Efficient synthesis of (+)-mk7607 and its c-1 epimer via the stereoselective transposition of a tertiary allylic alcohol

Article abstract of DOI:10.1021/ol9008987

These studies provide an efficient and stereoselective synthetic route to (+)-MK7607 and its C-1 epimer from a common intermediate in high overall yields. The synthetic methodologies mainly rely on the stereospecific 1,3-allylic transposition of the hinde

Full text of DOI:10.1021/ol9008987

ORGANIC
LETTERS
2009
Vol. 11, No. 12
2583-2586
Efficient Synthesis of (+)-MK7607 and
its C-1 Epimer via the Stereoselective
Transposition of a Tertiary Allylic
Alcohol
Chaemin Lim,† Dong Jae Baek,† Deukjoon Kim,† So Won Youn,‡ and
Sanghee Kim*,†
College of Pharmacy, Seoul National UniVersity, Seoul 151-742, Korea, and
Department of Chemistry, Pukyong National UniVersity, Busan 608-737, Korea
pennkim@snu.ac.kr
Received April 24, 2009
ABSTRACT
These studies provide an efficient and stereoselective synthetic route to (+)-MK7607 and its C-1 epimer from a common intermediate in high
overall yields. The synthetic methodologies mainly rely on the stereospecific 1,3-allylic transposition of the hindered tertiary alcohol group
through a palladium-catalyzed allylic rearrangement as well as a PBr3-mediated allylic-transposed bromination.
(+)-MK7607 (1, Figure 1) is an unsaturated carbapyranose
isolated from cultures of CurVularia eragrostidis D2452.1
R-galactose. Thus, this highly oxygenated cyclohexene com-
pound could be considered a mimetic of R-galactose,2 which
is an important component in many biological processes.
The first total synthesis of racemic MK7607 was reported
by Mehta in 2000,3 and the synthesis of the non-natural
enantiomer from (-)-shikimic acid was recorded by Singh
in 2001.4 However, the synthesis of the natural (+)-
enantiomer has not been reported so far. The C-1 epimer of
1, 1-epi-(+)-MK7607 (2), which could be regarded as a
mimetic of ꢀ-galactose, was recently synthesized.5 Very
recently, the syntheses of 5-fluorinated analogues of 1 and
2 were also reported.6
Figure 1. Structures of (+)-MK7607 (1) and its C-1 epimer 2.
As part of our ongoing research, we became interested in
the structures of 1 and 2, triggered by the fact that the latter
Structurally, it possesses four contiguous stereocenters, of
which the absolute configurations are identical to those of
(2) (a) Sears, P.; Wong, C.-H. Angew. Chem., Int. Ed. 1999, 38, 2300.
(b) Arjona, O.; Go´mez, A. M.; Lo´pez, J. C.; Plumet, J. Chem. ReV. 2007,
107, 1919.
(3) Mehta, G.; Lakshminath, S. Tetrahedron Lett. 2000, 41, 3509.
(4) Song, C.; Jiang, S.; Singh, G. Synlett 2001, 1983.
(5) Grondal, C.; Enders, D. Synlett 2006, 3507.
(6) Sardinha, J.; Rauter, A. P.; Sollogoub, M. Tetrahedron Lett. 2008,
49, 5548.
† Seoul National University.
‡ Pukyong National University.
(1) Nobuji, Y.; Noriko, C.; Takashi, M.; Shigeru, U.; Kenzou, H.;
Michiaki, I. Jpn. Kokai Tokkyo Koho, JP, 06306000, 1994.
10.1021/ol9008987 CCC: $40.75
Published on Web 05/21/2009
 2009 American Chemical Society

compound has a high affinity to a galactose-recognizing
lectin.7 We envisioned that the replacement of R- and
ꢀ-galactose moieties of carbohydrate epitopes with their
mimetics 1 and 2 could alter the inherent immunochemical
properties of the epitopes.8 In this regard, we undertook the
synthesis of these carbasugars and their derivatives. Herein,
we wish to report our efficient and stereoselective synthetic
route to (+)-MK7607 (1) and its C-1 epimer 2 from a
common intermediate.
as the only detectable isomer. The stereochemistry of the
newly generated stereocenter of 7 was tentatively assigned,
as shown, on the basis of a chelation-controlled transition
state model and established by its conversion to 8 (Scheme
4). The RCM of diene 7 was successfully performed with a
second-generation Grubbs catalyst 9 in CHCl3 at room
temperature to produce the desired cyclohexene derivative
10 in 92% yield.
With a facile route to 10 in hand, our study first focused
on the conversion of the tertiary allylic alcohol moiety of
10 to its transposed secondary allylic alcohol moiety of 1-epi-
(+)-MK7607 (2). Toward this end,10 we decided to utilize
the allylic rearrangement of the allylic ester with palladium.11
Thus, we examined the reaction of the tertiary allylic alcohol
of 10 with an acylating agent to give tertiary allylic acetate
11. The acetylation of 10 did not take place smoothly under
normal reaction conditions, but it was found that treatment
of 10 with LHMDS and acetyl chloride in THF successfully
produced the desired acetate 11 in 89% yield (Scheme 3).
Our synthetic plan for the target molecules is outlined in
Scheme 1. We envisaged that the C-1 stereocenters of 1 and
Scheme 1. Retrosynthetic Analysis of Compounds 1 and 2
Scheme 3. Synthesis of 1-epi-(+)-MK7607 (2)
2 could be selectively introduced by the 1,3-transposition
of the tertiary allylic hydroxyl group of 3, although steric
crowding around the tertiary hydroxyl group might limit the
types of reactions applicable for this transformation. The
cyclohexene ring of 3 would be constructed by a ring-closing
metathesis (RCM) reaction of diene 4. This diene 4 would
arise from galactose, which bears the same absolute con-
figurations at C-2, C-3, and C-4 as those in the final
compounds 1 and 2.
Our synthesis commenced with the preparation of the
known ketone 5 from the commercially available galactose
derivative 6, according to the reported four-step procedures
(Scheme 2).9 Addition of vinylmagnesium bromide to ketone
5 in THF at -78 °C gave the RCM substrate 7 in 91% yield
After investigating several reaction conditions for the
allylic rearrangement of 11, we found that treatment of 11
with a catalytic amount of (MeCN)2PdCl2 (10 mol %) in
refluxing EtOAc led to the exclusive formation of the desired
transposed secondary allylic acetate 12 in 74% yield. Under
these reaction conditions, no trace of the C-1 R-isomer was
Scheme 2. Synthesis of Key Intermediate Cyclohexenol 10
1
detected in the crude H NMR spectra. We established the
relative configuration of 12 ultimately through its conversion
(7) Block, O. Dissertation, University of Wuppertal, Germany, 2000.
(8) (a) Peri, F.; Cipolla, L.; La Ferla, B.; Nicotra, F. C. R. Chimica
2003, 6, 635. (b) Schmieg, J.; Yang, G.; Franck, R. W.; Tsuji, M. J. Exp.
Med. 2003, 198, 1631. (c) Tashiro, T.; Nakagawa, R.; Hirokawa, T.; Inoue,
S.; Watarai, H.; Taniguchi, M.; Mori, K. Tetrahedron Lett. 2007, 48, 3343.
(9) Agrofoglio, L. A.; Amblard, F.; Nolan, S. P.; Charamon, S.;
Gillaizeau, I.; Zevaco, T. A.; Guenot, P. Tetrahedron 2004, 60, 8397.
(10) All of our attempts to oxidize 10 to the transposed enone with a
variety of reagents were unsuccessful, presumably because of the high degree
of steric hindrance.
(11) (a) Overman, L. E.; Knoll, F. M. Tetrahedron Lett. 1979, 20, 321.
(b) Overman, L. E. Angew. Chem., Int. Ed. Engl. 1984, 23, 579.
2584
Org. Lett., Vol. 11, No. 12, 2009

to final product 2. The stereochemical outcome of this
stereoselective allylic transposition could be rationalized by
a [3,3]-sigmatropic reaction through transition state 13, in
which Pd(II) coordinates to the double bond on the less
hindered R-face opposite the acetate group and facilitates
the sigmatropic rearrangement on the ꢀ-face.11b,12
For the facile isolation of the hydrophilic final product 2,
the benzyl protecting groups of 12 were replaced with
acetates to give pentaacetate 14 by treating 12 with BCl3
and subsequent peracetylation in 90% overall yield. Finally,
the desired 1-epi-(+)-MK7607 (2) was delivered in 98%
yield upon treatment of 14 with triethylamine in MeOH and
purification on a Dowex 50 (H+) resin column. The
spectroscopic data (1H and 13C NMR) and optical rotation
for 2 were identical with those reported.5
Having achieved a selective synthesis of 2, we turned
our attention to the synthesis of (+)-MK7607 (1). We
chose to utilize 10 as a common intermediate and exploit
the route that would allow for the facile conversion to
the final natural product 1. Conceivably, the anti-SN2′ type
substitution of the tertiary hydroxyl group of 10 by an
oxygen nucleophile would provide a quick way to 1. Thus,
we investigated the feasibility of this transformation by
using various reaction conditions including the Mitsunobu
reaction.13 Unfortunately, all attempts to bring about this
transformation failed.
As an alternative approach for the desired transformation,
we decided to utilize the Winstein’s neighboring group
participation of C-2 acetate.14,15 To this end, the benzyl
groups of 10 were replaced with acetates to give 8 in 86%
overall yield (Scheme 4). The NOESY correlations of 8
was found that treatment of 8 with an excess of PBr3 in
CH2Cl2 led to the formation of the transposed allylic bromide
1517,18 as a single diastereomer in excellent yield (98%). On
the basis of the analysis of the NOE data, the relative
configuration of the newly introduced stereocenter was
assigned as ꢀ.16
Heating the obtained ꢀ-bromide 15 in refluxing wet ethanol
in the presence of Ag2CO3 afforded a 2:1 mixture of
inseparable acetates 16 and 17 in 80% combined yield. The
formation of both acetates 16 and 17, as well as the
configurational inversion of C-1, suggested that this reaction
involves the oxonium ion intermediate 18, which collapses
via path b but not via path a.14,19 Finally, removal of the
acetate groups in 16 and 17 with triethylamine in MeOH
and purification on a Dowex 50 (H+) resin column gave (+)-
MK7607 (1), whose [R]D value and other spectra were in
agreement with those reported.1
In this synthetic process, particularly worthy of note is
the stereospecific PBr3-mediated allylic-transposed bro-
mination of the tertiary allylic alcohol 8. It is known that
halogenation of tertiary allylic alcohols with PBr3 leads
to the transposed allylic bromides, and generally believed
that this transformation involves intra- or intermolecular
attack of the phosphate ester intermediate by bromide.20
To our knowledge, only a few precedents exist for the
cyclic systems, but the stereochemistry of the rearranged
products was ambiguously determined.21 Although more
systematic studies are needed to elucidate the origin of
the observed selective stereochemical outcome of 15, we
suggest that it could be a result of the SNi′ reaction
mechanism of intermediate phosphate ester 19. This
mechanistic assumption could be substantiated as follows.
To examine the possible stereodirecting effect of the C-2
acetoxyl group via anchimeric assistance, the nonanchimeric
Scheme 4. Synthesis of (+)-MK7607 (1)
(12) (a) Yuasa, Y.; Yuasa, Y. Synth. Commun. 2006, 36, 1671. (b)
Inomata, K.; Murata, Y.; Kato, H.; Tsukahara, Y.; Kinoshita, H.; Kotake,
H. Chem. Lett. 1985, 931.
(13) For related attempts, see: (a) Myers, A. G.; Glatthar, R.; Hammond,
M.; Harrington, P. M.; Kuo, E. Y.; Liang, J.; Schaus, S. E.; Wu, Y.; Xiang,
J.-N. J. Am. Chem. Soc. 2002, 124, 5380. (b) Young, J.-j.; Jung, L.-j.; Cheng,
K.-m. Tetrahedron Lett. 2000, 41, 3411.
(14) Winstein, S.; Buckles, R. E. J. Am. Chem. Soc. 1942, 64, 2780.
(15) For representative examples, see: (a) Davies, S. G.; Long, M. J. C.;
Smith, A. D. Chem. Commun. 2005, 4536. (b) Pei, Z.; Dong, H.; Ramstro¨m,
O. J. Org. Chem. 2005, 70, 6952.
(16) See Supporting Information for details.
(17) The synthesis of the enantiomer of bromide 15 as a mixture with
its C-1 isomer was previously reported, see: (a) Ogawa, S.; Hattori, T.;
Toyokuni, T. Bull. Chem. Soc. Jpn. 1983, 56, 2077. (b) Ogawa, S.; Sakata,
Y.; Ito, N.; Watanabe, M.; Kabayama, K.; Itoh, M.; Korenaga, T. Bioorg.
Med. Chem. 2004, 12, 995.
(18) Contrary to a report that bromide 15 is unstable (see ref 17), the
pure bromide 15 proved to be fairly stable in our hands.
(19) Woodward, R. B.; Brutcher, F. V., Jr. J. Am. Chem. Soc. 1958, 80,
209.
(20) For representative examples in an acyclic system, see: (a) Babler,
J. H. J. Org. Chem. 1976, 41, 1262. (b) Bakkestuen, A. K.; Gundersen,
L.-L.; Petersen, D.; Utenova, B. T.; Vik, A. Org. Biomol. Chem. 2005, 3,
1025.
(21) (a) Zhang, A.; Csutoras, C.; Zong, R.; Neumeyer, J. L. Org. Lett.
2005, 7, 3239. (b) Nampalli, S.; Bhide, R. S.; Nakai, H. Synth. Commun.
1992, 22, 1165.
allowed the unequivocal assignment of the configuration of
C-5.16 With compound 8 in hand, efforts were directed to
the activation of C-1 position by suprafacial allylic transposi-
tion. After some trials with several halogenating reagents, it
(22) (a) Robin, J.-P.; Landais, Y. Tetrahedron 1992, 48, 819. (b)
Ramage, R.; Griffiths, G. J.; Shutt, F. E. J. Chem. Soc., Perkin Trans. 1
1984, 1539.
(23) (a) Yajima, T.; Munakata, K. Chem. Lett. 1977, 891. (b) Koganty,
R. R.; Shambhu, M. B.; Digenis, G. A. Tetrahedron Lett. 1973, 14, 4511.
Org. Lett., Vol. 11, No. 12, 2009
2585

benzyl group containing alcohol 10 was subjected to the same
reaction conditions (Scheme 5). The reaction produced the
1H NMR spectra. This observation implied that the inter-
molecular mechanism involving oxonium ion might not be
a preferred pathway. In the case where the carbocation
intermediate was involved in the intermolecular mechanism,
a high selectivity would not be expected based on steric and
stereoelectronic considerations. To support these notions, we
treated compound 8 in CH2Cl2 with the complex of PBr3
with DMF as a brominating agent that was proposed to exist
as an iminium bromide salt and undergo a bromination
reaction via a stepwise intermolecular mechanism.21b,23 As
expected, the reaction of 8 gave an inseparable 1:1 mixture
of the transposed allylic bromide 15 and its C-1 R-isomer
in 69% combined yield (Scheme 5). All of the above results
indirectly supported that the SNi′ mechanism might be a
plausible explanation for the stereochemical outcome of 15.
In conclusion, here we describe an efficient and selective
synthetic route to (+)-MK7607 (1) and its C-1 epimer 2 from
a readily obtainable intermediate 10 in high overall yields.
An important feature of this synthesis was the stereospecific
1,3-allylic transposition of the hindered tertiary alcohol group
of compound 10. For this conversion, the palladium-catalyzed
allylic transposition reaction and the PBr3-mediated allylic-
transposed bromination reaction were employed. The syn-
thetic process was highly selective and efficiently provided
synthetically useful intermediates particularly in the prepara-
tion of MK7607-containing analogues.
Scheme 5
.
Allylic-Transposed Bromination of Compounds 10
and 8
C-1 ꢀ-bromide 20 as the only product isolated albeit in low
yield (33%). The low yield of 20 could be attributed to the
instability of the benzyl protecting groups under the reaction
conditions.22 This result suggested that the anchimeric
assistance of the C-2 acetoxyl group of 8 might not be a
major contributing factor in determining the selectivity in
the formation of 15.
If the bromination reaction proceeds via intermolecular
attack of bromide ion, the sole formation of 15 is unlikely,
regardless of the involvement of anchimeric assistance. When
the anchimeric assistance of the C-2 acetoxyl group was
involved in the intermolecular mechanism, the reaction would
proceed through the oxonium ion intermediate 18 (Scheme
4), and the bromide product 15 would be formed by attack
of bromide ion at C-1 (path a).
In this case, the C-1 oxygenated product might be formed
also, at least as a minor byproduct, upon competitive collapse
of the oxonium ion intermediate (path b).14,15 However, no
trace of the C-1 oxygenated product was detected in the crude
Acknowledgment. This work was supported by the Korea
Science and Engineering Foundation (KOSEF) through the
National Research Laboratory Program (M10500000055-
06J0000-05510) and the SRC/ERC program (R11-2007-107-
02001-0).
Supporting Information Available: Experimental pro-
1
cedures, product characterization, and copies of H NMR
and 13C NMR spectra. This material is available free of
charge via the Internet at http://pubs.acs.org.
OL9008987
2586
Org. Lett., Vol. 11, No. 12, 2009