Exploratory studies aimed at a synthesis of vinigrol. 2. Attempts to exploit ring-closing metathesis for construction of the central cyclooctane belt

Article abstract of DOI:10.1021/jo048458x

(Chemical Equation Presented) A program directed to the possible adaptation of ring closing metathesis to a total synthesis of vinigrol is described. With a convenient route to intermediates of general type 3 available from a prior investigation, several

Full text of DOI:10.1021/jo048458x

Exploratory Studies Aimed at a Synthesis of Vinigrol. 2. Attempts
to Exploit Ring-Closing Metathesis for Construction of the Central
Cyclooctane Belt
Leo A. Paquette,* Ivan Efremov, and Zuosheng Liu
Evans Chemical Laboratories, The Ohio State University, Columbus, Ohio 43210
paquette.1@osu.edu
Received September 1, 2004
A program directed to the possible adaptation of ring closing metathesis to a total synthesis of
vinigrol is described. With a convenient route to intermediates of general type 3 available from a
prior investigation, several candidate substrates were prepared. These included the epoxy dienes
10 and 22, the diacetoxy triene 42, and the heavily functionalized cyclohexane 48. The central
issue of this approach was to convey a maximum degree of conformational flexibility to these
functionalized intermediates, such that the olefinic termini of the side chains could enter into
intramolecular carbon-carbon bond formation. In no example was ring closure observed to operate.
Instead, the strategically placed π-bonds were seen to migrate internally to the chain in select
examples. Although the pivotal transformations failed, the deployment of a number of useful
stereocontrolled reactions has ultimately resulted in the preparation of heavily substituted cis-
decalins.
SCHEME 1
Vinigrol (1), a metabolite of Virgaria nigra, was shown
to possess unprecedented structural features on the basis
of a crystallographic determination.1 Early interest in
this substance was additionally fueled by its unusually
promising antihypertensive and platelet aggregation
inhibitory properties.2 The later discovery of its role as a
tumor necrosis factor antagonist3 has also been contribu-
tory to heightened synthetic attention,4-6 including our
projection of a possible enantioselective total synthesis.7
(1) Uchida, I.; Ando, T.; Fukami, N.; Yoshida, K.; Hashimoto, M.;
Tada, T.; Koda, S.; Morimoto, Y. J. Org. Chem. 1987, 52, 5292.
(2) (a) Ando, T.; Tsurumi, Y.; Ohata, N.; Uchida, I.; Yoshida, K.;
Okuhara, M. J. Antibiot. 1988, 41, 25. (b) Ando, T.; Yoshida, K.;
Okuhara, M. J. Antibiot. 1988, 41, 31.
(3) Norris, D. B.; Depledge, P.; Jackson, A. P. PCT Int. Appl. WO
91 07 953; Chem. Abstr. 1991, 115, 64776h.
(4) (a) Devaux, J.-F.; Hanna, I.; Lallemand, J.-Y.; Prange´, T. J. Org.
Chem. 1993, 58, 2349. (b) Devaux, J.-F.; Fraisse, P.; Hanna, I.;
Lallemand, J.-Y. Tetrahedron Lett. 1995, 36, 9471. (c) Devaux, J.-F.;
Hanna, I.; Lallemand, J.-Y. J. Org. Chem. 1997, 62, 5062. (d) Gentric,
L.; Hanna, I.; Ricard, L. Org. Lett. 2003, 5, 1139.
(5) (a) Kito, M.; Sakai, T.; Haruta, N.; Shirahama, H.; Matsuda, F.
Synlett 1996, 1057. (b) Matsuda, F.; Kito, M.; Sakai, T.; Okada, N.;
Miyashita, M.; Shirahama, H. Tetrahedron 1999, 55, 14369.
(6) Mehta, G.; Subba Reddy, K. Synlett 1996, 625.
(7) Paquette, L. A.; Guevel, R.; Sakamoto, S.; Kim, I. H.; Crawford,
J. J. Org. Chem. 2003, 68, 6096.
Detailed in part 1 of this series was a retrosynthetic plan
to arrive at the potential precursor 2 by means of the
SN2 intramolecular cyclization of iodo sulfone 3 (Scheme
1). While arrival at 3 via the oxyanionic-accelerated [3,3]-
sigmatropic rearrangement of 4 efficiently set the four
vicinal methine hydrogens in the targeted all-cis ar-
rangement, no subsequent ring closure to establish the
ansa bridge in the tricyclic product was accomplished.
The recalcitrance of 3 toward generation of the eight-
10.1021/jo048458x CCC: $30.25 © 2005 American Chemical Society
Published on Web 12/09/2004
J. Org. Chem. 2005, 70, 505-509
505

Paquette et al.
revealed by high-field 1H NMR analysis. Heating 10 with
the first-generation Grubbs catalyst9 in refluxing chlo-
roform or benzene gave no sign of reaction. When the
more reactive second-generation ruthenium catalyst10
was deployed instead, processing in hot benzene was
found to induce only double-bond migration as in 11 and
12. Related isomerizations have been previously re-
ported.11 In the present context, its operation suggests
that the intended internal coupling is kinetically inhib-
ited. A change in the functionalization pattern was
clearly mandated, and we therefore undertook to explore
that facet of Scheme 1 featuring precursors that have a
bulky substituent in â-orientation at C-7. Doing so was
expected to facilitate adoption of those conformational
changes potentially more conducive to the cyclization
reaction.
SCHEME 2a
The practicalities associated with diol 13, available via
a two-step deprotection of 5, allowed significant sums of
material to be brought forward by way of the doubly
protected acetal 14. Acid hydrolysis to unmask the ketone
carbonyl made possible the subsequent determination
that attack during sodium borohydride reduction pro-
ceeds preferentially from the more open â-surface. The
requisite inversion of configuration as 15 to 16 was
accomplished using an improved variant of the Mit-
sunobu reaction.12 The free hydroxyl group in 16, cleanly
liberated by exposure of 17 to lithium hydroxide in
aqueous THF, was protected as a tert-butyldiphenylsilyl
ether, and the primary hydroxyls were liberated by the
action of DDQ to furnish 20. The information garnered
earlier in connection with the preparation of 10 proved
to be transmittable to the elaboration of 22. We now
focused on the reactivity of 22 under RCM conditions.
In refluxing CH2Cl2 as the metathesis medium, neither
Grubbs catalyst promoted a chemical change. In agree-
ment with the previous observations, use of higher
temperatures (e.g., benzene at reflux) led to migration
of the double bond(s). Mass spectrometric analysis of
these reaction mixtures demonstrated unmistakably that
no 23 had been generated (Scheme 3).
a Reagents and Conditions: (a) m-CPBA, Na2HPO4, CH2Cl2
(99%); (b) TBAF, THF, rt (quant); (c) DDQ, CH2Cl2, H2O (quant);
(d) (COCl)2, DMSO, Et3N, CH2Cl2, -78 f -40 °C (89%); (e)
Ph3PdCH2, THF, 0 °C (77%); (f) Grubbs catalyst, C6H6, reflux (44%
of 11 and 36% of 12).
membered ring was attributed to conformational factors
that inhibit attainment of the proper SN2 reaction trajec-
tory.7 This particular methodology requires not only
spatial proximity of the reacting centers but also their
strictly defined orientation. A preferred protocol having
less stringent entropic demands should, on this basis,
serve better as a candidate for the intended mesocyclic
carbon-carbon bond formation. In light of the Chauvin
mechanism for ring closing metathesis (RCM),8 which
stresses the significance of metallocyclobutane and met-
allocarbene intermediates, the prospects for intramolecu-
lar formation of a cyclooctene double bond attracted our
attention. This choice was not without complications. The
difficulty arises from the fact that reliance must be placed
on adequate conformational flexibility within the cis-
fused decalin core.
In this paper, we delineate three attempts to harness
this particular power reaction for the purpose of accessing
vinigrol (1).
Migration of the Double Bond in the Octalin
Core. In light of the preceding developments, we next
considered the possibility of temporarily migrating the
double bond in several intermediates from its original
∆
Results and Discussion
3,4
location to the neighboring ∆2,3 site. As reflected in
Consequences of Preliminary Epoxidation. For
the RCM to proceed as planned, the endocyclic double
bond in 57 was initially protected to avoid its possible
participation. These considerations led us to advance by
means of the stereocontrolled epoxidation of acetal 5
(Scheme 2). This transformation was expedient in that
those stereochemical issues associated with reduction of
the carbonyl group were effectively skirted. By proceeding
in this direction, however, we were aware that potential
awkwardess arising from added steric congestion could
well manifest itself. Diene 10 was ultimately prepared
from 6 by sequential deprotection of the hydroxyl func-
tionalities, Swern oxidation, and 2-fold methylenation of
the resulting dialdehyde 9 by way of Wittig chemistry.
Relevantly, no epimerization at the allylic sites was
the model systems A and B (Figure 1), this relocation
was to be counted on for increased conformational flex-
ibility, such that the two side chains might be projected
axially with greater facility as suggested by MM3 calcu-
lations.
Three approaches to structural modification of the left-
hand domain were addressed in turn. The first-genera-
tion pathway was designed to involve 2-fold deployment
of the epoxide f allyl alcohol rearrangement under basic
conditions (Scheme 4). To evaluate this technology rela-
tive to the case at hand, resources were directed to the
preparation of 24, available from the disilylation of 19.
(9) Ulman, M.; Grubbs, R. H. J. Org. Chem. 1999, 64, 7202 and
references therein.
(10) Scholl, M.; Ding, S.; Lee, C. W.; Grubbs, R. H. Org. Lett. 1999,
1, 953.
(8) (a) Harrison, J.-L.; Chauvin, Y. Makromol. Chem. 1971, 141, 161.
(b) Soufflet, J.-P.; Commereuc, D.; Chauvin, Y. Compt. Rend. Acad.
Sci. Ser. C 1973, 276, 169. (c) Chauvin, Y.; Commereuc, D.; Zaborowski,
G. Makromol. Chem. 1978, 179, 1285.
(11) Corey, E. J.; Guzman-Perez, A. Angew. Chem., Int. Ed. 1998,
37, 388.
(12) Dodge, J. A.; Trujillo, J. I.; Presnell, M. J. Org. Chem. 1994,
59, 234.
506 J. Org. Chem., Vol. 70, No. 2, 2005

Exploratory Studies Aimed at a Synthesis of Vinigrol. 2
SCHEME 3a
FIGURE 1.
SCHEME 4a
a Reagents and conditions: (a) PMBCl (2.0 equiv), NaH, DMF
(43% for 14, 44% for the monoprotected alcohol); (b) 1.0 M HCl,
acetone; NaBH4, EtOH (78% for 15, 20% for 16); (c) p-nitrobenzoic
acid, PPh3, DEAD, THF (99%); (d) LiOH, H2O, THF (98%), (e)
TBDPSCl, imidazole, DMF; (f) DDQ, H2O, CH2Cl2 (84% for two
steps); (g) m-CPBA, Na2HPO4, CH2Cl2 (quant); (h) (COCl)2, DMSO,
Et3N, CH2Cl2, -78 f -40 °C (82%); (i)Ph3PdCH2, THF (65%); (j)
RCM conditions (see text).
standard conditions, as well as to lithium diisopropyla-
mide in ether.
At this juncture, dehydration of a suitable tertiary
carbinol became our next objective. Initial efforts to set
the stage for this eventuality explored the feasibility of
oxirane ring opening in 27 with potassium acetate and
18-crown-6 in DMF at 100 °C.15 These conditions gave
rise to the chromatographically separable diastereomers
30 and 31 in 52% and 12% yield, respectively (Scheme
5). A considerable enhancement in efficiency was realized
when recourse was made alternatively to sodium acetate
in 2-methoxyethanol and water.16 Selective protection of
the primary and secondary hydroxyl groups in 30 was
now mandated. The use of acetic anhydride in pyridine17
Exposure of 24 to peracid at 0 °C afforded 25. Treatment
of this intermediate with diethylaluminum 2,2,6,6-tet-
ramethylpiperidide (DATMP)13 led to regioselective for-
mation of 26 in 98% yield. Subsequent vanadium-
catalyzed epoxidation14 gave epoxy alcohol 27 as a single
isomer. When conditions were screened for the conversion
of 27 to 29, rearrangement invariably occurred to gener-
ate aldehyde 28 instead. The possibility that the proximal
free hydroxyl may have redirected the desired isomer-
ization prompted conversion of 27 to its trimethylsilyl
ether. However, this compound proved unreactive to the
(15) Momose, T.; Setoguchi, M.; Fujita, T.; Tamura, H.; Chida, N.
Chem. Commun. 2000, 2237.
(16) (a) Suami, T.; Ogawa, S.; Uematsu, Y.; Suga, A. Bull. Chem.
Soc. Jpn. 1986, 59, 1261. (b) Ogawa, S.; Uematsu, Y.; Yoshida, S.;
Sasaki, N.; Suami, T. J. Carbohydr. Chem. 1987, 6, 471.
(13) Yasuda, A.; Tanaka, S.; Oshima, K.; Yamamoto, H.; Nozaki,
H. J. Am. Chem. Soc. 1974, 96, 6513.
(14) Sharpless, K. B.; Michaelson, R. C. J. Am. Chem. Soc. 1973,
95, 6136.
J. Org. Chem, Vol. 70, No. 2, 2005 507

Paquette et al.
SCHEME 5a
SCHEME 6a
a Reagents and conditions: (a) Ac2O, DMAP, py (96%); (b) SeO2,
t-BuOOH, CH2Cl2 (69% at 79% conversion); (c) MsCl, Et3N, THF,
-78 to 0 °C (67% at 84% conversion).
SCHEME 7a
a Reagents and conditions: (a) KOAc, 18-cr-6, DMF, 100 °C; (b)
NaOAc, MeOCH2CH2OH, H2O, reflux; (c) Ac2O, DMAP, py (87%);
(d) NaOAc, MeOCH2CH2OH, H2O, reflux; (e) Ac2O, DMAP, py,
then Ac2O, DMAP, CH2Cl2 (67%).
afforded only products of monoacetylation (primary/
secondary ) 3:1 by NMR analysis). Submission of this
mixture to the action of acetic anhydride and DMAP in
CH2Cl2 solution led to the exclusive isolation of triacetate
33. Our inability to uncover a means for generating 3418
for projected dehydration studies prompted a shift in
emphasis to the third option.
The acetylation of 26 to generate 35 proved to be a
reliable first step (Scheme 6). When this intermediate
was reacted with selenium dioxide and tert-butyl hydro-
peroxide in CH2Cl2 at room temperature,19 relatively
efficient conversion to allyl alcohol 36 was observed. The
R-configuration of the newly introduced OH substituent
was ascertained on the basis of NOE data as illustrated
in C. The stage was now set for activation of the system
a Reagents and conditions: (a) n-Bu4NOAc, acetone, reflux
(92%); (b) p-TsOH, THF, MeOH (83%); (c) (COCl)2, DMSO, Et3N,
CH2Cl2, -78 °C (97%); (d) Ph3PCH3Br, n-BuLi, THF, -78 to 0 °C
(39%); (e) RCM conditions.
The desired mesylation was followed by spontaneous
attack of chloride ion at the alternative allyl terminus
(see 37) with concurrent repositioning of the double bond
to the intraannular location. This eventuality made pos-
sible the acquisition of diacetate 39 by SN2 displacement
with tetrabutylammonium acetate in acetone (Scheme
7).21 With 39 in hand, it proved an easy matter to effect
its conversion to 42 and to explore the susceptibility of
this polyolefin to RCM. No linkage between the two
terminal double bonds could be accomplished. Complete
recovery of starting material following, for example, the
heating of 42 with 20 mol % of second-generation Grubbs
catalyst in benzene for 16 h was routinely noted. More
drastic measures were therefore required.
with methanesulfonyl chloride and DMAP in CH2Cl2.20
(17) (a) Marotta, E.; Micheloni, L. M.; Scardovi, N.; Righi, P. Org.
Lett. 2001, 3, 727. (b) For comparison, consult: Kattnig, E.; Albert,
M. Org. Lett. 2004, 6, 945.
(18) Liu, Z. Unpublished observations.
(19) (a) Umbreit, M. A.; Sharpless, K. B. J. Am. Chem. Soc. 1977,
99, 5526. (b) Barrero, A. F.; Alvarez-Manzaneda, E.; Altarejos, J.;
Salido, S.; Ramos, J. M. Tetrahedron Lett. 1994, 35, 2945.
(20) Hatakeyama, S.; Numata, H.; Osanai, K.; Takano, S. J. Org.
Chem. 1989, 54, 3515.
(21) Tchilibon, S.; Mechoulam, R. Org. Lett. 2000, 2, 3301.
508 J. Org. Chem., Vol. 70, No. 2, 2005

Exploratory Studies Aimed at a Synthesis of Vinigrol. 2
SCHEME 8a
46 would be amenable to conversion to dialdehyde 47 via
Swern oxidation was matched by the subsequent Wittig
olefination. As encouraging as this sequence was, we
were again thwarted by an inability to arrive at 49 via
RCM. With either Grubbs catalyst in refluxing CH2Cl2,
only the recovery of 48 was evidenced. Recourse to higher
reaction temperatures (e.g., refluxing benzene) and longer
reaction times resulted in partial isomerization to chro-
matographically inseparable mixtures rich in 50 and 51.
Overview. In summary, the assembly of enantiopure
terminal dienes properly functionalized to serve as pos-
sible precursors to vinigrol has been achieved in conver-
gent fashion. The key deterrent to the serviceability of
these intermediates (viz., 10, 22, 42, and 48) is their
universal failure to engage in ring-closing metathesis.
Our inability to integrate this protocol into rather late
stages of the proposed vinigrol synthesis is perceived to
arise from steric congestion generated by the numerous
side chains that conspire to preclude the necessary
proximity of reaction centers. With the structurally less
substituted congener 52, Matsuda has found it possible
to arrive at 53 in 98% yield by reduction with samarium
iodide under high dilution conditions.5 The accompanying
a Reagents and conditions: (a) O3, CH2Cl2, MeOH, -78 °C then
NaBH4, Me2S, -78 °C, 1 h; (b) NaBH4, EtOH, 0 °C (74% for two
steps); (c) PivCl, DMAP, py, rt (89%); (d) p-TsOH, THF, MeOH, rt
(99%); (e) (COCl)2, DMSO, Et3N, CH2Cl2, -78 °C (91%); (f)
Ph3PCH3Br, n-BuLi, THF, -78 f 0 °C (87%); (g) RCM conditions;
(h) Grubbs II catalyst, C6H6, ∆ (72%).
Reversal in the Timing of Cyclooctene Belt In-
stallation. At this point, the assumption was made that
the eight-membered ring might be more amenable to
construction if greater degrees of conformational mobility
became available. If it is further assumed that this state
of affairs would be realized by cleavage of the ∆3,4 π-bond,
several candidate precursors are available for consider-
ation. Of these, 24 was selected for investigation. This
pragmatic choice was matched with a series of transfor-
mations designed to permit regeneration of the cyclohex-
ene ring at the proper time after installation of the belt.
This phase of the study began by sequential ozonolysis
and reduction with sodium borohydride. As anticipated,
diol 44 was efficiently generated (Scheme 8), thereby
making possible diesterification with pivaloyl chloride to
deliver 45. Next to be explored was its selective desi-
lylation with p-toluenesulfonic acid in a solvent system
constituted of THF and methanol. The expectation that
two papers describe alternative exploratory efforts to
achieve the same objectives by way of a conformational
lock model22 and practical recourse to ring contraction
alternatives.23
Acknowledgment. Partial financial support was
provided by Aventis and the Yamanouchi USA Founda-
tion whom we thank.
Supporting Information Available: Experimental de-
tails and high-field 1H and 13C NMR spectra for all compounds
described herein. This material is available free of charge via
the Internet at http://pubs.acs.org.
JO048458X
(22) Paquette, L. A.; Efremov, I. J. Org. Chem. 2005, 70, 510.
(23) Paquette, L. A.; Liu, Z.; Efremov, I. J. Org. Chem. 2005, 70,
514.
J. Org. Chem, Vol. 70, No. 2, 2005 509