room temperature, 3 h, then allyl bromide, room temperature, 1 h (70%);
d) CBr4, PPh3, Et3N, CH2Cl2, À608C, 1 h (89%); e) TBAF, THF, 458C,
20 h (81%). TPP ˆ 5,10,15,20-tetraphenyl-21H,23H-porphine, TBAF ˆ
tetra-n-butylammonium fluoride.
hydroxyallene moiety in 10 by means of SN2' hydride
reduction[9] of the conjugated ethynylepoxide group in 9 has
been reported. We thus concentrated on the stereocontrolled
preparation of 9. The stereocontrolled construction of the
conjugated polyene is the central problem in the synthesis of
carotenoids, and palladium-catalyzed sp sp2 and sp2 sp2 is
effective. Epoxyaldehyde 4 was transformed into acetylene
derivative 8.[8] Sonogashira cross-coupling[10] of 8 with vinyl
iodide 12 (which was prepared from 11[11]) in the presence of
catalytic amounts of tetrakis(triphenylphosphine)palladium
and cuprous iodide in diisopropylamine gave the desired ester
9 in 84% yield, with complete stereochemical control. Stereo-
specific SN2' reduction of 9 with DIBAL produced triol 10 in
80% yield. The C3 hydroxy group was acetylated by a sequence
of regioselective oxidation with MnO2, acetylation,[12] and
then reduction to give 2 with efficient stereocontrol.
also be obtained from 13 in a one-pot procedure: treatment of
13 with 1O2, allyl ester formation, and dibromide formation in
CH2Cl2 gave 17 in 53% overall yield from 4. Dibromide 17
was transformed into 18 in 81% yield by treatment with
TBAF.[18]
The novel and efficient one-pot stereocontrolled formation
of 3 from 18 was the highlight of our synthesis (Scheme 5). A
mixture of allyl ester 18 (Scheme 4) and vinyl iodide 11
(Scheme 3, 1 equivalent) was stirred in the presence of
catalytic amounts of tetrakis(triphenylphosphine)palladium
and cuprous iodide in triethylamine at room temperature for
1 h. After the complete consumption of 18 was ascertained by
TLC, formic acid was added to the reaction mixture at room
The most striking point in the synthesis of 1 was the
preparation of 3, which includes a characteristic conjugated
(Z)-g-ylidenebutenolide function. We have reported a highly
stereoselective synthesis of the sesquiterpene freelingyne[13]
by means of a PdII-catalyzed intramolecular lactonization of a
conjugated ene-yne carboxylic acid.[14] Analogously, a con-
jugated ethynylcarboxylic acid such as 20 (which could be
obtained from a conjugated (Z)-b-alkoxycarbonyldienal such
as 16, Schemes 4 and 5) could serve as a precursor of 3. We
therefore had to synthesize 16 in a stereocontrolled fashion.
Thus, epoxyaldehyde 4 was treated with the ylide derived
from our silylfurylmethane-Wittig salt 14[15] to give 13 as a
single stereoisomer (Scheme 4). The silylfuran moiety of 13
[16]
was chemoselectively oxidized with 1O2
to afford the
corresponding g-hydroxybutenolide 15. The stereocontrolled
ring opening of 15 by treatment with ethyldiisopropylamine
and allyl bromide in DMSO produced 16 in excellent yield.
Aldehyde 16 was transformed into the relatively stable
dibromide 17 by the standard method.[17] Dibromide 17 could
Scheme 5. Synthesis of 3: a) 11, [Pd(PPh3)4], CuI, Et3N, room temperature,
1 h; HCO2H, room temperature, 20 h (49%).
1024
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4106-1024 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 6

COMMUNICATIONS
OH
O
H
•
O
O
N
S
H
S
AcO
OH
1
•
O
O
a, b
2
AcO
OH
21
e
d
15
OH
H
O
•
c
3
OHC
O
AcO
OH
O
O
O
22
O
23
HO
Scheme 6. Total synthesis of 1: a) 2-mercaptobenzothiazole, PPh3, DIAD, THF, room temperature, 1.5 h (78%); b) (NH4)6Mo7O24 ¥ 4H2O, H2O2 (aq. 30%),
EtOH, 08C, 1 h (89%); c) MnO2, diethyl ether, room temperature, 5 min; d) NaHMDS, THF, À788C, 5 min, in the dark (50% from 22, E/Z 1:3); e) C6H6,
258C, 3 d, in the dark (E/Z > 5:1); preparative HPLC purification. DIAD ˆ diisopropyl azodicarboxylate, NaHMDS ˆ sodium bis(trimethylsilyl)amide,
HPLC ˆ high-performance liquid chromatography.
[1] F. Schutt, Ber. Dtsch. Bot. Ges. 1890, 8, 9.
[2] a) H. A. Frank, R. J. Cogdell in Carotenoids in Photosynthesis (Eds.:
temperature, and the resulting mixture was stirred for 20 h. As
expected, we obtained the desired ylidenebutenolide 3 in
49% yield as a single stereoisomer. Sonogashira coupling of
18 with 11, reductive deallylation,[19] and highly stereoselec-
A. Young, G. Britton), Chapman & Hall, London, 1993; b) E.
Hofmann, P. M. Wrench, F. P. Sharples, R. G. Hiller, W. Welte, K.
Diederichs, Science 1996, 272, 1788.
tive intramolecular lactonization proceeded successfully via
19 in one pot[20] by the successive action of Pd0 and PdII
catalysts.
[3] a) H. H. Strain, W. A. Svec, K. Aitzetmuller, M. C. Grandolfo, J. J.
Katz, H. Kjosen, S. Norgard, S. Lieean-Jensen, F. T. Haxo, P. Wegfahrt,
H. Rapoport, J. Am. Chem. Soc. 1971, 93, 1823; b) J. E. Johansen, G.
Borch, S. Lieean-Jensen, Phytochemistry 1980, 19, 441.
The final step in the synthesis of peridinin was the coupling
of 2 and 3. After several trials, we knew that the halides
derived from the C17-allene segment 2 were unstable under
the reaction conditions required for the preparation of the
corresponding Wittig salt, although the very similar C15-Wittig
salt has been reported.[20] We thus turned our attention to the
modified Julia Kocienski olefination,[21] which made the
olefination possible at a relatively lower temperature. Allene
2 was transformed into benzothiazole sulfone 21 by the
Mitsunobu reaction with 2-mercaptobenzothiazole, followed
by molybdenum(vi)-catalyzed oxidation[22] (Scheme 6). The
reaction between the sulfone group of 21 and the aldehyde
function of 22 (derived from 3) proceeded successfully within
5 min by using NaHMDS at À788C in the dark. Although the
crude product obtained was a mixture of the desired all-trans-
[4] a) Y. Yamano, M. Ito, J. Chem. Soc. Perkin Trans. 1 1993, 1599; b) M.
Ito, Y. Yamano, S. Sumiya, A. Wada, Pure Appl. Chem. 1994, 66, 939.
[5] a) T. Katsuki, K. B. Sharpless, J. Am. Chem. Soc. 1980, 102, 5976;
b) R. M. Hanson, K. B. Sharpless, J. Org. Chem. 1986, 51, 1922; c) Y.
Gao, R. M. Hanson, J. M. Klunder, S. Y. Ko, H. Masamune, K. B.
Sharpless, J. Am. Chem. Soc. 1987, 109, 5765.
[6] Y. Yamano, C. Tode, M. Ito, J. Chem. Soc. Perkin Trans. 1 1998, 2569.
[7] a) A. Schoenberg, I. Bartoletti, R. F. Heck, J. Org. Chem. 1974, 39,
3318; b) J. K. Stille, P. K. Wong, J. Org. Chem. 1975, 40, 532; c) M.
Hidai, T. Hikita, Y. Wada, Y. Furikura, Y. Uchida, Bull. Chem. Soc.
Jpn. 1975, 48, 2075.
[8] A. Baumeler, W. Brade, A. Haag, C. H. Eugster, Helv. Chim. Acta.
1990, 73, 700.
[9] E. Widmer, Pure Appl. Chem. 1985, 57, 741; alternatively, we reported
the novel synthesis of the allene moiety of carotenoids by means of
biomimetic photosensitized oxygenation: M. Nakano, N. Furuichi, H.
Mori, S. Katsumura, S. Tetrahedron Lett. 2001, 42, 7307.
[10] K. Sonogashira in Comprehensive Organic Synthesis, Vol. 3 (Eds.:
B. M. Trost, I. Fleming, G. Pattenden), Pergamon, Oxford, 1991, p. 52 1.
[11] Vinyl iodide 12 was prepared from the commercially available
2-butyn-1-ol by means of stannylcupration with nBu3Sn(nBu)CuCN-
Li2 (B. H. Lipshutz, J. A. Kozlowski, R. S. Wilhelm, J. Org. Chem.
1984, 49, 3943) followed by tin halogen exchange.
[12] Y. Yamano, S. Sumiya, M. Ito, J. Chem. Soc. Perkin Trans. 1 1995, 167.
[13] H. Mori, H. Kubo, H. Hara, S. Katsumura, Tetrahedron Lett. 1997, 38,
5311.
[14] Original reports: a) C. Lambert, K. Utimoto, H. Nozaki, Tetrahedron
Lett. 1984, 25, 5323; b) X. Lu, X. Huang, S. Ma, Tetrahedron Lett. 1993,
34, 5963; this method was developed for natural products syntheses by
other groups: c) M. Kotora, E. Negishi, Tetrahedron Lett. 1996, 37,
9041; d) F. Liu, E. Negishi, J. Org. Chem. 1997, 62, 8591; e) M. Kotora,
E. Negishi, Synthesis 1997, 121; f) E. Negishi, M. Kotora, Tetrahedron
1
peridinin (1) and its 15-cis isomer 23 (1:3 based on H NMR
spectroscopic analysis), we found that the Z isomer was
converted into the thermodynamically more stable E isomer
in a solution of benzene at room temperature in the dark after
1
3 days (E/Z > 5:1 based on H NMR spectroscopic analysis
and analytical HPLC). The desired enantiomerically pure
peridinin (1) was obtained after purification by means of
preparative HPLC in the dark. The spectral data of the
synthetic peridinin were in good agreement with those
reported.[3, 4]
Received: November 16, 2001 [Z18233]
Angew. Chem. Int. Ed. 2002, 41, No. 6
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4106-1025 $ 17.50+.50/0
1025

COMMUNICATIONS
age.[5] Strand cleavage occurs selectively at guanine, the most
readily oxidized of the nucleobases, and is sequence selec-
tive.[6, 7] Preferential cleavage at GG and GGG sequences has
been attributed to stabilization of the hole by delocalization
1997, 53, 6707; g) R. Rossi, F. Bellina, M. Biagetti, L. Mannina,
Tetrahedron Lett. 1998, 39, 7799; h) C. Xu, E. Negishi, Tetrahedron
Lett. 1998, 40, 431; i) R. Rossi, F. Bellina, A. Catanese, L. Mannina, D.
Valensin, Tetrahedron 2000, 56, 479.
[15] a) K. Tanaka, M. Kamatani, H. Mori, S. Fujii, K. Ikeda, M. Hisada, Y.
Itagaki, Y, S. Katsumura, Tetrahedron 1999, 55, 1657; b) T. Hata, K.
Tanaka, S. Katsumura, Tetrahedron Lett. 1999, 40, 1731; c) N. Furuichi,
T. Hata, H. Soetjipto, M. Kato, S. Katsumura, Tetrahedron 2001, 57,
8425.
10]
over two or three guanine residues.[8
We recently reported the use of transient absorption
spectroscopy with kinetic modeling to obtain the rate
.
‡
constants for reversible hole transports between G and
GG or GGG sequences separated by a single A:T base
pair.[11, 12] The resulting Gibbs energy differences, 52mV for
GG and 77 mV for GGG, are much smaller than those
previously estimated from calculations of gas-phase ioniza-
tion potentials.[8] However, recent theoretical studies by
Bixon and Jortner[13, 14] and by Kurnikov et al.[15] have
provided results in good agreement with our experimental
data. Our kinetic data have also been used by Giese and co-
workers[5, 16] to model the results of strand-cleavage studies in
duplexes containing G, GG, and GGG sites.
[16] S. Katsumura, K. Hori, S. Fujiwara, S. Isoe, Tetrahedron Lett. 1985, 26,
4625.
[17] a) E. J. Corey, P. L. Fuchs, Tetrahedron Lett. 1972, 13, 3769; b) D.
Grandjean, P. Pale, J. Chuche, Tetrahedron Lett. 1994, 35, 3529.
[18] a) F. Naso, L. Ronzini, J. Chem. Soc. Perkin Trans. 1 1974, 340; b) K. C.
Nicolaou, C. V. C. Prasad, P. K. Somers, C.-K. Hwang, J. Am. Chem.
Soc. 1989, 111, 5330; c) I. C. Gonzalez, C. J. Forsyth, J. Am. Chem. Soc.
2000, 122, 9099.
[19] a) T. Mandai, K. Ohta, N. Baba, M. Kawada, J. Tsuji, Synlett 1992, 671;
b) C. J. Solomon, E. G. Mata, O. A. Mascaretti, Tetrahedron 1993, 49,
3714; c) J. Tsuji, T. Mandai, Synthesis 1996, 1.
[20] Y. Yamano, C. Tode, M. Ito, J. Chem. Soc. Perkin Trans. 1 1995, 1895.
[21] a) J. B. Baudin, G. Hareau, S. A. Julia, O. Ruel, Tetrahedron Lett. 1991,
32, 1175; b) J. B. Baudin, G. Hareau, S. A. Julia, O. Ruel, Bull. Soc.
Chim. Fr. 1993, 130, 336; c) J. B. Baudin, G. Hareau, S. A. Julia, R.
Lorne, O. Ruel, Bull. Soc. Chim. Fr. 1993, 130, 856; d) P. R.
Blakemore, W. J. Cole, P. J. Kocienski, A. Morley, Synlett 1998, 1, 2 6.
[22] H. S. Schultz, H. B. Freyermuch, S. R. Buc, J. Org. Chem. 1963, 28,
1140.
The importance of obtaining experimental values for the
dynamics and energetics of hole transport in DNA led us to
investigate the use of the modified nucleobase 7-deazagua-
nine, Z, as a hole trap. The oxidation potential of Z is reported
to be 0.3 eV lower than that of G.[17] Z has been employed in
investigations of single-step (superexchange) electron transfer
20]
in DNA.[18
Both Z and 8-oxoguanine, which also has a
lower oxidation potential than G, have been used as hole traps
in strand cleavage studies.[6, 21] We report here the results of an
investigation of the dynamics of forward and return hole
transport from G to Z, which establish that Z is a much deeper
hole trap than GG or GGG. In addition, the dynamics of the
hole transport processes are found to be strongly dependent
upon the number of A:T base pairs separating the hole donor
and acceptor.
The structures of the synthetic DNA hairpins prepared for
the present study are shown in Scheme 1. The stilbene-4,4'-
dicarboxamide (Sa) serves as a linker connecting comple-
mentary polyT and polyA arms containing one or more C:G
or C:Z base pairs.[22] All of these hairpins have high melting
temperatures (>758C) and circular dichroism spectra similar
to that of a stilbene-linked hairpin known to adopt a normal
B-form structure in which the stilbene is parallel to the
Dynamics and Energetics of Hole Trapping in
DNA by 7-Deazaguanine**
Frederick D. Lewis,* Jianqin Liu, Xiaoyang Liu,
Xiaobing Zuo, Ryan T. Hayes, and
Michael R. Wasielewski*
Oxidative cleavage of DNA by photonucleases can occur at
sites separated from the nuclease by as many as several dozen
4]
base pairs.[1 Such long-distance processes are known to
occur by a multistep hole-hopping mechanism which is
initiated by photoinduced electron transfer between the
nuclease (electron acceptor) and a neighboring base (electron
donor) resulting in creation of a hole on the base. Migration of
the hole over long distances is more rapid than the chemical
reactions of the oxidized bases which lead to strand cleav-
O
O
N
N
N
NH
NH2
NH
guanine, G
Sa =
7-deazaguanine, Z
NH2
N
N
[*] Prof. F. D. Lewis, Prof. M. R. Wasielewski, J. Liu, Dr. X. Liu, X. Zuo,
R. T. Hayes
O
H
N
O
( )3
( )3
N
Department of Chemistry
Northwestern University
O
H
O
Evanston, IL 60208-3113 (USA)
Fax : (‡1)847-467-2184
Sa
Sa
Sa
Sa
Sa
Sa
T
A
T
T
C
T
C
A
A
G
A
G
T
T
C
T
C
A
T
T
C
T
T
A
A
G
A
A
T
T
C
T
T
A
A
G
A
A
T
T
C
T
T
A
A
Z
A
A
E-mail: lewis@chem.northwestern.edu
wasielew@chem.northwestern.edu
T
C
T
T
A
G
A
A
A
G
A
Z
[**] We thank R. S. Kalgutkar for advice on the kinetic modeling. This
research was supported by grants from the Division of Chemical
Sciences, Office of Basic Energy Sciences, US Department of Energy
under contracts DE-FG02-96ER14684 (FDL) and DE-FG02-
99ER14999 (MRW).
C
C
G
G
C
T
G
A
T
T
A
A
C
T
Z
A
T
T
A
A
T
T
A
A
3G
3Z
3G,6Z
3,5,6G
3G,5Z
3,6,7G
Supporting information for this article is available on the WWW under
http://www.angewandte.com or from the author.
Scheme 1. Structures of the synthetic DNA hairpins used in this study.
1026
¹ WILEY-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002
1433-7851/02/4106-1026 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2002, 41, No. 6