Efficient hydrolysis of RNA by a PNA-diethylenetriamine adduct

Full text of DOI:10.1002/(SICI)1521-3773(20000117)39:2<369::AID-ANIE369>3.0.CO;2-A

COMMUNICATIONS
Typical CI source conditions were as follows: source temperature 1008C;
repellor voltage 0.0 V; ion-extraction voltage 8 kV; source pressure 0.1 ±
0.3 Torr. The NR experiments were performed in the first of the collision
cells positioned between the magnet and the second electrostatic analyzer,
using Xe or CH4 as the neutralizing colliders, at a pressure adjusted to
achieve a 80% transmittance. Reionization was achieved utilizing O2 as the
collider, approximately at the same transmittance. Any ions remaining
after the first collision event were deflected from the primary neutral beam
using a high-voltage electrode (1 kV), whose efficiency was checked by
suitable control experiments. The NR spectra were averaged over 20 ±
50 acquisitions to achieve a satisfactory signal-to-noise ratio.
Efficient Hydrolysis of RNA by a PNA ±
Diethylenetriamine Adduct**
Jeroen C. Verheijen, Birgit A. L. M. Deiman,
Esther Yeheskiely, Gijs A. van der Marel, and
Jacques H. van Boom*
A few years ago, Komiyama et al. reported[1] that a hybrid
composed of diethylenetriamine (DETA) anchored to the 5'-
end of DNA by means of a urethane bond (Figure 1a)
hydrolyzed linear RNA selectively at the 3'-side of cytosine 22
(C22, marked with an arrow) to give a 22-mer RNA fragment
The gases were of research grade from commercial sources with a stated
purity exceeding 99.95 mol% and were used without further purification.
Ozone was prepared from dry O2 in a commercial ozonizer, collected in a
silica trap at 77 K and recovered by controlled warming of the trap. 18O2
and H218O (>99 18O atom%) were obtained from Isotec (Miamisburg,
‡
‡
USA). The (H2O ´ O2) ion was prepared by hydration of O2 in the positive
O2/CI of water[11]. Much higher yields were achieved by displacement of O3
by H2O in the O5 complex[12], a very efficient process as demonstrated by
‡
kinetic experiments performed utilizing a Fourier-transform ion-cyclotron
resonance mass spectrometer (47e Apex, Bruker Spectrospin AG, Bremen,
Germany). The (H2O ´ O2) ion was prepared by hydration of O2 in the
negative CI of moist oxygen.[13]
Received: July 22, 1999
Revised: September 29, 1999 [Z13767]
Figure 1. a) Complex I of RNA and DNA ± DETA as employed by
Komiyama et al.; b) complex II of RNA and PNA ± DETA. Arrows
indicate cleavage positions of RNA for hydrolysis by DNA ± DETA or
PNA ± DETA. Nucleotide units are written in uppercase, PNA units in
lowercase.
[1] C. T. R. Wilson, Philos. Trans. R. Soc. London A 1899, 192, 403.
[2] a) F. J. M. Farley, Proc. R. Soc. London A 1951, 207, 527; b) W. A.
Hoppel, D. E. Dinger, J. Atmos. Sci. 1973, 30, 331; c) H. Reiss, D. C.
Marvin, R. H. Heist, J. Colloid Interface Sci. 1977, 58, 155; d) D. Clark,
J. F. Noxon, Science 1971, 174, 941; e) F. C. Wen, T. McLaughlin, J. L.
Katz, Phys. Rev. A 1982, 26, 2235.
with a 3'-phosphate terminus. This selective scission was
ascribed to intramolecular acid ± base cooperation of an
ammonium ion and a neutral amine in the ethylenediamine
moiety [N(CH2)2NH2] of the DNA ± DETA adduct.[2] Inter-
estingly, the total conversion for the RNA hydrolysis was only
10 mol% after incubation of the RNA ´ DNA ± DETA com-
plex I (Figure 1a) at pH 8 for 4 h at 508C. The relatively low
conversion of the RNA substrate may be attributed to less
effective hydrogen bonding in complex I. We surmised that a
higher conversion could be attained by decreasing signifi-
cantly the freedom in dangling motion of the duplex in the
complex.
It is generally accepted that peptide nucleic acids (PNAs)[3]
nicely mimic the physical properties of DNA. Thus, PNAs
hybridize sequence specifically to RNA (DNA). They bind,
however, more strongly to RNA (DNA)[4] due to the presence
of a neutral polyamide backbone. The latter features, together
[3] W. Byers Brown, Chem. Phys. Lett. 1995, 235, 94.
[4] a) W. Byers Brown, M. A. Vincent, K. Trollope, I. H. Hillier, Chem.
Phys. Lett. 1992, 192, 213; b) W. Byers Brown, I. H. Hillier, A. J.
Masters, I. J. Palmer, H. V. Dos Santos, M. Stein, M. A. Vincent,
Faraday Discuss. Chem. Soc. 1995, 100, 253; c) I. J. Palmer, W. Byers
Brown, I. H. Hillier, J. Chem. Phys. 1996, 104, 3198.
[5] H. V. Dos Santos, S. J. Vaughn, E. V. Akhmatskaya, M. A. Vincent,
A. J. Masters, J. Chem. Soc. Faraday Trans. 1997, 93, 2781.
[6] a) D. V. Zagorevskij, J. L. Holmes, Mass Spectrom. Rev. 1994, 13, 133;
b) C. A. Schalley, G. Hornung, D. Schröder, H. Schwarz, Chem. Soc.
Rev. 1998 27, 91.
[7] The 6 kcalmol 1 value recently reported in the case of H2OO survival
is to be regarded as a lower limit, see: D. Schröder, C. A. Schalley, N.
Ï Â
Goldberg, J. Hrusak, H. Schwarz, Chem. Eur. J. 1996, 2, 1235, and
references therein.
[8] J. O. Hirschfelder, C. F. Curtiss, R. B. Bird, Molecular Theory of Gases
and Liquids, Wiley, New York, 1964.
[9] a) J. L. Holmes, Mass Spectrom. Rev. 1989, 8, 513; b) N. Goldeberg, H.
Schwarz, Acc. Chem. Res. 1994, 27, 347.
[10] Other conceivable candidates would be neutral species formed in
‡
‡
‡
NR and NR experiments in highly excited, long-lived Rydberg-
states, see: a) G. Herzberg, Annu. Rev. Phys. Chem. 1987, 38, 27.
However, their role is denied by the results of our NR and NR
experiments, since electron attachment to a Rydberg-state species is
expected to be difficult, see ref. [7].
[*] Prof. Dr. J. H. van Boom, J. C. Verheijen, Dr. B. A. L. M. Deiman,
Dr. E. Yeheskiely, Dr. G. A. van der Marel
Leiden Institute of Chemistry
‡
Gorlaeus Laboratories
[11] N. G. Adams, D. K. Bohme, D. B. Dunkin, F. C. Fehsenfeld, E. E.
Ferguson, J. Chem. Phys. 1970, 52, 3133.
P.O. Box 9502, 2300
RA Leiden (The Netherlands)
[12] F. Cacace, R. Cipollini, G. de Petris, F. Pepi, M. Rosi, A. Sgamellotti,
Inorg. Chem. 1998, 37, 1398.
Fax : (‡ 31)71-527-4307
E-mail: j.boom@chem.leidenuniv.nl
[13] J. D. Payzant, P. Kerbarle, J. Chem. Phys. 1972, 56, 3482.
[**] This work was supported by the Netherlands Foundation for Chemical
Research (SON) with financial aid from the Netherlands Organization
for Scientific Research (NWO). We thank Hans van den Elst for
synthesizing the RNA sequence and recording the mass spectra.
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. 2
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3902-0369 $ 17.50+.50/0
369

COMMUNICATIONS
with the fact that PNAs resist biodegradation,[5] were an
incentive to replace the DNA ± DETA adduct in complex I by
a PNA ± DETA adduct as in complex II (Figure 1b).
As part of an ongoing program[6] towards the design and use
of antisense probes based on PNA, we here report the
preparation and ribonucleolytic properties of a PNA attached
by means of a urea bond to DETA.
nBu3SnH resulted, after extraction, in the isolation of the
crude free amine 4. Reaction of 4 with para-nitrophenyl
chloroformate led, after purification on silica gel, to the
activated derivative 5 in an overall yield of 42%.
At this stage, the PNA decamer complementary to the
G7 ± A16 stretch of the RNA in complex II was assembled
following a well-established solid-phase synthesis.[6] Thus,
sequential elongation of glycine anchored by an ester bond to
functionalized polystyrene beads (i. e. 6 in Scheme 1) with
Fmoc/N-acyl-protected building blocks 7a ± d[9] under the
agency of the coupling reagent HATU and subsequent
reaction of the resulting resin-bound 10-mer PNA 8 with
excess 5 in the presence of DIPEA for 16 h at 208C gave
immobilized 9 (see Experimental Section). Removal of the N-
acyl protecting groups from the nucleobases in 9 and
concomitant release from the solid support was effected by
ammonolysis (NH3/MeOH) to give, after purification by
reversed-phase high-performance liquid chromatography
(RP-HPLC), the homogeneous N,N'-Boc protected PNA ±
DETA adduct 10. Treatment of the latter adduct with TFA
(75%) resulted in the isolation of the fully deprotected
PNA ± DETA adduct 11, the homogeneity and identity of
which was firmly established by RP-HPLC and matrix-
assisted laser desorption/ionization time-of-flight mass spec-
trometry (MALDI-TOF MS).
Having the 10-mer PNA ± DETA adduct 11 and the 25-mer
RNA in hand, we examined the ribonucleolytic behavior of 11
by degrading the target 25-mer RNA at 408C in a Tris ´ HCl
(10mm) buffer containing NaCl (100 mm) and EDTA
(0.1 mm) at pH 7. In this respect, it is of interest to note that
the pH is one unit below that required for optimal cleavage of
the phosphodiester bonds in RNA by intramolecular acid ±
base cooperation. The hydrolysates were analyzed by electro-
phoresis on 20% denaturing polyacrylamide gel.
Based on the RNA substrate employed by Komiyama et al.
(Figure 1a), a 25-mer RNA (Figure 1b) containing a 10-mer
sequence (i. e. G7 ± A16) complementary to the adduct, as
well as two potential neighboring C ± A scission sites[7]
adjacent to A16, was chosen as the target RNA. The RNA
25-mer, prepared according to a well-established solid-phase
approach,[8] was 32P-labeled at the 5'-end using adenosine 5'-
[g-32P]-triphosphate and T4 polynucleotide kinase. The
PNA ± RNA complex in which the diethylenetriamine moiety
in PNA 11 is replaced by an acetyl group (PNA-Ac) showed a
Tm of 698C (pH 7, 100mm NaCl), indicating that the PNA ±
DETA adduct forms a sufficiently stable duplex with the
target RNA.
The construction of the 10-mer PNA ± DETA adduct 11
commences with the four-step transformation of DETA (1)
into the N,N'-Boc-protected and activated derivative 5
(Scheme 1). Treatment of excess 1 in dichloromethane with
MMTrCl (see Scheme 1 for abbreviations) led to the isolation
of the crude mono-MMTr derivative 2. Treatment of the latter
with excess Boc2O in dioxane gave, after purification on silica
gel, the fully protected derivative 3. Removal of the N-MMTr
group of 3 with TFA in the presence of the scavenger
R'
N
Boc
N
O
d
R'HN
NHR
BocHN
N
H
OpNO2Ph
1 R = R' = H
a
b
c
5
2 R = MMTr, R' = H
3 R = MMTr, R' = Boc
4 R = H, R' = Boc
A typical electrophoresis pattern is presented in Figure 2.
Cursory inspection of the traces in lanes 4 ± 8 shows, after a
B
H Gly HMBA
O
O
6
N
FmocHN
OH
e
7a B = A(Bz)
b B = C(Bz)
c B = G(iBu)
d B = T
B
O
O
N
H N
Gly-HMBA-
n
H
8 B = A(Bz), C(Bz), G(iBu), T
f
B
O
O
R
N
O
N
N
R'
10
RHN
N
H
H
9 B = A(Bz), C(Bz), G(iBu), T, R = Boc, R' = Gly-HMBA-
10 B = A, C, G, T, R = Boc, R' = Gly-NH2
11 B = A, C, G, T, R = H, R' = Gly-NH2
g
h
Figure 2. Autoradiograph for the cleavage of the 5'-labeled RNA 25-mer
in Tris ´ HCl buffer at pH 7 and 408C in the presence of EDTA (0.1 mm).
Lane 1: PNA ± DETA (t ˆ 0); lane 2: T1 digestion; lane 3: alkaline
hydrolysis ladder; lane 4: PNA ± DETA (t ˆ 1 h); lane 5: PNA ± DETA
(t ˆ 2 h); lane 6: PNA ± DETA (t ˆ 4 h); lane 7: PNA ± DETA (t ˆ 8 h);
lane 8: PNA ± DETA (t ˆ 24 h). [RNA]0 ˆ 60nm, [PNA ± DETA] ˆ 2mm.
EDTA ˆ ethylenediaminetetraacetic acid, Tris ˆ tris(hydroxymethyl)ami-
nomethane.
Scheme 1. a) Monomethoxytrityl chloride (MMTrCl), CH2Cl2; b) Boc2O
(di-tert-butyldicarbonate), dioxane, 73% (2 steps); c) 5% trifluoroacetic
acid (TFA), Bu3SnH, CH2Cl2, 91%; d) pNO2PhOC(O)Cl, Et3N, CH2Cl2,
63%; e) stepwise elongation with 7a ± d; f) 5, N,N-diisopropylethylamine
(DIPEA), DMF; g) NH3/MeOH, 508C; h) 75% TFA, H2O. Boc ˆ tert-
butyloxycarbonyl, Fmoc ˆ fluorene-9-ylmethoxycarbonyl, HMBA ˆ para-
hydroxymethylbenzoic acid. The polystyrene support is represented by
shaded circles.
370
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3902-0370 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 2

COMMUNICATIONS
lapse of 24 h, the presence of two major 5'-end labeled RNA
fragments. Comparison of the mobility of the two cleavage
products with T1 digestion (lane 2) and alkaline hydrolysis
(lane 3) of the target RNA (lane 1) revealed that both
fragments resulted from hydrolysis at the 3'-side of C17 and
C19, respectively. The total conversion for the RNA hydro-
lysis was approximately 29 mol% after 4 h at 408C, as gauged
by densitometry (Figure 3). It is also worthwhile mentioning
that the cleavage of the phosphodiester bond in close
proximity of the duplex structure in complex II is relatively
less efficient (see, for instance, Figure 3). A similar preference
for the more distant cleavage site was also reported in other,
closely related studies.[7a,b, 10]
nucleases based on PNAwill be of great interest for the future
design of other types of chemical nucleases.
Experimental Section
Solid-phase synthesis of PNA ± DETA adduct 11: All solvents (Biosolve,
DNA synthesis grade) were used as received. Solid-phase synthesis was
performed on a Pharmacia Gene Assembler using highly cross-linked
polystyrene beads as the solid support (loading: 26 ± 28 mmolg 1) on a
1 mmol scale. The support was functionalized with a glycine moiety
attached to a para-hydroxymethylbenzoic acid linker. Assembly of the
PNA part was established using solutions of 0.3m of monomers 7a ± d, 0.3m
DIPEA and 0.3m HATU (O-(azobenzotriazol-1-yl)-1,1,3,3-tetramethy-
luronium hexafluorophosphate) in acetonitrile/dimethylformamide (1/
1 v/v). Prior to coupling, the monomers were preactivated for 1 min by
mixing equal amounts of the PNA monomer (15 equiv per mmol support),
HATU and DIPEA solutions. The protocol for one PNA chain extension
cycle consisted of 1) wash: acetonitrile/dimethylformamide (1/1 v/v),
2.5 mL; 2) coupling: PNA ‡ HATU ‡ DIUPEA in acetonitrile/dimethyl-
formamide (1/1 v/v), 15 min; 3) wash: acetonitrile/dimethylformamide (1/
1 v/v), 2.5 mL; 4) capping: Ac2O/lutidine/N-methylimidazole/tetrahydro-
furan (1/1/1/7 v/v/v/v), 2.0 mL; 5) wash: acetonitrile/dimethylformamide
(1/1 v/v), 2.5 mL; 6) Fmoc removal: 20% piperidine in acetonitrile/
dimethylformamide (1/1 v/v), 3 min; 7) wash: acetonitrile/dimethylforma-
mide (1/1 v/v), 2.5 mL. After the last elongation step, the solid support was
transferred to a separate reaction vessel, and a solution of 5 (12 mg,
25 mmol) and diisopropylethylamine (5 mL) in dimethylformamide
(0.5 mL) was added. After mixing for 16 h, the reagents were drained
from the vessel, and the solid support was washed with dimethylformamide
(2 mL) followed by dichloromethane (2 mL). At this stage, the oligomers
were cleaved from the support with concomitant deprotection of the
exocyclic amino groups by treatment with methanolic ammonia (1.5 mL) at
508C for 16 h. RP-HPLC purification and analysis were carried out on a
Jasco HPLC system equipped with a LiChrospher 100 RP-18 end-capped
column (10.0 Â 250 mm and 4.0 Â 250 mm, respectively). Gradient elution
was performed at 408C by building up a gradient starting with buffer A
(0.1% TFA in water) and applying buffer B (0.1% TFA in acetonitrile/
water, 9/1 v/v) with a flow rate of 1.0 mLmin 1. The Boc groups of purified
10 were removed by treatment with 75% TFA. Homogeneous deprotected
PNA ± DETA adduct 11 was obtained after freeze-drying. The identity and
homogeneity of oligomers 10 and 11 were confirmed by MALDI-TOF MS
Figure 3. Relative amounts (y) of intact and cleaved RNA at different
&
*
reaction times. Intact RNA 25-mer ( ), fragment 1 (RNA 19-mer) ( ),
~
^
fragment 2 (RNA 17-mer) ( ), summed amount of fragments 1 and 2 ( ).
In contrast with the latter observations, no scission was
observed when the 10-mer PNA-Ac was used instead of the
PNA ± DETA conjugate 11. The same holds for incubation of
the 25-mer RNA with diethylenetriamine in the presence or
absence of the PNA-Ac decamer, thus illustrating the
sequence specificity of RNA hydrolysis by the PNA ± DETA
adduct 11.
‡
and RP-HPLC. 10: MALDI-TOF MS: m/z: 3139.1 [M‡H] , 3160.2
‡
[M‡Na] ; calcd for C125H165N65O35: m/z: 3138.1. 11: MALDI-TOF MS:
‡
‡
m/z: 2939.6 [M‡H] , 2962.3 [M‡Na] ; calcd for C115H149N65O31: m/z:
2937.9.
Apart from the two main cleavage products an additional
minor degradation product, resulting from scission after U6,
could be observed upon prolonged incubation (see lanes 6 ±
8). Since participation of the glycinamide in the cleavage is
ruled out by the inactivity of the PNA-Ac, the formation of
the latter fragment may indicate that the spatial arrangement
of the RNA ´ PNA ± DETA complex II positions the ethyl-
enediamine residue of the PNA ± DETA in close proximity
(i. e. approximately one helical turn) to the phosphodiester
bond between U6 and G7 in the RNA substrate.
RNA degradation with PNA ± DETA adduct 11: All buffers were made
from highly purified Milli Q water and were sterilized before use. Labeled
RNA 25-mer and PNA ± DETA were mixed in Tris ´ HCl buffer (10mm,
pH 7) containing NaCl (100 mm) and EDTA (0.1 mm) to give the following
concentrations: [RNA] ˆ 60nm, [PNA ± DETA] ˆ 2mm. The samples
(70 mL) were incubated at 408C. After incubation, the RNA was
immediately precipitated by addition of 3m NaOAc (pH 5, 7 mL), EtOH
1
(225 mL), and 10 mgmL tRNA (1 mL). The precipitated RNA was
recovered by centrifugation, and was then dried and redissolved in 5 mL
water and 5 mL loading buffer. The solutions were heated at 808C for 1 min,
centrifuged and analyzed on a 20% denaturing electrophoresis gel.
These results show for the first time that a PNA-based
adduct cleaves an RNA sequence specifically at micromolar
concentration under physiological conditions. An additional
but nonetheless essential feature of this type of adduct is the
fact that almost complete conversion of RNA can be attained
(see e. g. Figure 3) with a relatively short stretch of PNA, thus
obviating the need for an excessively long sequence of DNA
to reach the same goal.[11] In this respect, we note that our
PNA approach nicely complements the recently reported
selective cleavage of HIV-1 TAR-RNAwith a peptide ± cyclen
conjugate.[12] In summary, the ease of preparing artificial
Received: July 22, 1999 [Z13768]
[1] M. Komiyama, T. Inokawa, K. Yoshinari, J. Chem. Soc. Chem.
Commun. 1995, 77 ± 78.
[2] a) K. Yoshinari, K. Yamazaki, M. Komiyama, J. Am. Chem. Soc. 1991,
113, 5899 ± 5901; b) M. Komiyama, K. Yoshinari, J. Org. Chem. 1997,
62, 2155 ± 2160.
[3] P. E. Nielsen, M. Egholm, R. H. Berg, O. Buchardt, Science 1991, 254,
1497 ± 1500.
[4] M. Egholm, O. Buchardt, L. Christensen, C. Behrens, S. M. Freier,
D. A. Driver, R. H. Berg, S. K. Kim, B. Norden, P. E. Nielsen, Nature
1993, 265, 566 ± 568.
Angew. Chem. Int. Ed. 2000, 39, No. 2
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3902-0371 $ 17.50+.50/0
371

COMMUNICATIONS
[5] V. V. Demidov, V. N. Potaman, M. D. F. Kamenetskii, M. Egholm, O.
Buchardt, S. H. Sonnichsen, P. E. Nielsen, Biochem. Pharmacol. 1994,
48, 1310 ± 1313.
Me
HO
Me
H
H
H
H
H
O
O
H
H
O
H
O
O
[6] a) A. C. van der Laan, N. J. Meeuwenoord, R. S. Oosting, R. Brands,
E. Kuyl-Yeheskiely, J. H. van Boom, Recl. Trav. Chim. Pays-Bas 1995,
114, 295 ± 297; b) A. C. van der Laan, R. Brill, R. G. Kuimelis, E. Kuyl-
Yeheskiely, J. H. van Boom, Tetrahedron Lett. 1997, 38, 2249 ± 2252;
c) A. C. van der Laan, P. Havenaar, R. S. Oosting, E. Kuyl-Yehskiely,
E. Uhlmann, J. H. van Boom, Bioorg. Med. Chem. Lett. 1998, 8, 663 ±
668; d) J. C. Verheijen, A.-M. M. van Roon, A. C. van der Laan, G. A.
van der Marel, J. H. van Boom, Nucleosides Nucleotides 1999, 18,
493 ± 508; e) J. C. Verheijen, G. A. van der Marel, J. H. van Boom,
S. F. Bayly, M. R. Player, P. F. Torrence, Bioorg. Med. Chem. 1999, 7,
449 ± 455.
Me
H
O
O
H
O
O
H
O
H
H
H
H
H
H
H
HO
OH
H
H
O
H
O
O
Me
Me
Figure 1. The structure of CTX-3C.
targets, providing a stimulus for the development of many
new reactions and strategies for the construction of polycyclic
systems.[3] This work has culminated in the recent total
syntheses of brevetoxins A and B, completed by Nicolaou and
co-workers,[4] as well as several syntheses of the smaller
congener hemibrevetoxin B.[5]
[7] C ± A sites in RNA are known to be particularly susceptible to
hydrolysis through intramolecular acid ± base cooperation. See: a) M.
Komiyama, T. Inokawa, T. Shiiba, N. Takeda, K. Ysohinari, M.
Yashiro, Nucleic Acids Symp. Ser. 1993, 29, 197 ± 198; b) V. Vlassov, T.
Â
Abramova, T. Godovikova, R. Giege, V. Silnikov, Antisense Nucleic
Acid Drug Dev. 1997, 7, 39 ± 42; c) V. V. Vlassov, G. Zuber, B. Felden,
Â
J.-P. Behr, R. Giege, Nucleic Acids Res. 1995, 23, 3161 ± 3167; d) M. A.
We recently presented a powerful strategy for the con-
struction of subunits found in the brevetoxins and ciguatoxins.
Catalytic ring-closing metathesis (RCM)[6] reactions of enol
ethers,[7a,b] allylic ethers,[7c,d] or alkynyl ethers[7e] were promot-
ed by the Schrock catalyst [2,6-iPr2C6H3NMo{OC(CF3)2-
Me}2CHCMe2Ph] (1)[8] or the Grubbs catalyst [Cl2{(c-C6H11)3P}2-
RuCHPh] (2),[9] leading to the formation of cyclic ethers.[10]
Stereoselective elaboration of the RCM products then gave
fully functionalized, saturated or unsaturated, cyclic ethers
suitable for conversion into polyoxacyclic units of the type
found in the natural products.[7d]
Â
Podyminigin, V. V. Vlassov, R. Giege, Nucleic Acids Res. 1993, 21,
5950 ± 5956.
[8] R. T. Pon in Methods in Molecular Biology, Protocols for Oligonu-
cleotides and Analogs, Vol. 20 (Ed.: S. Agrawal), Humana Press,
Totowa, NJ, 1993, chap. 19.
[9] F. Bergmann, W. Bannwarth, S. Tam, Tetrahedron Lett. 1995, 36,
6823 ± 6826.
Â
[10] V. Silnikov, G. Zuber, J.-P. Behr, R. Giege, V. Vlassov, Phosphorus
Sulfur Silicon 1996, 109 ± 110, 277 ± 280.
[11] R. Häner, J. Hall, Antisense Nucleic Acid Drug Dev. 1997, 7, 423 ± 430,
and references therein.
[12] K. Michaelis, M. Kalesse, Angew. Chem. 1999, 111, 2382 ± 2385;
Angew. Chem. Int. Ed. 1999, 38, 2243 ± 2245.
We now report a new strategy for polycyclic ether assembly,
the essence of which is illustrated in Scheme 1.[11] We
envisaged the construction of rings in a two-directional
H
H
R
H
R
H
H
R
H
H
R
O
O
O
O
Synthesis of Polycyclic Ethers by Two-
a
Directional Double Ring-Closing Metathesis**
O
O
H
H
H H
J. Stephen Clark* and Olivier Hamelin
b, c
H
O
H
H
Neurotoxic marine polycyclic ethers of the brevetoxin and
ciguatoxin family continue to be interesting synthetic targets
as a consequence of their architectural complexity and potent
biological activity.[1] They constitute a major challenge in
terms of medium-size ring construction as illustrated by CTX-
3C (Figure 1).[2] This ciguatoxin contains a total of 13 rings
and possesses a daunting combination of saturated and
unsaturated medium-sized rings. The size of the brevetoxins
and ciguatoxins, coupled with their structural and stereo-
chemical complexity, has made them irresistible synthetic
d
O
H
O
H
O
HO
OH
H
O
H
O
O
H
H
H
O
H
R
R
H
H
Scheme 1. Strategy for the synthesis of polycyclic ethers using two-
directional RCM reactions. a) Ring-closing metathesis; b) ring functional-
ization; c) side-chain functionalization; d) side-chain introduction.
manner by performing double RCM reactions of allylic ethers,
enol ethers, or alkynyl ethers, or any combination thereof.[12]
We expected that the reaction would be of general applic-
ability, permitting the preparation of tricyclic ethers with any
permutation of ring sizes. Elaboration of the tricyclic prod-
ucts, using our recently described methods,[7b] followed by
side-chain functionalization would then provide appropriately
functionalized systems for iteration of the two-directional
ring-closure sequence.
The precursors were prepared as shown in Scheme 2.
Treatment of commercially available tri-O-acetyl-d-glucal
(3) with sodium methoxide afforded d-glucal, which was
converted into 4,6-O-di-(tert-butyl)silanediyl-d-glucal accord-
[*] Dr. J. S. Clark, Dr. O. Hamelin
School of Chemistry
University of Nottingham
University Park, Nottingham NG7 2RD (UK)
Fax: (‡44)115-9513564
E-mail: j.s.clark@nottingham.ac.uk
[**] The European Commission is gratefully acknowledged for the award
of a Marie-Curie Fellowship to O.H. as part of the TMR Programme.
We thank Dr. A. J. Blake and Dr. P. Cooke for performing the X-ray
crystallographic analysis of compound 14.
Supporting information for this article is available on the WWW under
http://www.wiley-vch.de/home/angewandte/ or from the author.
372
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
0570-0833/00/3902-0372 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2000, 39, No. 2