Highly selective DNA alkylation at the 5' side G of a 5'GG3' sequence by an aglycon model of pluramycin antibiotics through preferential intercalation into the GG step

Article abstract of DOI:10.1021/ja980801q

Altromycin B and kapurimycin A₃ are new members of the pluramycin family antibiotics that alkylate N7 of guanine (G) in duplex DNA at an epoxide subunit attached to the C2 position of the pyranone ring. While 5'AG3' selective alkylation by altr

Full text of DOI:10.1021/ja980801q

J. Am. Chem. Soc. 1998, 120, 11219-11225
11219
Highly Selective DNA Alkylation at the 5′ Side G of a 5′GG3′
Sequence by an Aglycon Model of Pluramycin Antibiotics through
Preferential Intercalation into the GG Step
Kazuhiko Nakatani,* Akimitsu Okamoto, Takahiro Matsuno, and Isao Saito*
Contribution from the Department of Synthetic Chemistry and Biological Chemistry,
Faculty of Engineering, Kyoto UniVersity, and CREST, Japan Science and Technology Corporation,
Kyoto 606-8501, Japan
ReceiVed March 10, 1998
Abstract: Altromycin B and kapurimycin A3 are new members of the pluramycin family antibiotics that alkylate
N7 of guanine (G) in duplex DNA at an epoxide subunit attached to the C2 position of the pyranone ring.
While 5′AG3′ selective alkylation by altromycin B was accounted for by selective binding to the sequence, it
remained uncertain why kapurimycin A3, which is structurally similar to an aglycon part of altromycin B but
has an epoxide subunit with an opposite absolute configuration, selectively alkylates 5′ G of the 5′GG3′ sequence.
To clarify the molecular basis for the sequence-selective G alkylation by kapurimycin A3, we have examined
the DNA cleavage sequence selectivity of an enantiomeric pair of an aglycon model of pluramycin antibiotics
and calculated total energy change upon intercalation of the model into a specific sequence by means of the
DFT-HF hybrid method at the B3LYP/6-31G(d) level. Only an enantiomer with the same epoxide absolute
configuration as that of kapurimycin A3 showed a remarkable G alkylation reactivity with a high selectivity
for 5′ G of the 5′GG3′ sequence. Molecular mechanics calculations suggested that the intercalation of the
aglycon model in an orientation perpendicular to neighboring base pairs is essential for the effective G alkylation
at 5′ G of GN sequences, and the efficiency for G alkylation is not dependent upon the sequence being
intercalated. DFT-HF hybrid calculations suggested that the intercalation of the model is energetically most
favorable into GG step. These results suggested that highly 5′ G selective alkylation of the GG sequence by
kapurimycin A3 arises from a selective intercalation into the GG step. Stacking interaction with both 5′ and
3′ side Gs is a basis for the stabilization of the intercalated complex.
Introduction
Altromycin B (1)1,2 and kapurimycin A3 (2)3,4 are new
members of the pluramycin family antibiotics5 and alkylate N7
of guanine (G) in duplex DNA at an epoxide subunit attached
to the C2 position of the pyranone ring.6-10 NMR studies on
the DNA adduct of 1 have shown that the antibiotic intercalates
into the DNA duplex at the 5′ side of the alkylated G in an
orientation perpendicular to neighboring base pairs.11,12 The
* Corresponding author: (phone) (+81)-75-753-5656; (fax) (+81)-75-
753-5676; (e-mail) saito@sbchem.kyoto-u.ac.jp.
(1) Jackson, M.; Karwowski, J. P.; Theriault, R. J.; Hardy, D. J.; Swanson,
S. J.; Barlow, G. J.; Tillis, P. M.; McAlpine, J. B. J. Antibiot. 1990, 43,
223-228.
(2) Brill, G.; McAlpine, J. B.; Whittern, D. N.; Buko, A. M. J. Antibiot.
1990, 43, 229-237.
(3) Hara, M.; Mokudai, T.; Kobayashi, E.; Gomi, K.; Nakano, H. J.
Antibiot. 1990, 43, 1513-1518.
(4) Yoshida, M.; Hara, M.; Saitoh, Y.; Sano, H. J. Antibiot. 1990, 43,
1519-1523.
observed 5′AG*3′ selectivity (* denotes the alkylated site) for
the G alkylation by 1 was accounted for by a preferential binding
to this sequence through a DNA-sugar interaction.8,11 In
contrast, preferential alkylation of 5′ G of the 5′GG3′ sequence
was observed with 2,10,13 which is structurally similar to an
aglycon part of 1 but has an alkenyl epoxide subunit with an
opposite absolute configuration (14S, 16S).14 Since there is no
(5) Se´quin, U. Fortschr. Chem. Org. Naturst. 1986, 50, 57-122.
(6) Sun. D.; Hansen, M.; Clement, J. J.; Hurely, L. H. Biochemistry 1993,
32, 8068-8074.
(7) Hansen, M.; Yun, S.; Hurley, L. H. Chem. Biol. 1995, 2, 229-240.
(8) Hansen, M. R.; Hurley, L. H. Acc. Chem. Res. 1996, 29, 249-258.
(9) Hara, M.; Yoshida, M.; Nakano, H. Biochemistry 1990, 29, 10449-
10455.
(10) Chan, K. L.; Sugiyama, H.; Saito, I. Tetrahedron Lett. 1991, 52,
7719-7722.
(11) Hansen, M.; Hurley, L. H. J. Am. Chem. Soc. 1995, 117, 2421-
2429.
(13) Nakatani, K.; Okamoto, A.; Saito, I. Angew. Chem. Int. Ed. Engl.
1997, 36, 2794-2797.
(14) Uosaki, Y.; Saito, H. 69th annual meeting of the Chemical Society
of Japan, Kyoto, 1995; Abstr. p 1013.
(12) Sun, D.; Hansen, M.; Hurley, L. H. J. Am. Chem. Soc. 1995, 117,
2430-2440.
10.1021/ja980801q CCC: $15.00 © 1998 American Chemical Society
Published on Web 10/21/1998

11220 J. Am. Chem. Soc., Vol. 120, No. 44, 1998
Nakatani et al.
Scheme 1
Figure 1. Cleavage of 43-mer ODN via G alkylation by (a) 13S-3
and (b) 13R-3. 32P 5′-end-labeled ODN was annealed with a comple-
mentary strand and incubated with drug (50 µM) in Tris-HCl (pH 7.6)
at 37 °C for the period indicated. ODN was recovered by ethanol
precipitation, heated at 90 °C for 30 min with 10% piperidine, and
analyzed on 15% denatured PAGE. Lanes: 1, control; 2, 1 h; 3, 3 h;
4, 5 h; 5, 7 h; 6, 10 h; 7, 24 h; G + A, Maxam-Gilbert G + A lane.
A partial sequence of ODN is shown in the left.
Wittig olefination to produce 9 (Z:E ) ∼10:1), which was
desilylated to give alkyne 10. Lithiated 10 was coupled with
1-(tert-butyldimethylsilyloxy)-2-anthraldehyde (11) to give ad-
duct 12. Oxidation followed by pyranone ring cyclization15,16
efficiently produced tetracyclic compound 13, which was finally
transformed after deprotection to 13S-3 by intramolecular
Mitsunobu reaction.
Scheme 2
Analysis of G Alkylation Sequence Selectivity for Aglycon
Models. Sequence selectivities of 13R-3 and 13S-3 were
investigated using a 43-mer oligodeoxynucleotide (ODN),
d(TTTTTGTTTGTTAGTTCGTTTGCTTGGTTGATTGTTTG-
TTTTT), having all seven possible G-containing sequences
(italic). The 32P 5′-end-labeled ODN was annealed with a
complementary strand, and the duplex was incubated with 13R-3
and 13S-3. Alkylated G sites were detected as cleavage bands
after hot piperidine treatment by means of denatured polyacryl-
amide gel electrophoresis (PAGE). As is clear from Figure 1,
13R-3 having an epoxide group with the same absolute
configuration as 1 showed only weak and equal G bands (Figure
1b, lanes 4-7). In sharp contrast, strong and highly selective
G bands were observed with 13S-3 preferentially at 5′ G of the
5′GG3′ sequence and to a lesser extent at the GA sequence
(Figure 1a, lanes 4-7). The alkylation susceptibility of GN
sequences (N ) A, C, G, T) decreased in the order of 5′GG3′
> GA . GT > GC, and the normalized relative G band
intensities (IG*N) for G*G, G*A, G*T, and G*C as obtained by
densitometric analysis of lane 4 (Figure 1a) were 64.3, 26.6,
5.7, and 3.4, respectively. A large difference in both reactivity
and sequence selectivity between these enantiomers is particu-
larly noteworthy.17 Efficient and highly sequence selective
alkylation at 5′ G of the 5′GG3′ by 13S-3 possessing the same
epoxide absolute configuration as 2 provided strong evidence
that such 5′ G selective alkylation proceeds via a DNA-drug
complex which can be produced only from a 13S isomer.
Simulation of Intercalated Complex of 13S-3 into DNA.
The 5′G*G3′ alkylation selectivity observed for 2 may arise
apparent structural determinant for 2 to bind selectively to the
5′GG3′ sequence, the molecular basis for the 5′G*G3′ selectivity
has remained uncertain. To clarify this point, we have carried
out comparative G alkylation experiments using an enantiomeric
pair of a simplified aglycon model of pluramycins and calculated
the total energy change upon intercalation of this model into a
specific DNA sequence by means of the DFT-HF hybrid method
at the B3LYP/6-31G(d) level. On the basis of these theoretical
and experimental results, it was demonstrated that the highly
GG selective alkylation observed for 2 is derived from an
intrinsic property of the GG step in B-form duplex DNA, which
stabilizes the intercalated complex much more strongly than any
other G-containing sequences do by stacking interaction with
both 5′ and 3′ side Gs.
Results and Discussion
Synthesis of Aglycon Models. To assess the effect of
structure on the sequence selectivity, we have investigated G
alkylation by aglycon models 13R-3 and 13S-3 bearing an
alkenyl epoxide subunit of absolute configurations exactly
corresponded to 1 and 2, respectively. Both 13R-3 and 13S-3
isomers were synthesized via a synthetic route to that similar
to that reported earlier (Scheme 1).13,15,16 Since the enantiomeric
purity of 5 obtained by asymmetric dihydroxylation of 4 with
AD-mix-â was not high (about 80% ee), alcohol 8 was
recrystallized to obtain enantiomerically pure (4R,5R)-8 ([R]25
D
) +19.34, c ) 0.610, MeOH). In the synthesis of the
enantiomer of 8 (ent-8), HPLC separation of a diastereomeric
mixture of (S)-MTPA ester 15 followed by reductive deacylation
(17) For recent examples of different DNA alkylation properties of drug
enantiomers, see: (a) Boger, D. L.; McKie, J. A.; Nishi, T.; Ogiku, T. J.
Am. Chem. Soc. 1997, 119, 311-325. (b) Patel, V. F.; Andis, S. L.; Enkema,
J. K.; Johnson, D. A.; Kennedy, J. H.; Mohamadi, F.; Schultz, R. M.; Soose,
D. J.; Spees, M. M. J. Org. Chem. 1997, 62, 8868-8874. (c) Borger, D.
L.; Yun, W. J. Am. Chem. Soc. 1994, 116, 7996-8006. (d) Boger, D. L.;
Johnson, D. S.; Yun, W. J. Am. Chem. Soc. 1994, 116, 1635-1656. (e)
Kigoshi, H.; Imamura, Y.; Mizuta, K.; Niwa, H.; Yamada, K. J. Am. Chem.
Soc. 1993, 115, 3056-3065. (f) Boger, D. L.; Yun, W.; Terashima, S.
Fukuda, Y.; Nakatani, K.; Kitos, P.; Jin, Q. Bioorg. Med. Chem. Lett. 1992,
2, 759-765.
provided enantiomerically pure ent-8 ([R]25 ) -19.17, c )
D
0.120, MeOH) (Scheme 2). Oxidation of 8 was followed by
(15) Nakatani, K.; Okamoto, A.; Yamanuki, M.; Saito, I. J. Org. Chem.
1994, 59, 4360-4361.
(16) Nakatani, K.; Okamoto, A.; Saito, I. Tetrahedron 1996, 52, 9427-
9446.

DNA Alkylation at the 5′ Side G of a 5′GG3′ Sequence
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11221
thermal bath time constant of 0.2 ps, simulation time step of
0.75 fs, and simulation length of 100 ps. These simulations
indicated that the intercalation of 13S-3 in an orientation
perpendicular to neighboring base pairs is essential to steer the
epoxide subunit proximal to the N7 of 5′ G with an appropriate
alignment for the SN2 epoxide ring-opening reaction (Figure
3). The average nonbonding distance between N7 of 5′ G and
the epoxy carbon (C15) and the angle for N7 of G, C15, and the
epoxy oxygen (O3) are shown in Table 1 and are virtually the
same for all complexes. These results suggested that the
efficiency for G alkylation is insensitive to the sequence to be
intercalated.19 On the basis of these arguments, it was presumed
that (i) the crucial step for 5′G*G3′ selectivity is a noncovalent
binding step and (ii) the sequence selectivity for the G alkylation
directly reflects the susceptibility for the intercalative binding
of 13S-3 to the GN/MC step.
Figure 2. Unwinding of supercoiled pBR322 DNA by benzo-, naptho-,
and anthrapyranone derivatives 16-18, respectively. Supercoiled
pBR322 DNA (250 ng) (lane 1) was first treated with topoisomerase
l (topo l) (4 units) for 30 min at 37 °C in a buffer (10 mM Tris-HCl,
pH 7.9, 1 mM EDTA, 150 mM NaCl, 0.1 mM spermidine, 5% glycerol,
0.1% BSA) and further incubated for 30 min in the absence (lane 2)
and presence of 16-18 (100 µM) (lanes 4, 6, and 8, respectively). In
lanes 3, 5, and 7, DNAs were incubated with 16-18, respectively, in
the absence of topo l. The resulting DNAs were analyzed by
electrophoresis on 1% native agarose gel at 1.3 v/cm for 18 h. Lanes:
1, supercoiled DNA; 2, DNA treated with topo l; 3, 16; 4, 16 plus
topo l; 5, 17; 6, 17 plus topo l; 7, 18; 8, 18 plus topo l.
from a noncovalent binding step and/or a covalent alkylating
step.8 Previous structure-reactivity studies on truncated models
of 2 showed that G alkylation reactivity increased with
increasing ability for their intercalative binding.13 DNA-
unwinding assays with benzo-, naphtho-, and anthrapyranone
derivatives 16, 17, and 18, all of which lack an epoxide subunit
Energy Calculation of GN and GN‚13S-3 Complex. To
gain further insight into the relative stability of the complex
formed by intercalation of 13S-3 into the GN/MC step (GN‚
13S-3), total energy change upon intercalation was calculated
by the DFT-HF hybrid method at the B3LYP/6-31G(d) level
using the Gaussian 94 program.20,21 Geometries of GN‚13S-3
and the two-base pair unit of GN/MC (GN) used for energy
calculations were prepared as follows: for the GN‚13S-3
complex, 10 structures were sampled every 10 ps from MD
simulation of the complex of 13S-3 intercalated into the GN/
MC step of the duplex 6-mer d(ATGNTA/TAMCAT). The
structures were energy minimized and simplified for quantum
calculations by removing two bases from both the 5′ and 3′
ends of the duplex 6-mer and the sugar phosphate backbone. A
methyl group was attached to N9 and N1 of the purine and
pyrimidine bases, respectively, instead of deoxyribose, giving
the GN‚13S-3 structure. The structure of GN was made from
the energy-minimized duplex by a similar procedure as described
above. One of the 10 structures for each complex of GG‚13S-
3, GA‚13S-3, GT‚13S-3, and GC‚13S-3 used for energy calcula-
tion is shown in Figure 4. These coordinates of GN, 13S-3,
and GN‚13S-3 were transformed into Gaussian input data by
using Spartan (version 5.0). Single-point energy calculation of
each GN, 13S-3, and GN‚13S-3 complex was carried out at the
for DNA alkylation, were examined. The DNA-unwinding
assay indicated that 18 intercalates into DNA much more
strongly than 16 and 17 (Figure 2).
Molecular mechanics (MM) calculations of the intercalated
complex of 13S-3 into the GN/MC step, where M is a
complementary base for N of duplex 6-mer d(ATGNTA/
TAMCAT) were carried out by MacroModel (version 6.0) using
the Amber* force field with GB/SA solvation treatment of water
(distant-dependent dielectric electrostatic, cutoff for van der
Waals, electrostatic, and hydrogen bonding were 8, 20, and 4
Å, respectively).18 13S-3 was manually inserted into the GN/
MC step of B-form duplex d(ATGNTA/TAMCAT) and the
complexes were energy minimized to give initial geometries
for molecular dynamics (MD) simulation. Stochastic dynamics
simulation was carried out under the following conditions:
initial temperature of 300 K, simulation temperature of 300 K,
B3LYP/6-31G(d) level to give their total energies EGN, E13S-3
,
and EGN‚13S-3, respectively. EGN‚13S-3 was obtained as an average
energy of 10 independent structures for each GN‚13S-3 (Table
2). The total energy change upon intercalation of 13S-3 into
the GN/MC step (∆EGN) and the relative energy difference of
intercalation between the GN/MC and GG/CC steps (∆∆EGN-
GG) were obtained by eqs 1 and 2, respectively.
(18) Still, W. C.; Tempczyk, A.; Hawley, R. C.; Hendrickson, T. J. Am.
Chem. Soc. 1990, 112, 6127-6129.
(19) Average nonbonding distance between C15 of 13R-3 and 3′ side G
N7 in GG‚13R-3 complex obtained by MD simulation was 5.08 Å (0.52).
(20) Energy calculations at the B3LYP/6-31G(d) level with Gaussian
94 were carried out on a VPP300/16 supercomputer at the JST supercom-
puter complex.
(21) Gaussian 94, Revision D.4. Frisch, M. J.; Trucks, G. W.; Schlegel,
H. B.; Gill, P. M. W.; Johnson, B. G.; Robb, M. A.; Cheeseman, J. R.;
Keith, T.; Petersson, G. A.; Montgomery, J. A.; Raghavachari, K.;
Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J. V.; Foresman, J. B.;
Cioslowski, J.; Stefanov, B. B.; Nanayakkara, A.; Challacombe, M. Peng,
C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.; Andres, J. L.; Replogle, E.
S.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Binkley, J. S.; Defrees, D. J.;
Baker, J.; Stewart, J. P.; Head-Gordon, M.; Gonzalez, C.; Pople, J. A.
Gaussian, Inc., Pittsburgh, PA, 1995.
∆EGN ) EGN‚13S-3 - EGN - E13S-3
∆∆EGN-GG ) ∆EGN - ∆EGG
(1)
(2)
Relative Intercalation Susceptibility into the GN/MC Step.
Our calculations pointed out that the intercalative binding of
13S-3 was energetically most favorable to the GG/CC step
(Table 3). Τhe order of GG > GA > GT > GC for the stability
of the intercalated complex (i.e., ∆EGN) is consistent with

11222 J. Am. Chem. Soc., Vol. 120, No. 44, 1998
Nakatani et al.
Figure 3. Energy-minimized structure of complex of d(ATGGTA/TACCAT) with 13S-3, in which 13S-3 intercalated into the GG step. DNA and
13S-3 are shown in gray and black, respectively. Views from (a) a major groove and (b) a direction parallel to GC base pairs.
Figure 4. Representative structures of four 5′GN3′‚13S-3 complexes used for B3LYP/6-31G(d) energy calculations: (a) GG‚13S-3; (b) GA‚13S-3;
(c) GT‚13S-3; (d) GC‚13S-3.
Table 1. Average Nonbonding Distance of 5′G N7-C15 and the
Angle for N7-C15-O3 for the Complexes of d(ATGNTA/
TAMCAT) with 13S-3a
molecule), but there was a large difference in both calculated
∆E values and the experimentally observed G alkylation
susceptibility. On the basis of these results, it is strongly
suggested that highly selective 5′ G alkylation by 2 and 13S-3
is due to their preferential intercalation into the GG step.
Stacking interaction of the intercalator with both 5′ and 3′ side
Gs is the basis for such complex stabilization. It is well-known
that the G base is the highest in HOMO energy among
nucleobases.22 Therefore, it is conceivable that the HOMO-
LUMO interaction between the HOMO of 3′ G at the intercala-
tion site and the LUMO of the intercalator may play a significant
role for the highly selective intercalation into the GG step.
GN/MC
5′G N7-C15 (Å)b
N7-C15-O3 (deg)b
GG/CC
GA/TC
GT/AC
GC/GC
3.49 (0.42)
3.35 (0.30)
3.35 (0.33)
3.44 (0.36)
134.30 (8.7)
137.19 (8.0)
139.91 (8.5)
137.35 (7.9)
a The average nonbonding distance of 5′ G N7-C15 and the angle
of N7-C15-O3 were obtained for 100 structures being sampled during
MD simulations of the complexes of d(ATGNTA/TAMCAT) with 13S-
3. b The value in parentheses is the standard deviation.
experimentally observed G alkylation susceptibility. The rela-
tive alkylation reactivity of the Gs of the GN sequences as
compared with GG [ln(IG*G/IG*N)] obtained by the normalized
relative G band intensity of Figure 1 is well correlated with the
Experimental Section
1H NMR spectra were measured with Varian Mercury 400
(400 MHz) or JEOL JNM R-400 (400 MHz) spectrometers. 13
NMR spectra were measured with a Varian Mercury 400 (100
C
calculated energy difference between GN and GG (∆∆EGN-GG
)
(22) (a) Sugiyama, H.; Saito, I. J. Am. Chem. Soc. 1996, 118, 7063-
7068. (b) Prat, F.; Houk, K. N.; Foote, C. S. J. Am. Chem. Soc. 1998, 120,
845-846. (c) Sponer, J.; Leszczynski, J.; Hobza, P. J. Phys. Chem. 1996,
100, 5590-5596.
(Figure 5). It is particularly noteworthy that the component
molecules for GG‚13S-3 and for GC‚13S-3 complexes are the
same (i.e., two guanines and two cytocines, and one 13S-3

DNA Alkylation at the 5′ Side G of a 5′GG3′ Sequence
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11223
Table 2. Calculated Energies of GN‚13S-3 at the B3LYP/6-31G(d) Level
runa
EGA‚13S-3
EGC‚13S-3
EGG‚13S-3
EGT‚13S-3
b
b
b
b
1
11
21
31
41
51
61
71
81
91
-3129.111 356 05
-3129.111 989 00
-3129.110 893 34
-3129.109 069 01
-3129.109 432 35
-3129.112 687 10
-3129.109 790 01
-3129.109 719 15
-3129.109 674 12
-3129.109 312 98
-3145.148 432 00
-3145.147 885 82
-3145.146 527 62
-3145.145 856 46
-3145.146 633 84
-3145.148 324 42
-3145.150 900 95
-3145.152 326 85
-3145.150 200 44
-3145.150 255 16
-3145.160 633 82
-3145.160 620 22
-3145.160 854 67
-3145.160 184 99
-3145.161 372 33
-3145.159 341 72
-3145.161 530 37
-3145.162 733 91
-3145.163 784 51
-3145.160 784 61
-3129.105 785 79
-3129.105 751 21
-3129.109 294 66
-3129.110 436 63
-3129.110 801 80
-3129.111 097 29
-3129.110 155 55
-3129.109 262 37
-3129.107 808 03
-3129.106 036 33
av
-3129.110 392 31
-3145.148 734 36
-3145.161 184 12
-3129.108 642 97
a Run number corresponds to the structure number of GN‚13S-3 complex being sampled during MD simulations. b Energy is reported in hartree.
Table 3. Relative Energy Difference of the Intercalation of 13S-3
was added p-methoxybenzyl chloride (15.5 mL, 114.3 mmol)
at 0 °C, and the mixture was stirred for 30 min. The resulting
mixture was diluted with saturated NH4Cl and extracted with
ethyl acetate. The crude product was purified by column
chromatography on silica gel, eluting with 5% ethyl acetate in
hexane to give (E)-1-(4-methoxybenzyloxy)-3-methyl-2-penten-
4-yne (4) (20.6 g, 102 mmol, 98%) as a yellow oil: 1H NMR
(400 MHz, CDCl3) δ 7.24 (d, 2H, J ) 8.8 Hz), 6.86 (d, 2H, J
) 8.7 Hz), 6.07 (m, 1H), 4.42 (s, 2H), 4.05 (dt, 2H, J ) 0.7,
6.6 Hz), 3.79 (s, 3H), 2.81 (s, 1H), 1.78 (s, 1H); IR (CHCl3)
3304, 3021, 3009, 1612, 1513, 1249, 1217, 1173, 1035, 773,
769 cm-1. A mixture of AD-mix-â (12.0 g) and methane-
sulfonamide (0.82 g, 8.62 mmol) in tert-butyl alcohol (40 mL)
and H2O (40 mL) was stirred at 0 °C for 30 min. To the mixture
was added 4 (1.73 g, 8.54 mmol) at 0 °C, and the whole mixture
was stirred for 2 days. The reaction was quenched by adding
Na2SO3 (15 g) followed by stirring for 2 h. The resulting
mixture was extracted with ethyl acetate, and the combined
organic layer was washed with 1 N NaOH. The crude product
was purified by column chromatography on silica gel, eluting
with 20-100% ethyl acetate in hexane to give 5 (1.63 g, 76%)
as a colorless oil: 1H NMR (400 MHz, CDCl3) δ 7.24 (d, 2H,
J ) 8.7 Hz), 6.86 (d, 2H, J ) 8.7 Hz), 4.49 and 4.48 (d × 2,
each 1H, J ) 11.4 Hz), 3.79 (s, 3H), 3.77 (m, 1H), 3.72 (dd,
1H, J ) 4.7, 9.2 Hz), 3.59 (dd, 1H, J ) 6.3, 9.2 Hz), 3.26 (s,
1H), 2.53 (d, 1H, J ) 5.0 Hz), 2.46 (s, 1H), 1.47 (s, 3H); 13C
NMR (100 MHz, CDCl3) δ 159.06, 129.28, 129.26, 113.69,
85.28, 74.71, 73.16, 72.73, 70.56, 69.77, 55.20, 25.41; IR
(CHCl3) 3562, 3306, 3011, 2937, 1514, 1250, 1174, 1087, 1036,
734 cm-1; MS (EI) m/z (%) 250 (M+) (21), 137 (100), 121
(100); HRMS (EI) calcd for C14H8O4 (M+), 250.1204; found,
250.1201.
into the GN/MC and GG/CC Stepsa
d,e
f
∆EGN
(kcal/mol)
∆∆EGN-GG
(kcal/mol)
b
b,c
N
EGN
EGN‚13S-3
G
A
T
-2032.26467 -3145.16118 -5.53 (0.80)
0.00
7.11
10.79
16.43
-2016.22521 -3129.11039
-2016.22932 -3129.10864
-2032.27840 -3145.14873
1.58 (0.79)
5.26 (1.34)
10.89 (1.33)
C
a Energy calculations of GN, 13S-3, and GN‚13S-3 complex were
carried out at the DFT-HF hybrid B3LYP/6-31G(d) level. b Energies
(E) are reported in hartree. c An average energy of GN‚13S-3 was
obtained from Table 2. d ∆EGN ) EGN‚13S-3 - EGN - E13S-3, where
E13S-3 was -1112.887 70 hartree. e The value in parentheses is the
standard deviation. f ∆∆EGN-GG ) ∆EGN - ∆EGG
.
Figure 5. Calculated ∆∆EGN-GG vs relative alkylation reactivity [ln-
(lG*G/lG*N)] experimentally obtained. Correlation coefficient r ) 0.966.
MHz) spectrometer. Coupling constants (J values) are repre-
sented in hertz. The chemical shifts are expressed in ppm
downfield from tetramethylsilane, using residual solvent as an
internal standard. IR spectra were recorded on a Jasco FT-IR-
5M spectrophotometer. Mass spectra were recorded on a JEOL
JMS DX-300 or a JEOL JMS SX-102A spectrometer. Sequence
gel electrophoresis was carried out on a Gibco BRL model S2
apparatus. Densitometric analysis of the gel was carried out
using a Bio-Rad GS-700 imaging densitometer with analytical
software Molecular Analyst (version 2.1). Wakogel C-200 was
used for silica gel flash chromatography. Precoated TLC plates
Merck silica gel 60 F254 were used for monitoring reactions and
also for preparative TLC. Gel permeation chromatography
(GPC) was carried out on JAI LC-908 using a polystyrene
column. Ether and tetrahydrofuran (THF) were distilled under
N2 from sodium/benzophenone ketyl prior to use. All enzymes
used in these studies were from commercial sources. [γ-32P]-
ATP (6000 Ci/mmol) was obtained from Amersham.
(4R,5R)-4-Ethynyl-5-(4-methoxybenzyloxymethyl)-2,2,4-
trimethyl-1,3,-dioxolane (6). To a solution of 5 (1.97 g, 7.87
mmol), 2,2-dimethoxypropane (4.9 mL, 39.9 mmol), and
anhydrous CuSO4 (6.29 g, 39.4 mmol) in acetone (50 mL) was
added dl-camphorsulfonic acid (92.4 mg, 0.40 mmol) at 0 °C,
and the mixture was stirred for 3 h. The resulting mixture was
filtered, diluted with saturated NaHCO3, concentrated in vacuo
to remove acetone, and extracted with ethyl acetate. The crude
product was purified by column chromatography on silica gel,
eluting with 15% ethyl acetate in hexane to give 6 (2.17 g, 95%)
as a colorless oil: 1H NMR (400 MHz, CDCl3) δ 7.26 (d, 2H,
J ) 8.8 Hz), 6.86 (d, 2H, J ) 8.7 Hz), 4.58 (d, 1H, J ) 11.7
Hz), 4.46 (d, 1 H, J ) 11.7 Hz), 4.42 (dd, 1H, J ) 5.0, 6.5
Hz), 3.79 (s, 3H), 3.555 (d, 1H, J ) 5.0 Hz), 3.551 (d, 1H, J )
6.4 Hz), 2.48 (s, 1H), 1.46 (d, 3H, J ) 0.5 Hz), 1.42 (d, 3H, J
) 0.5 Hz), 1.34 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 158.92,
129,50, 129,10, 113.54, 109.34, 85.17, 81.44, 74.26, 73.02,
(2R,3R)-1-(4-Methoxybenzyloxy)-3-methyl-4-pentyne-2,3-
diol (5). To a suspension of sodium hydride (60%, 4.50 g, 113
mmol) in N,N-dimethylformamide (10 mL) was added (E)-3-
methylpent-2-en-4-yn-1-ol (10.01 g, 104 mmol) at 0 °C, and
the reaction mixture was stirred for 10 min. To this mixture

11224 J. Am. Chem. Soc., Vol. 120, No. 44, 1998
Nakatani et al.
72.29, 67.79, 55.10, 28.15, 25.70, 23.18; IR (CHCl3) 3020, 2993,
1513, 1375, 1249, 1088, 780 cm-1; MS (EI) m/z (%) 290 (M+)
(13), 275 (47), 232 (81), 137 (97), 121 (100); HRMS (EI) calcd
for C17H22O4 (M+), 290.1518; found, 290.1524.
the mixture was stirred at -78 °C for 15 min and at 0 °C for
15 min. The resulting mixture was diluted with methanol (10
mL), stirred at ambient temperature for 10 min, further diluted
with 5 mL each of ethyl acetate and saturated Rochelle salt,
and stirred at ambient temperature for 30 min. The mixture
was extracted with ethyl acetate. The crude product was purified
by column chromatography on silica gel, eluting with 30% ethyl
acetate in hexane to give optically pure (4S,5S)-8 (41.0 mg,
(4R,5R)-5-(4-Methoxybenzyloxymethyl)-2,2,4-trimethyl-4-
(2-trimethylsilylethynyl)-1,3-dioxolane (7). To a solution of
6 (1.63 g, 5.62 mmol) in THF (20 mL) was added lithium
hexamethyldisilazide (LHMDS) (1.0 M solution in THF, 6.2
mL, 6.20 mmol) at -78 °C, and the mixture was stirred for 10
min. To the mixture was added trimethylchlorosilane (0.78 mL,
6.16 mmol) at -78 °C, and the reaction mixture was warmed
to 0 °C and stirred for another 20 min. The resulting mixture
was diluted with saturated NH4Cl and extracted with ethyl
acetate. The crude product was purified by column chroma-
tography on silica gel, eluting with 15% ethyl acetate in hexane
to give 7 (1.79 g, 88%) as a colorless oil: 1H NMR (400 MHz,
CDCl3) δ 7.27 (d, 2H, J ) 8.8 Hz), 6.86 (d, 2H, J ) 8.7 Hz),
4.58 (d, 1H, J ) 11.6 Hz), 4.46 (d, 1H, J ) 11.6 Hz), 4.39 (dd,
1H, J ) 4.4, 7.0 Hz), 3.79 (s, 3H), 3.57 (dd, 1H, J ) 4.4, 10.3
Hz), 3.52 (dd, 1H, J ) 7.0, 10.3 Hz), 1.46 (s, 3H), 1.41 (s,
3H), 1.32 (s, 3H), 0.13 (s, 9H); 13C NMR (100 MHz, CDCl3)
δ 159,00, 129.75, 129.14, 113.65, 109.37, 106.72, 88.55, 81.76,
74.84, 73.05, 68.13, 55.25, 28.33, 25.89, 23.35, -0.04; IR
(CHCl3) 3020, 1513, 1250, 1180, 1087, 848 cm-1; MS (FAB)
m/z (%) 362 (M+) (5), 347 [(M - Me)+] (10), 303 (17), 121
(100); HRMS (FAB) calcd for C20H30O4Si (M+), 362.1913;
found, 362.1896.
25
0.19 mmol): [R]D ) -19.17 (c 0.120, MeOH).
(4R,5R,Z)-5-Propenyl-2,2,4-trimethyl-4-(trimethylsilyleth-
ynyl)-1,3,-dioxolane (9). To a solution of 8 (0.81 g, 3.34 mmol)
in dichloromethane (30 mL) was added Dess-Martin periodi-
nane (1.7 g, 4.01 mmol) at ambient temperature, and the mixture
was stirred at that temperature for 1 h. The resulting mixture
was diluted with saturated Na2S2O3 and saturated NaHCO3 and
then extracted with ethyl acetate. The crude product was
purified by column chromatography on silica gel, eluting with
25% ethyl acetate in hexane to give (4R,5S)-5-formyl-2,2,4-
trimethyl-4-(2′-trimethylsilylethynyl)-1,3,-dioxolane (0.76 g,
94%) as a colorless oil: 1H NMR (CDCl3, 400 MHz) δ 9.61
(d, 2 H, J ) 2.2 Hz), 4.54 (d, 2 H, J ) 2.4 Hz), 1.56 (d, 3 H,
J ) 0.5 Hz), 1.53 (d, 3 H, J ) 0.5 Hz), 1.43 (s, 3 H), 0.16 (s,
9 H); MS (EI) m/e (%) 235[(M - Me)+], (10), 153 (55), 100
(74), 85 (100). To a solution of aldehyde (0.67 g, 2.80 mmol)
in toluene (20 mL) was added a toluene solution of ethyliden-
triphenylphosphorane, which was prepared from ethyltri-
phenylphosphonium bromide (1.86 g, 5.01 mmol) and potassium
hexamethyldisilazide (10.0 mL, 0.5 M in toluene, 5.0 mmol)
in toluene (10 mL) under refluxing for 2 h, at -78 °C, and the
mixture was stirred 1 h at that temperature and then for another
1 h at 0 °C. The resulting mixture was diluted with saturated
NH4Cl and extracted with ethyl acetate. The crude product was
purified by column chromatography on silica gel, eluting with
30% toluene in hexane to give 9 (0.55 g, 78%, Z/E ) 10:1
mixture) as a colorless oil. Pure Z-isomer for analysis could
(4R,5R)-5-(Hydroxymethyl)-2,2,4-trimethyl-4-(2-trimeth-
ylsilylethynyl)-1,3,-dioxolane (8). To a solution of 7 (0.31 g,
0.85 mmol) in dichloromethane (18 mL) and H2O (1 mL) was
added DDQ (0.23 g, 1.01 mmol) at 0 °C, and the mixture was
stirred at ambient temperature for 4 h. The resulting mixture
was diluted with saturated NaHCO3, and extracted with ethyl
acetate. The crude product was purified by column chroma-
tography on silica gel, eluting with 60% toluene in hexane to
give 8 (0.18 g, 99%) as colorless needles. Recrystallization
from 5% benzene in hexane at -30 °C provided optically pure
1
be isolated by recycling preparative HPLC: H NMR (400 MHz,
CDCl3) δ 5.82 (ddq, 1H, J ) 1.3, 7.0, 11.1 Hz), 5.40 (m, 1H),
5.05 (d, 1H, J ) 8.5 Hz), 1.78 (dd, 3H, J ) 1.8, 7.0 Hz), 1.47
(s, 3H), 1.43 (s, 3H), 1.30 (s, 3H), 0.13 (s, 9H); 13C NMR (100
MHz, CDCl3) δ 130.47, 124.65, 108.85, 106.99, 88.58, 78.96,
76.13, 28.32, 25.95, 24.20, 14.09, -0.09; IR (CHCl3) 3017,
1730, 1250, 723 cm-1; MS (EI) m/z (%) 237 [(M - Me)+] (11),
195 (21), 182 (28), 167 (38), 97 (100); HRMS (EI) calcd for
C8H21O2Si [(M - Me)+], 237.1311; found, 237.1322.
(4R,5R,Z)-4-Ethynyl-5-propenyl-2,2,4-trimethyl-1,3-diox-
olane (10). To a solution of 9 (105.1 mg, 0.42 mmol) in
methanol (1 mL) was added a catalytic amount of sodium
methoxide at 0 °C, and the mixture was stirred at ambient
temperature for 4 h. The resulting mixture was diluted with
saturated NH4Cl and extracted with ether. Concentration of the
organic layer afforded 10 (51.6 mg, 69%) as a colorless oil:
1H NMR (400 MHz, CDCl3) δ 5.84 (ddq, 1H, J ) 1.2, 7.0,
11.1 Hz), 5.40 (ddq, 1H, J ) 1.8, 8.6, 11.1 Hz), 5.08 (dd, 1H,
J ) 1.1, 8.6 Hz), 2.47 (s, 1H), 1.78 (dd, 3H, J ) 1.8, 7.0 Hz),
1.48 (s, 3H), 1.44 (s, 3H), 1.34 (s, 3H); 13C NMR (100 MHz,
CDCl3) δ 130.98, 124.27, 109.02, 85.47, 78.68, 75.66, 72.33,
28.32, 25.91, 24.23, 14.03; IR (CHCl3) 3017, 2932, 1721, 1291,
1222, 1077, 788 cm-1; MS (EI) m/z (%) 165 [(M - Me)+]
(100), 149 (13), 110 (80), 97 (60); HRMS (EI) calcd for
C10H13O2 [(M - Me)+], 165.0915; found, 165.0920.
25
1
8: mp 45 °C; [R]D ) +19.34 (c 0.610, MeOH); H NMR
(400 MHz, CDCl3) δ 4.31 (dd, 1H, J ) 3.9, 7.4 Hz), 3.73 (2H),
1.73 (m, 1H), 1.49 (s, 3H), 1.43 (s, 3H), 1.37 (s, 3H), 0.14 (s,
9H); 13C NMR (100 MHz, CDCl3) δ 109.48, 106.77, 88.86,
83.38, 74.72, 61.32, 28.39, 25.95, 23.43, -0.05; IR (CHCl3)
3020, 1217, 847, 754 cm-1; MS (EI) m/z (%) 227 [(M - Me)+]
(82), 209 (35), 182 (35), 167 (100), 155 (70), 124 (88); HRMS
(EI) calcd for C11H19O3Si [(M - Me)+], 227.1103; found,
227.1098. Anal. Calcd for C12H22O3Si: C, 59.46; H, 9.15.
Found: C, 59.20; H, 9.30.
To the solution of alcohol (4S,5S)-8 (196 mg, 0.92 mmol) in
dichloromethane (5 mL) were added (-)-(S)-MTPA acid (215
mg, 0.92 mmol), DCC (191 mg, 0.92 mmol), and DMAP (12.0
mg, 0.098 mmol) at ambient temperature, and the mixture was
stirred at ambient temperature for 1 h. The mixture was diluted
with ethyl acetate, filtered, and concentrated in vacuo. The
crude product was purified by column chromatography on silica
gel, eluting with toluene to give a diastereomeric mixture of
MTPA ester 15 (354 mg). The mixture was separated by HPLC
on Daicel Chiralcel OJ (10 × 250 mm, elution with 2% hexane
in 2-propanol at a flow rate of 1.0 mL/min). The major fraction
was collected and concentrated in vacuo to give pure 15: 1H
NMR (CDCl3, 400 MHz,) δ 7.55-7.53 (2 H), 7.42-7.37 (3
H), 4.47-4.37 (3 H), 3.55 (d, 3 H, J ) 1.1 Hz), 1.44 (s, 3 H),
1.40 (s, 3 H), 1.36 (s, 3 H), 0.13 (s, 3 H). To the solution of
15 (101 mg, 0.22 mmol) in dichloromethane (2 mL) was added
DIBAL (1.0 M in toluene, 0.5 mL, 0.5 mmol) at -78 °C, and
(4R,5R,Z)-4-[3-[1-(tert-Butyldimethylsilyloxy)-2-anthryl]-
3-hydroxy-1-propynyl]-5-(1-propenyl)-2,2,4-trimethyl-1,3-di-
oxorane (12). To a solution of 10 (99.2 mg, 0.55 mmol) and
anhydrous CeCl3 (411 mg, 1.7 mmol) in THF (5 mL) stirred at

DNA Alkylation at the 5′ Side G of a 5′GG3′ Sequence
J. Am. Chem. Soc., Vol. 120, No. 44, 1998 11225
ambient temperature for 10 min was added LHMDS (1.1 mL,
1.0 M in THF, 1.1 mmol) at -78 °C, and the mixture was stirred
at -78 °C for 15 min. A solution of 11 (187 mg, 0.55 mmol)
in THF (1 mL) was added to the mixture at -78 °C, and the
whole mixture was stirred at -78 °C for 15 min. The reaction
was quenched by adding saturated NH4Cl at -78 °C, and the
resulting mixture was extracted with ethyl acetate. The crude
product was purified by column chromatography on silica gel,
eluting with 35% toluene in hexane to give a mixture of
hexene (13S-3). To a solution of 14 (6.9 mg, 19.1 µmol) and
triphenylphosphine (7.0 mg, 26.7 µmol) in toluene (1 mL) was
added DEAD (6 µL, 37.9 µmol) at 0 °C, and the mixture was
stirred at ambient temperature for 24 h. The resulting mixture
was concentrated in vacuo and purified by column chromatog-
raphy on silica gel, eluting with 13% toluene in hexane to give
13S-3 (4.1 mg, 63%) as a white solid: 1H NMR (400 MHz,
CDCl3) δ 9.00 (s, 1H), 8.47 (s, 1H), 8.14 (m, 1H), 8.08-8.04
(2H), 7.88 (d, 1H, J ) 9.0 Hz), 7.63-7.56 (2H), 6.61 (s, 1H),
5.81 (ddd, 1H, J ) 1.1,7.1, 11.2 Hz), 5.18 (dq, 1H, J ) 1.7,
8.6 Hz), 3.95 (d, 1H, J ) 8.6 Hz), 1.96 (s, 3H), 1.85 (dd, 3H,
J ) 1.7, 7.1 Hz); IR (CHCl3) 3129, 1651, 1223, 909 cm-1; MS
(EI) m/e (%) 342 (M+) (100), 300 (33), 220 (76); HRMS (EI)
calcd for C23H18O3 (M+), 342.1256; found, 342.1245.
1
diastereoisomer of 12 (98.8 mg, 35%) as a yellow oil: H NMR
(400 MHz, CDCl3) δ 8.67 (s, 1H), 8.34 (s, 1H), 7.99, 7.96 (s ×
2, total 2H), 7.73-7.63 (2H), 7.50-7.43 (2H), 6.06-6.04 (1H),
5.83-5.75 (1H), 5.42-5.36 (1H), 5.07, 5.03 (dd × 2, total 1H,
J ) 1.1, 8.6 Hz,), 2.16-2.12 (1H), 1.70, 1.60 (dd × 2, total
3H, J ) 1.7, 7.0 Hz), 1.48-1.43 (6H), 1.37, 1.35 (s × 2, total
3H), 1.20, 1.19 (s × 2, total 9H), 0.24-0.22 (6H); IR (CHCl3)
3571, 3019, 1216 cm-1; MS (EI) m/e (%) 516 (M+) (2), 222
(70), 75 (100).
Cleavage of 32P 5′-End-Labeled Oligodeoxynucleotide by
13S-3 and 13R-3. Single-stranded 43-mer DNA oligomer
5′-d(TTTTTGTTTGTTAGTTCGTTTGCTTGGTTGATTGTTT-
GTTTTT)-3′ (400 pmol), purchased from Greiner Japan Co.
Ltd., was 5′-end-labeled by phosphorylation with [γ-32P]ATP
(4 µL) (Amersham, 370 MBq/µL) and T4 polynucleotide kinase
(4 µL) (Takara, 10 units/µL) using standard procedures. The
5′ end-labeled DNA was recovered by ethanol precipitation,
further purified by 15% nondenatured gel electrophoresis, and
isolated by a crush and soak method. The isolated DNA was
incubated with equimolar of the complementary strand in 100
µL of water at 90 °C for 5 min and cooled slowly to ambient
temperature for duplex formation. 13S-3 and 13R-3 (50 µL)
was incubated with 10 µL of calf thymus DNA and ∼1.0 ×
106 cpm 32P 5′-end-labeled ODN duplex in 20 mM Tris-HCl
buffer (100 µL, pH 7.6) at 37 °C. After a certain time interval,
an aliquot of the sample (10 µL) was separated from the reaction
mixture and precipitated with methanol. The recovered DNA
was dissolved in 100 µL of 10% (v/v) piperidine and heated at
90 °C for 30 min. The mixture was concentrated in vacuo and
resuspended in 10 µL of 80% formamide loading buffer (80%
formamide, 1 mM EDTA, 0.1% xylene cyanole, and 0.1%
bromophenol blue). The samples (1 µL) were loaded onto 15%
polyacrylamide and 7 M urea sequence gel and electrophoresed
at 1900 V for ∼2 h. The gel was dried, exposed to X-ray film
with intensifying sheet at -70 °C, and analyzed by densitometer.
Supercoiled DNA-Unwinding Assay. To a solution of
pBR322 plasmid DNA (250 ng, Nippon Gene) in topo I reaction
buffer (10 mM Tris-HCl, pH 7.9, 1 mM EDTA, 150 mM NaCl,
0.1 mM spermidine, 5% glycerol, 0.1% BSA) was added human
topoisomerase I (TopoGEN, 4 units). The reaction mixture was
incubated for 30 min at 37 °C. Drugs 16, 17, and 18 (100 µM)
were added, and incubation was continued for another 30 min
at 37 °C. The reaction was terminated by addition of SDS to
1%. After proteinase K was added to 50 µg/mL, the mixture
was digested for 20 min at 56 °C. Addition of 0.1 volume of
10× gel loading buffer was followed by chloroform extraction.
Different forms of DNA were separated at room temperature
on a 1% agarose gel. The gel was stained for 30 min with
ethidium bromide (0.5 µM/mL) and destained for 20 min in
water. It was placed on a UV transilluminator (313 nm) and
photographed with Polaroid 665 film.
(4S,5R,Z)-4-(4H-Anthra[1,2-b]pyran-2-yl)-5-(1-propenyl)-
2,2,4-trimethyl-1,3-dioxorane (13). To a solution of 12 (56.1
mg, 109 µmol) in dichloromethane (3 mL) was added manga-
nese(IV) oxide (50 mg), and the reaction mixture was stirred
for 3 h at ambient temperature. The mixture was diluted with
ethyl ether, filtered, and concentrated in vacuo. The crude
product was purified by column chromatography on silica gel,
eluting with 13% ethyl acetate in hexane to give (4R,5R,Z)-4-
[3-[1-(tert-butyldimethylsilyloxy)-2-anthryl]-3-oxo-1-propynyl]-
5-propenyl-2,2,4-trimethyl-1,3-dioxorane (38.7 mg, 75.2 µmol,
69%) as a yellow oil: MS (EI) m/e (%) 457 [(M - t-Bu)+]
(10), 368 (18), 294 (37), 149 (100); HRMS (EI) calcd for
C32H38O4Si [(M - t-Bu)+], 457.1836; found, 457.1840. This
ketone was used immediately for the next reaction. To a
solution of ketone (29.8 mg, 57.9 µmol) and 18-crown-6 (30.7
mg, 116 µmol) in N,N-dimethylformamide (3 mL) was added
a potassium fluoride (6.8 mg, 117 µmol) at 0 °C, and the mixture
was stirred at ambient temperature for 1 h. The resulting
mixture was diluted with saturated NH4Cl and extracted with
ethyl acetate. The crude product was purified by column
chromatography on silica gel, eluting with 13% ethyl acetate
in hexane to give 13 (17.3 mg, 43.2 µmol, 75%) as a yellow
oil: 1H NMR (400 MHz, CDCl3) δ 8.84 (s, 1H), 8.46 (s, 1H),
8.08-8.04 (3H), 7.87 (d, 1H, J ) 9.0 Hz), 7.63-7.56 (2H),
6.85 (s, 1H), 6.12-6.04 (m, 1H), 5.85-5.78 (m, 1H), 5.10 (d,
1H, J ) 10.0 Hz), 1.68 (s, 3H), 1.63 (s, 3H), 1.56 (s, 3H), 1.43
(dd, 3H, J ) 7.0, 1.8 Hz); IR (CHCl3) 3263, 2995, 1650, 1386,
1105 cm-1; MS (EI) m/e (%) 400 (M+) (59), 289 (100), 288
(75), 149 (43); HRMS (EI) calcd for C26H24O4 (M+), 400.1674;
found, 400.1658.
(2S,3R,4Z)-2-(4H-Anthra[1,2-b]pyran-2-yl)-4-hexene-2,3-
diol (14). To a solution of 13 (14.4 mg, 36.0 µmol) in THF (1
mL) and AcOH (1 mL) was added 0.2 M HCl (0.1 mL) at 0
°C, and the mixture was stirred at ambient temperature for 3
days. The resulting mixture was poured into saturated NaHCO3
and extracted with ethyl acetate. The crude product was purified
by column chromatography on silica gel, eluting with 50% ethyl
acetate in hexane to give 14 (11.8 mg, 32.7 µmol, 91%) as a
white solid: 1H NMR (400 MHz, CDCl3) δ 8.92 (s, 1H), 8.39
(s, 1H), 8.12-8.04 (2H), 7.90 (d, 1H, J ) 9.0 Hz), 7.72 (d,
1H, J ) 9.0 Hz), 7.64-7.57 (2H), 6.82 (s, 1H), 5.93-5.85 (m,
1H), 5.74-5.67 (m, 1H), 5.13 (d, 1H, J ) 9.3 Hz), 3.34 (br,
1H), 2.53 (br, 1H), 1.80 (dd, 3H, J ) 1.8, 7.0 Hz), 1.64 (s,
3H); MS (FAB) (NBA) m/e 361 [(M + H)+]; HRMS (FAB)
calcd for C23H21O4 [(M + H)+], 361.1440; found, 361.1446.
(2S,3R,4Z)-2-(4H-Anthra[1,2-b]pyran-2-yl)-2,3-epoxy-4-
Acknowledgment. We thank Prof. S. Rychnovsky (Univer-
sity of California, Irvine) for many helpful discussions. We are
grateful to the JST supercomputer complex for allocation of
supercomputer time. This research was in part supported by a
Grant-in-Aid for Priority Research from the Ministry of Educa-
tion, Science, Sports and Culture.
JA980801Q