Total synthesis and repudiation of the helianane family

Article abstract of DOI:10.1021/ol2022214

Total syntheses of two structures purported as (+)-heliananes were completed in six pots. Spectral comparisons, between the synthetic and natural compounds, revealed a misassignment of the eight-membered ring in the heliananes. The key step in the synthes

Full text of DOI:10.1021/ol2022214

ORGANIC
LETTERS
2011
Vol. 13, No. 20
5500–5503
Total Synthesis and Repudiation of the
Helianane Family
ꢀ
Jason C. Green, Sandra Jimenez-Alonso, Eric. R. Brown, and Thomas R. R. Pettus*
Department of Chemistry and Biochemistry, University of California, Santa Barbara,
California 93106, United States
Pettus@chem.ucsb.edu
Received August 16, 2011
ABSTRACT
Total syntheses of two structures purported as (þ)-heliananes were completed in six pots. Spectral comparisons, between the synthetic and
natural compounds, revealed a misassignment of the eight-membered ring in the heliananes. The key step in the syntheses of the proposed
structures and the confirmation of their actual structures was a diastereoselective inverse-demand DielsꢀAlder reaction between an optically
active enol ether and an ortho-quinone methide species, which was generated in situ at low temperature by the sequential addition of
methylmagnesium bromide and di-tert-butyl dicarbonate to a salicylaldehyde.
(þ)-Helianane (1) was isolated in 1997 from the marine
sponge Haliclona fascigera.1a Some years thereafter, two
additional halogenated derivatives 2 and 3 thought to
display the same ethereal ring system were reported from
Spirastrella hartmani.2 All three metabolites exhibited (þ)
optical activity. Crews proposed that (S)-(þ)-curcuphenol
(4) was the biosynthetic precursor for this family, which
appeared to be a very small subset of aromatic bisabolene
derived sesquiterpenes exclusively found in marine organ-
isms (Figure 1). The (þ)-enantiomer of 4 had been isolated
10 years earlier by Wright from the marine sponge Didiscus
flavus3 and then subsequently found in both Epipolasis4
(1) Isolation: (a) Harrison, B.; Crews, P. J. Org. Chem. 1997, 62,
Figure 1. (i) Heliananes’ supposed biosynthesis. (ii) Other sponge
2646–2648. Previous syntheses:(b) Stefinovic, M.; Snieckus, V. J. Org.
metabolites (5ꢀ8a) isolated alongside (þ)-curcuphenol (4).
Chem. 1998, 63, 2808–2809. (c) Sabui, S. K.; Venkateswaran, R. V.
Tetrahedron Lett. 2004, 45, 9653–9655. (d) Sen, P. K.; Biwas, B.;
Venkateswaran, R. V. Tetrahedron Lett. 2005, 46, 8741–8743. (e) Ghosh,
S.; Tuhina, K.; Bhowmik, D. R.; Venkateswaran, R. V. Tetrahedron
and Arenochalina sp.5 In the latter study, six additional
2007, 53, 644–651. (f) Bidyut, B.; Sen., P. K.; Venkateswaran, R. V.
natural products (5ꢀ8a) were also isolated alongside 4.
Tetrahedron 2007, 63, 12026–12036. (g) Sabui, S.; Ghosh, S.; Sarkar, D.;
Conversely, the 12 heliannuols AꢀL were speculated to
Venkateswaran, R. V. Tetrahedron Lett. 2009, 50, 4683–4684. (h) Soga,
K.; Kanematsu, M.; Yoshida, M.; Shishido, K. Synlett 2011, 8, 1171–
1173.
comprise a separate albeit larger subset of aromatic bisa-
bolene sequiterpenes that Crews and others proposed to
emanate from (R)-(ꢀ)-curcuquinone (11). Members ofthis
family have all been exclusively isolated from terrestrial
(2) Martin, M. J.; Berrue, F.; Amade, P.; Fernandez, R.; Francesch,
A.; Reyes, F.; Cuevas, C. J. Nat. Prod. 2005, 68, 1554–1555.
(3) Wright, A. E.; Pomponi, S. A.; McConnell, O. J.; Kohmoto, S.;
McCarthy, P. J. J. Nat Prod. 1987, 50, 976–978.
(4) Fusetani, N.; Sugano, M.; Matsunaga, S.; Hashimoto, K. Cell.
Mol. Life Sci. 1987, 43, 1234–1235.
(5) Butler, M. S.; Capon, R. J.; Nadeson, R.; Beveridge, A. A. J. Nat.
Prod. 1991, 54, 619–623.
(6) Macıas, F. A.; Varela, R. M.; Torres, A.; Molinillo, J. M. G.
J. Chem. Ecol. 2000, 26, 2173–2186.
r
10.1021/ol2022214
Published on Web 09/19/2011
2011 American Chemical Society

Scheme 1. Diastereoselective Cycloaddition
at generating a tertiary carbocation that might be trapped
bythe tetheredphenol toprovide the cyclicether. While the
idea of constructing an eight-membered ring in this fashion
seemed improbable, the mere existence of the natural
product gave this unconventional notion some credence.
However, to test the transformation we first required an
enantioselective construction of the corresponding ben-
zylic stereocenter.
Addition of MeMgBr (2 equiv) to the salicylaldehydes
14 or 15 (1.0 equiv, 0.1 M in Et2O, ꢀ40 °C) resulted in
sequential deprotonation of the phenol and addition to the
aldehyde (Scheme 1). The respective dianion underwent
monocarbonylation upon addition of Boc2O (1.1 equiv,
ꢀ40 °C) and β-elimination thereby leading to a short-lived
o-quinone methide intermediate (o-QM) that engaged the
nonracemic enol ether (ꢀ)-169 in a diastereoselective in-
verse-demand DielsꢀAlder reaction.10 Thus, three new
bonds were formed in a single pot. It appeared that the
aryl bromine substituent provided increased diastereocon-
trol as the chroman ketal 18 formed with >20:1 selectivity,
whereas its hydrido counterpart 17 afforded a 16:1 ratio.
Hydrolysis (camphorsulfonic acid, 0.1 equiv) of the ketal
provided the respective lactol epimers 19ꢀ20 in 90% yield
and returned the chiral alcohol that was recycled for
subsequent preparations of 16.
Figure 2. (i) Two (9, 10) of the twelve known helianuols (AꢀL)
and their terrestrial biosynthetic precursor. (ii) Prior synthesis of
10. (iii) Our proposed divergent biosynthetic pathway to 9 and 10.
plants, principally the sunflower Helianthus annuus that
deploys these compounds for phytotoxic alleopathic de-
fenses (Figure 2).6,7 However, among its members were
examples of six-, seven-, and eight-membered etherial ring
systems, all of which also displayed an additional phenol
and alcohol or alkene residues.
The supposed configurational differences and possible
biosynthetic similaries between the marine heliananes and
terrestrial heliannuols captured our interest. Macıas pre-
viously demonstrated a 7-exo-tet epoxide opening of 12 to
furnish the seven-membered ring present in six of the
heliannuols (B, C, D, F, I, J, Figure 2ii).8 While an
8-endo-tet epoxide opening had been proposed for the
biosynthesis,7b it seemed an unlikely strategy for the
five known eight-membered ring containing heliannuols
(A, G, H, K, L). We propose a biosynthetic pathway in
which curcuhydroquinone (11) undergoes oxidation to a
quinone, indiscriminate dihydroxylation of the tethered
olefin, and subsequent ketalization to produce the tricyclic
quinoid ketals 13. Reductive aromatization of one diaster-
eomer releases the secondary alcohol to afford heliannuol
A (9), whereas reduction of the other diastereomer releases
the tertiary alcohol to furnish heliannuol D (10).
After attempting several new processes aimed at con-
verting the masked aldehyde of 19 into the prenyl residue
found within 4 in a single pot, a transformation predicated
On the other hand, the origin oftheeight-memberedring
belonging to the heliananes (1ꢀ3) was more of a conun-
drum for us. The simplest explanation would be a direct
conversion of curcuphenol (4) to helianane (1), as pro-
posed by Crews. We therefore set out totest this hypothesis
by treatment of curcuphenol (4) to acidic conditions aimed
(9) Okimoto, Y.; Sakaguchi, S.; Ishii, Y. J. Am. Chem. Soc. 2002, 68,
5225–5227.
(10) For prior examples of related cycloadditions, see: (a) Van de
Water, R. W.; Magdziak, D. J.; Chau, J. N.; Pettus, T. R. R. J. Am.
Chem. Soc. 2000, 27, 6502–6503. (b) Jones, R. M.; Van de Water, R. W.;
Lindsey, C. C.; Pettus, T. R. R. J. Org. Chem. 2001, 10, 3435–3341.
(c) Selenski, C.; Mejorado, L. H.; Pettus, T. R. R. Synlett 2004, 6, 1101–
1103. (d) Selenski, C.; Pettus, T. R. R. J. Org. Chem. 2004, 26, 9196–
9203.
(11) Fleming, I.; Patterson, I. Synthesis 1979, 446–448.
(12) (a) Wang, S.; Gates, B. D.; Swenton, J. S. J. Org. Chem. 1991, 56,
1979–1981. (b) Berard, D.; Jean, A.; Canesi, S. Tetrahedron Lett. 2007,
48, 8238–8241.
(7) (a) Macıas, F. A.; Varela, R. M.; Torres, A.; Milinillo, J. M. G.;
Fronzek, F. R. Tetrahedron Lett. 1993, 34, 1999–2002. (b) Macıas, F. A.;
Molinillo, J. M. G.; Varella, R. M.; Torres, A.; Fronzek, F. D. J. Org.
Chem. 1994, 59, 8261–8266. (c) Macıas, F. A.; Varela, R. M.; Torres, A.;
Molinillo, J. M. G. Tetrahedron Lett. 1999, 40, 4725–4728.
(8) Macıas, F. A.; Chinchilla, D.; Molinillo, J. M. G.; Marin, D.;
Varela, R. M.; Torres, A. Tetrahedron 2003, 59, 1679–1683.
(13) Green, J.; Pettus, T. R. R. J. Am. Chem. Soc. 2011, 133, 1603–
1608.
Org. Lett., Vol. 13, No. 20, 2011
5501

Our next line of thinking was to explore oxidative formats
and perhaps use the dearomatization of 4 and subsequent
formation of the phenoxonium C to induce an intramole-
cular [3 þ 2] cycloaddition with tethered olefin (Scheme 4).
Our initial confidence was bolstered by reports of inter-
molecular [3 þ 2] reactions under oxidative conditions.12
We reasoned that if D were formed in any appreciable
amount, then it might restore aromaticity either by dis-
placement from addition atthe C5site or by deprotonation
of the C4 site and eliminative opening of the fused five-
memberedrings. While these notions were notsuccessful in
that aspect, they led to the discovery of a novel [5 þ 2]
cyclizationꢀacetoxylation reaction that provided adduct
22 in 61% yield.13
Scheme 2. Synthesis of (þ)-Curcuphenol
upon the homologation of the carbonyl into an allyl silane
and subsequent protodesilylation,11 we chose instead to
employ two sequentialWittigreactionswithanintervening
hydrolysis to the aldehyde 21 (Scheme 2). This sequence
provided curcuphenol (þ)-4 in a respectable 84% yield
from (þ)-19 over the two pots.
Scheme 3. Attempts for Closing the Eight-Membered Ringa
Scheme 5. Syntheses of Benzopyran Natural Products
a Hþ = F3CCO2H, HBr, HCl, H2SO4, MeSO3H, p-TsOH, camphor-
sulfonic acid, HCO2H, AcOH. Xþ = BF3•Et2O, AlCl3, I2, Br2, PhSeCl,
PhSeBr, SnCl4, ZnCl2, Tl(NO3)3, Tl(O2CCF3)3.
Our first attempts at a biomimetic construction of the
etherial ring were guided by the previous investigations of
Crews and Macıas (Scheme 3). We began by examining a
range of nonaqueous Brønsted acids in the hope of forming
the tertiary cation intermediate A. We also attempted using
Lewis acids to form a species resembling B, which might
undergo cyclization to either the seven- or eight-membered
rings. Unfortunately, these efforts all failed to afford any of
the desired eight-membered cyclic ether. While we did
observe trace amounts of the corresponding seven-mem-
bered rings, these experiments mostly led to undesired electro-
philic aromatic substitutions or recovered starting material.
Having exhausted the proposed biomimetic applications
emanating from curcuphenol (4), our attention turned
toward examining the reactivity of the allylic cation that
could be generated upon protonation of metabolite 6.
Addition of excess (2-methylprop-1-en-1-yl)-magnesium
bromide (2.1 equiv) to the hemiketal 19 afforded an easily
separable mixture of the secondary alcohols 23 and 24 in a
respective 4.1:1 diastereomeric ratio and 71% combined
yield (Scheme 5). This is a surprising stereoselection of
special note, as we postulate that a series of A-1,3 steric
effects guided the addition to the aldehyde, which emerged
from deprotonationof19. Whereas treatment of6 to acidic
conditions would generate the transoid allylic cation F,
we had hoped that protonation of 23 or 24 might afford
a fractional amount of the cisoid intermediate G and
lead to an eight-membered ring. Unfortunately, these at-
tempts afforded only a mixture of benzopyrans 25 and 26.
Scheme 4. Attempted Intramolecular [3 þ 2] Cycloaddition
(14) Lipshutz, B. H.; Chung, D. W.; Rich, B.; Corral, R. Org. Lett.
2006, 8, 5069–5072.
(15) Isolation:Jakupovic, J.; Warning, U.; Bohlmann, F.; King,
R. M. Rev. Latinoam. Quim. 1987, 18, 75–76.
5502
Org. Lett., Vol. 13, No. 20, 2011

Individual submission of either 23 or 24 to an intramole-
cular Mitsunobu reaction afforded the respective ethers in
pure form.14 Remarkably, an inseperable mixture of these
two benzopyrans had been partially identified from ex-
tracts of the aerial parts of an Argentinian Baccharis
species.15 By examination of the various protonꢀproton
consequence of conformational flexibility on the NMR
time scale. For the natural material, however, the supposed
aliphatic etherial carbon resonance differed by ∼10 ppm
from the synthetic material. Similar inconsistencies were
also apparent for comparisons of synthetic and natural
(þ)-2. Our continued analyses led to the inescapable
conclusion that the synthetic and natural compounds
had different structures. From our analyses, we speculated
that the three compounds previously claimed as the helia-
nanes (1ꢀ3) were in fact curcudiol 8a and its previously
unknown halogenated derivatives 8b and 8c. Our hypoth-
esis was confirmed by separate oxymercuration and reduc-
tion of curcuphenol (4)16 and bromocurcuphenol (30),17
which furnished the synthetic compounds 8a and 8b
(Scheme 7), the spectra of which were identical to those
of the proposed heliananes 1 and 2.
1
relationships apparent from the respective H-NOESY
spectra for each pure compound, we have confirmed their
prior speculative stereochemical assignments. The cis dia-
stereomer 25 shows a strong NOE effect between the
benzylic and allylic protons, whereas the trans diastereo-
mer 26 displays a similar interaction between the benzylic
methyl substituent and the allylic proton.
Unable to mimic a plausible biosynthetic reaction, we
turned to a metathesis strategy, which had been employed
for similar eight-membered rings.1b,d,fꢀh Treatment of (þ)-
20 with excess ylide generated from addition of n-BuLi
(2.05 equiv) to Ph3PCH3I (2.1 equiv) resulted in a Wittig ole-
fination and furnished the terminal olefin (þ)-27 (Scheme 6).
The phenol of (þ)-27 smoothly underwent reverse O-pre-
nylation upon treatment with methyl (2-methylbut-3-en-
2-yl) carbonate (10 equiv) and palladium tetrakis(triphenyl-
phosphine) (0.01 equiv). Ring closing metathesis afforded
the expected eight-membered heterocycle (þ)-29. Palladium-
catalyzed hydrogenation in ethanol provided (þ)-1, where-
as solvation with ethyl acetate afforded (þ)-2.
Scheme 7. Syntheses of the Curcudiols
In conclusion, our efforts have led to (1) the enantiose-
lective total syntheses of the two structures previouly
assignedasheliananesin six pots, (2) the structural revision
and repudiation of the helianane family, (3) total syntheses
of curcudiol (8a) and the previously unrecognized bromo-
curcudiol (8b), (4) the structural confirmation of two addi-
tional benzopyran natural products 24 and 25, (5) the
discovery of a new intramolecular [5 þ 2] cycloaddition-
acetoxylation reaction, and (6) the demonstration of con-
cise applications of our method for construction of ben-
zylic stereocenters.
Scheme 6. Completion of the Eight-Membered Ring
Acknowledgment. T.R.R.P. is deeply grateful that the
National Science Foundation (CHE-0806356) has sup-
ported this and other o-quinone methide work. J.C.G.
would like to thank the Robert H. DeWolfe Fellowship for
additional support.
The 1H NMR spectra for our synthetic compound (þ)-1
were identical to those of all seven of the previously
reported total syntheses.1bꢀh However, we noted several
inconsistencies with the proton resonances of the natural
product orignally assigned as (þ)-1. Contradictions be-
tween the 13C NMR spectra for natural and synthetic
compounds were more obvious. In the case of the synthetic
material, many of the carbon resonances were broad as a
Supporting Information Available. Experimental pro-
cedures and characterization data for all compounds.
This material is available free of charge via the Internet
at http://pubs.acs.org.
(17) See the Supporting Information for a one-step synthesis of
racemic curcuphenol and bromocurcuphenol.
(16) Machiko, O.; Yamamoto, Y.; Akita, H. Chem. Pharm. Bull.
1995, 43, 553–558.
Org. Lett., Vol. 13, No. 20, 2011
5503