Stereoselective synthesis of the atropisomers of myristinin B/C

Article abstract of DOI:10.1021/ol060511b

The first stereoselective synthesis of a potent DNA damaging agent, (-)-myristinin B/C, has been accomplished. This efficient synthesis allowed for unambiguous confirmation of the structure and absolute stereochemistry of the atropisomeric natural product

Full text of DOI:10.1021/ol060511b

ORGANIC
LETTERS
2006
Vol. 8, No. 9
1925-1927
Stereoselective Synthesis of the
Atropisomers of Myristinin B/C
David J. Maloney, Shengxi Chen, and Sidney M. Hecht*
Departments of Chemistry and Biology, UniVersity of Virginia,
CharlottesVille, Virginia 22904
sidhecht@Virginia.edu
Received March 1, 2006
ABSTRACT
The first stereoselective synthesis of a potent DNA damaging agent, (
allowed for unambiguous confirmation of the structure and absolute stereochemistry of the atropisomeric natural product. The antipode,
)-myristinin B/C, was also synthesized, providing ample material for biological evaluation of both enantiomers.
−)-myristinin B/C, has been accomplished. This efficient synthesis
(
+
Our laboratory has long been interested in the isolation and
biological evaluation of compounds which possess activity
as DNA damaging agents or polymerase â inhibitors.1 The
myristinins (Figure 1) were first isolated by Sawadajoon and
as potent DNA damaging agents and as DNA polymerase â
inhibitors.3,4 Additionally, they have demonstrated strong
potentiation of the cytotoxicity of bleomycin. Given the
intriguing biological activity of these compounds, it was of
interest to synthesize these compounds to gain access to
material sufficient for detailed biological evaluation. Initial
studies focused on the synthesis of myristinin A (1) as
preliminary biochemical assays revealed this to be the more
potent DNA damaging agent. Ultimately, these efforts
culminated in the first synthesis of 1, providing structural
confirmation and absolute stereochemistry determination.4
With the completion of the synthesis of the 2,4-trans isomer
1, we sought to gain synthetic access to the 2,4-cis isomer
myristinin B/C (2a,b) which exists naturally as an inseparable
mixture of atropisomers.2 In this account, we report the
successful synthesis of both enantiomers of 2a,b, permitting
the absolute stereochemistry of the natural product to be
Figure 1. Structures of the myristinins.
(2) Sawadjoon, S.; Kittakoop, P.; Kirtikara, K.; Vichai, V.; Tanticharoen,
M.; Thebtaranonth, Y. J. Org. Chem. 2002, 67, 5470. Compound 1 has
also been isolated from Horsfieldia amygdaline. See: Katuhiko, H.; Akio,
K.; Hirokazu, Y.; Akira, M.; Akira, I.; Akinori, S.; Shengji, P.; Yanhui, L.;
Chun, W. Patent WO 9208712, 1992.
(3) Deng, J.-Z.; Starck, S. R.; Li, S.; Hecht, S. M. J. Nat. Prod. 2005,
68, 1625.
co-workers from Myristica cinnamomea2 and more recently
isolated by our laboratory from Knema elegans.3 These
compounds exhibit impressive dual biochemical activity both
(1) (a) Hecht, S. M. J. Nat. Prod. 2000, 63, 158. (b) Hecht, S. M. Pharm.
Biol. 2003, 41, S68 and references therein.
(4) Maloney, D. J.; Deng, J.-Z.; Starck, S. R.; Gao, Z.; Hecht, S. M. J.
Am. Chem. Soc. 2005, 127, 4140.
10.1021/ol060511b CCC: $33.50
© 2006 American Chemical Society
Published on Web 04/01/2006

determined unambiguously. A key step involved a Lewis acid
promoted condensation,5 and the synthesis also provided an
opportunity to investigate the factors controlling this dias-
tereoselective reaction.
neighboring group participation effect from the acetate,
yielding the 2,4-trans coupled product in a stereoselective
manner (Figure 2A). Thus, for the synthesis of myristinin
Although the absolute stereochemistry of the 2,4-trans
isomer (1) was determined previously,4 the absolute config-
uration of naturally occurring 2a,b could not be inferred from
that study. Thus, the synthesis of both enantiomers seemed
prudent. The approach utilized (2R,3S)-flavan-3-ol 56 which
is accessible in six steps from the substituted 1,3-diphenyl-
propene derivative 34 shown in Scheme 1.7 The absolute
Scheme 1. Approach to Myristinin B/C
Figure 2. Approach of the nucleophile in the Lewis acid promoted
condensation reaction.
B/C, it seemed logical to anticipate that the 2,3-trans
configuration of the phenyl and OAc groups would favor
approach of the nucleophile from the bottom face in the
Lewis acid promoted condensation reaction5 (Figure 2B).
Nonetheless, it was anticipated that the steric interaction of
the incoming nucleophile from the bottom face with the C-2
phenyl group might result in reduced diastereoselectivity.
In fact, the coupling of 6 with the tri-O-benzyl-protected
acetophenone derivative 74 used in the synthesis of 1 gave
9, albeit as an unsatisfactory 1:1 mixture of cis/trans isomers
with respect to C-2 and C-4. Reasoning that the benzyl
protecting groups were too large, we changed these to a
smaller protecting group to limit this steric interaction.9 Thus,
the tri-O-methyl-protected acetophenone derivative 8 was
synthesized from commercially available 1,3,5-trimethoxy-
benzene via acid-catalyzed acylation with lauroyl chloride.
Gratifyingly, the diastereoselectivity of the condensation
reaction was improved to 4:1 cis/trans (100% overall yield);
however, attempts to improve this ratio using TiCl4, SnCl4,
BF3‚OEt2, or varying reaction conditions failed. Although
the selectivity was not as high as had been hoped, the 2,4-
cis product could be separated easily from the undesired 2,4-
trans isomer by column chromatography. Removal of the
C-3 hydroxyl group proceeded as planned (Scheme 3).
Deacetylation of 10 using K2CO3 in 1:1 THF-MeOH gave
alcohol 11 in 92% yield. Subsequent deoxygenation of the
C-3 hydroxyl group was carried out in a manner analogous
to that reported for the synthesis of 1 using phenyl chlo-
rothionoformate, DMAP in MeCN followed by treatment
with Bu3SnH, and catalytic AIBN to afford 12 in 72% yield
over two steps.10 Notably, formation of the thioester precursor
stereochemistry of 5 was confirmed using circular dichroism
as previously described by van Rensburg et al.8 Subsequent
acetylation of 5 followed by C-4 oxidation using DDQ and
ethylene glycol gave 6 in yields of 93% and 82%, respec-
tively (Scheme 2).5 In our approach to myristinin A (1), the
Scheme 2. Diastereoselective Coupling Reaction
C-3 hydroxyl group of 5 was inverted to facilitate the desired
(5) (a) Tu¨ckmantel, W.; Kozikowski, A. P.; Romanczyk, L. J., Jr. J. Am.
Chem. Soc. 1999, 121, 12073. (b) Saito, A.; Nakajima, N.; Tanaka, A.;
Ubukata, M. Tetrahedron 2002, 58, 7829. (c) Kozikowski, A. P.; Tu¨ck-
mantel, W.; Hu, Y. J. Org. Chem. 2001, 66, 1287.
(6) Notably, the enantiomer of 5 (i.e., having the 2S,3R configuration)
can be obtained by using AD-Mix â in the asymmetric dihydroxylation7
reaction of 3.
(7) Sharpless, K. B.; Amberg, W.; Beller, M.; Chen, H.; Hartung, J.;
Kawanami, Y.; Lu¨bben, D.; Manoury, E.; Ogino, Y.; Shibata, T.; Ukita, T.
J. Org. Chem. 1991, 56, 4585.
(8) van Rensburg, H.; Steynberg, P. J.; Burger, J. F. W.; van Heerden,
P. S.; Ferreira, D. J. Chem. Res., Synop. 1999, 450.
(9) Initially, the use of MOM ethers was planned as they are generally
more easily removed as compared to methyl ethers. However, all attempts
to protect the three phenolic OH groups led to significant C-alkylation of
the electron-rich aromatic ring.
1926
Org. Lett., Vol. 8, No. 9, 2006

stereochemistry of authentic (-)-myristinin B/C is 2S,4S and
exists as a 1.2:1 mixture of atropisomers myristinin B (2a)
and myristinin C (2b), respectively.13 To gain further insight
into the biological activity of these compounds, the enantio-
mer (2R,4R)-(+)-myristinin B/C was also synthesized in
comparable yields starting from the antipode of 5 (Scheme
1).14
Scheme 3. Completion of (-)-(2a,b) Synthesis
In summary, both enantiomers of myristinin B/C (2a,b)
were synthesized in a facile manner permitting unequivocal
determination of the absolute stereochemistry of (-)-
myristinin B/C as 2S,4S. Both enantiomers were identical
1
with the authentic sample as judged by comparisons of H
NMR, 13C NMR, HRMS, and mobility on C18 reversed-phase
HPLC.15 With the syntheses of both enantiomers of 2a,b
completed, a detailed biochemical and biological evaluation
of these interesting compounds is currently underway.
on the hindered alcohol required the use of rather forcing
conditions, whereas the radical deoxygenation was facile.
With the fully protected myristinin B/C (2a,b) in hand, a
method to remove all of the protecting groups in one step
was sought. BBr3 was chosen, given its ability to remove
methyl ethers and the known sensitivity of benzyl ethers to
this reagent. Although epimerization at the doubly benzylic
C-4 position was anticipated, the extent to which this would
happen was unclear. Initial attempts at complete deprotection
using minimal amounts of BBr3 (i.e., 1.1 equiv per protecting
group) led to incomplete deprotection and very complex
product mixtures. After several attempts at optimizing
reaction conditions, complete deprotection was finally ac-
complished using 15 equiv of BBr3 in CH2Cl2. As anticipated,
some epimerization at the C-4 position was observed,
yielding an approximately 4:1 mixture of cis/trans isomers
as judged by HPLC analysis. The isomeric compounds 1
and 2a,b have the same mobility on silica gel yet could be
separated by C18 reversed-phase HPLC (retention times: 22.5
min for 1 and 23.4 min for 2a,b). Interestingly, treatment of
12 with 40 equiv of BBr3 yielded approximately a 1:5
mixture of cis/trans isomers, essentially inverting the ratio
of the products.11 The fact that increasing the amount of BBr3
(which should favor the thermodynamic product) results in
the formation of more 1 argues that this species is the more
energetically stable isomer.12 Comparison of the synthetic
2a,b with the natural sample revealed that the absolute
Acknowledgment. This work was supported by NIH
Research Grant CA50771, awarded by the National Cancer
Institute.
Supporting Information Available: Experimental pro-
cedures and characterization data for all new compounds.
1H and 13C NMR for compounds 2a,b and 5-12 are also
available. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL060511B
(11) In addition to observing epimerization of 2a,b to 1, we also observed
the formation of what appears to be a conformational isomer of 1. This
compound has mass (by HRMS) and retention time on C18 reversed HPLC
identical to 1. However, the 1H NMR spectra of the two compounds are
quite different (see Supporting Information). The possibility that this species
is a constitutional isomer of 1 formed under the forcing conditions cannot
presently be excluded.
(12) Attempts to measure the relative energies of compounds 1 and 2a,b
using molecular modeling techniques led to inconsistent results.
(13) The ratio was determined by integration of H-4 and 2′-OH in the
1H spectrum of 2a,b as these protons possess different resonances due to
the restricted rotation about the substituents on C-4 and C-3′. Sawadajoon
and co-workers2 reported the same ratio and assigned the structures of
atropisomers 2a and 2b through NOESY spectral data by observing a cross-
peak with the chelated hydroxyl (2′-OH) of 2a with the pseudoaxial H-4.
In comparison, this correlation was absent in 2b.
(14) See Supporting Information for the optical rotations of intermediates
from the enantiomeric series.
(15) The optical rotation of synthetic 2a,b, [R]19D -48.3 (c 0.4, MeOH),
(10) Robins, M. J.; Wilson, J. S.; Hansske, F. J. Am. Chem. Soc. 1983,
105, 4059.
had the same sign as that of the authentic sample, [R]22 -55.3 (c 0.5,
D
MeOH); (+)-myristinin B/C, [R]23 +44.6 (c 0.6, MeOH).
D
Org. Lett., Vol. 8, No. 9, 2006
1927