Total synthesis of (±)-hyphodermins A and D

Article abstract of DOI:10.1021/jo800227p

(Chemical Equation Presented) An efficient formal synthesis of (±)-hyphodermins A and D, metabolites of Hyphoderma radula, has been completed in 12 and 11 steps, respectively. The tricyclic carbon skeleton of enone 6 was rapidly assembled from diester 11 via an α brominationn - elimination sequence followed by anhydride formation. Regioselective reduction of the lactone group of enone 6 with LiA1H(t-BuO)₃ gave lactol 15. Lactol 15 was converted in two steps to (±)-hyphodermin D, without the need for complex protection-deprotection strategies. Lactol 15 was converted in three steps to (±)-hyphodermin A, via the key step of epoxidation of an enone in the presence of a THP lactol. A combination of NMR and ab initio studies suggests that the structures of hyphodermin C and D should be interchanged.

Full text of DOI:10.1021/jo800227p

Total Synthesis of (()-Hyphodermins A and D
Wendy A. Loughlin,*,†,‡ Ian D. Jenkins,‡ Luke C. Henderson,‡ Marc R. Campitelli,‡ and
Peter C. Healy†,‡
School of Biomolecular and Physical Sciences, Griffith UniVersity,
Brisbane, QLD, 4111, Australia, and Eskitis Institute for Cell and Molecular Therapies, Griffith
UniVersity, Brisbane, QLD, 4111, Australia.
w.loughlin@griffith.edu.au
ReceiVed January 28, 2008
An efficient formal synthesis of (()-hyphodermins A and D, metabolites of Hyphoderma radula, has
been completed in 12 and 11 steps, respectively. The tricyclic carbon skeleton of enone 6 was rapidly
assembled from diester 11 via an R brominationn-elimination sequence followed by anhydride formation.
Regioselective reduction of the lactone group of enone 6 with LiAlH(t-BuO)3 gave lactol 15. Lactol 15
was converted in two steps to (()-hyphodermin D, without the need for complex protection-deprotection
strategies. Lactol 15 was converted in three steps to (()-hyphodermin A, via the key step of epoxidation
of an enone in the presence of a THP lactol. A combination of NMR and ab initio studies suggests that
the structures of hyphodermin C and D should be interchanged.
Introduction
Hyphodermins A, B, C, and D (assigned as structures 1, 2,
3, and 4, respectively) are novel naphtho[1,2-c]furan-3,9-dione
derivatives isolated in 1995 from a culture of the basidiomycete
Hyphoderma radula (WP 2184), obtained from the trunk of a
wild cherry tree in Wuppertal (Germany) as the major metabo-
lites, in conjunction with other metabolites hyphodermins E-H.1
Biological studies identified these metabolites as being potential
drug leads for the treatment and prophylaxis of asthma, chronic
bronchitis, as well as heart and CNS illnesses.1 The hyphoder-
mins are only available in milligram amounts from Hyphoderma
radula, so there is considerable interest in developing syntheses
for them.
example, the simple antibacterial corollosporine3 has been made,
whereas syntheses of more complex structures,4,5 such as
betolide,6 remain unreported. Incorporation of the lactol unit,
even in simple aromatic compounds, is often complicated by
over-reduction,7,8 generation of regioisomers,8 and generation
of unstable species,9 or it occurs as a byproduct.10 However,
We recently reported the first total synthesis of (()-hypho-
dermin B (2)2 but the synthesis of other major hyphodermins,
such as A, C, and D (1, 3, and 4) has remained a challenge.
Literature syntheses of related compounds containing an aro-
matic lactol structure (3-hydroxyphthalide) are limited. For
* To whom correspondence should be addressed. Phone: + (07) 3735 7567.
Fax: + (07) 3735 7656.
(3) Ohzeki, T.; Mori, K. Biosci., Biotechnol., Biochem. 2001, 65, 172.
(4) Miyase, T.; Yamamoto, R.; Ueno, A. Chem. Pharm. Bull. 1996, 44, 1610.
(5) Rajasekhar, D.; Subbaraju, G. V. Fitoterapia 2000, 71, 598.
(6) (a) Bankova, V.; Koeva-Todorovska, J.; Stambolijska, T.; Ignatova-
Groceva, M-D.; Todorova, D.; Popov, S. Z. Naturforsch., C: J. Biosci. 1999,
54, 876. (b) Tkachev, V. V.; Nikonov, G. K.; Atovmyan, L. O.; Kobzar, A. Ya.;
Zinchenko, T. V. Khim. Prir. Soedin. 1987, 811.
† School of Biomolecular and Physical Sciences, Griffith University.
‡ Eskitis Institute for Cell and Molecular Therapies, Griffith University.
(1) Henkel, T.; Mueller, H.; Schmidt, D.; Wollweber, H. Ger. Offen. DE
19611366, 1997; Chem. Abstr. CAN 127:26006.
(2) Henderson, L. C.; Loughlin, W. A.; Jenkins, I. D.; Healy, P. C.;
Campitelli, M. R. J. Org. Chem. 2006, 71, 2384.
10.1021/jo800227p CCC: $40.75
Published on Web 03/26/2008
2008 American Chemical Society
J. Org. Chem. 2008, 73, 3435–3440 3435

Loughlin et al.
SCHEME 1. A Retrosynthetic Approach to Hyphodermin
A (1)
SCHEME 2. Synthetic Approach to Lactol 15
regioselective reduction of anhydride 7 was exploited in the
synthesis of (()-hyphodermin B (2).2 Few literature syntheses
exist of related compounds containing an aromatic conforma-
tionally rigid cyclic R,ꢀ-epoxy ketone. For example, epoxidation
of conformationally rigid cyclic R,ꢀ-unsaturated ketones, with
optically active hydroperoxides11 or via a chlorohydrin,12 give
poor to moderate yields of the corresponding epoxy ketone. The
biological activity and unique skeleton incorporating the 3-hy-
droxyphthalide and adjacent aromatic conformationally rigid
cyclic R,ꢀ-epoxy ketone make hyphodermins A, C, and D (1,
3, and 4) challenging and important synthetic targets. There is
only one other example in the literature of a compound that
contains both an R-epoxycyclohexanone and a lactol, and it has
not been synthesized.13 Herein we report the first total synthesis
of (()-hyphodermins A and D (from commercially available
4,4-dimethylcyclohexan-1,3-dione), which represents the first
successful synthesis of a molecule that contains this unique
combination of reactive functional groups.
by using Spartan ’04.15 Analogous to our previous study with
anhydride 7,2 the C1 carbon of 5 was the most electron deficient
carbonyl and thus most likely to undergo reduction.
Treatment of anhydride 7 either with phenylselenium chloride
then hydrogen peroxide16 or with DDQ17 gave complex mixtures.
Installation of the R,ꢀ-double bond via a ꢀ-hydrohalo elimination
was an alternative, and would require the R-bromo anhydride 8.
Treatment of anhydride 7 with NBS in THF/CCl4 gave R-bromo
anhydride 8 (19%). Addition of anhydride 7 (in CHCl3) to a
solution of copper(II) bromide in EtOAc at reflux gave the
bromoethyl ester 9 (52%). The solid state structure of 9 was
determined by X-ray crystallography (see the Supporting
Information). Repeating the latter reaction in THF resulted in
the formation of R-bromo anhydride 8 (20%) and bromo diacid
10 (35%). Treatment of anhydride 7 with elemental bromine18
in THF at rt gave exclusively bromo anhydride 8 (64%).
Results and Discussion
The previous synthesis of hyphodermin B (2)2 had provided
us access to the intermediate anhydride 7. Initially we considered
that conversion of anhydride 7 to enone 6, followed by
epoxidation to give 5 and subsequent reduction, would provide
a facile route to hyphodermin A (1, Scheme 1). Whereas in the
synthesis of the massileunicellins, an epoxy ketone intermediate
was selectively reduced by LiAlH(t-BuO)3 to the corresponding
epoxy alcohol,14 no precedent exists for the reduction of an
epoxy ketone anhydride to an epoxy ketone lactol. To check
the validity of the proposed reduction of epoxy ketone anhydride
5, molecular modeling was carried out at the HF/6-31G* level
Treatment of the anhydride 8 with potassium tert-butoxide
gave a complex mixture along with recovered 8 (29%). Similar
results were obtained with 2,6-lutidine, while DABCO or DBU
gave diacid 14 (13–22%) as part of a complex mixture. It was
clear that selective dehydrobromination was not possible in the
presence of the anhydride moiety, so we turned our attention
to the corresponding dimethyl ester 11.2 R bromination of diester
11 with either bromine or copper(II) bromide in MeOH gave
bromo diester 12 in 50% or 78% yield, respectively. Treatment
of bromo diester 12 with DABCO gave unsaturated diester 13
(81%), which was hydrolyzed with sodium hydroxide to diacid
14 (75%). The synthetic route from diester 11 to diacid 14
represented a robust and clean synthesis and was used in the
generation of larger quantities of 14 (8–9 g) (Scheme 2).
(7) (a) Donati, C.; Prager, R. H.; Weber, B. Aust. J. Chem. 1989, 42, 787.
(b) Delorme, D.; Ducharme, Y.; Brideau, C.; Chan, C.-C.; Chauret, N.; Desmarais,
S.; Dubé, D.; Falgueyret, J.-P.; Fortin, R.; Guay, J.; Hamel, P.; Jones, T. R.;
Lépine, C.; Li, C.; McAuliffe, M.; McFarlane, C. S.; Nicoll-Griffith, D. A.;
Riendeau, D.; Yergey, J. A.; Girard, Y. J. Med. Chem. 1996, 39, 3951.
(8) (a) Sugimoto, A.; Tanaka, H.; Eguchi, Y.; Ito, S.; Takashima, Y.;
Ishikawa, M. J. Med. Chem. 1984, 27, 1300. (b) Coucy, C.; Favreau, D.; Kayser,
M. M. J. Org. Chem. 1987, 52, 129.
(9) Troll, T.; Schmid, K. Tetrahedron Lett. 1984, 25, 2981.
(10) Moriconi, E. J.; O’Connor, W. F.; Wallenberger, F. T. J. Am. Chem.
Soc. 1959, 81, 6466.
(15) Spartan ’04; Wavefunction Inc: Irvine, CA.
(16) (a) Engman, L. J. Org. Chem. 1988, 53, 4031. (b) Honda, T.; Yoshizawa,
H.; Sundararajan, C.; Gribble, G. W. J. Org. Chem. 2006, 71, 3314.
(17) (a) Barrett, A. G. M.; Blaney, F.; Campbell, A. D.; Hamprecht, D.;
Meyer, T.; White, A. J. P.; Witty, D.; Williams, D. J. J. Org. Chem. 2002, 67,
2735–2750. (b) Dey, S.; Mal, D. Tetrahedron Lett. 2005, 46, 5483–5486. (c)
Chin, C.-L.; Tran, D. D.-P.; Shia, K.-S.; Liu, H.-J. Synlett 2005, 3, 417–420.
(18) For over-halogenation of ketones using elemental bromine see: Cheng,
Y.; Zhang, F.; Rano, T. A.; Lu, Z.; Schleif, W. A.; Gabryelski, L.; Olsen, D. B.;
Stahlhut, M.; Rutkowski, C. A.; Lin, J. H. Bioorg. Med. Chem. Lett. 2002, 12,
2419–2422.
(11) (a) Adam, W.; Rao, P. B.; Degen, H.-G.; Saha-Moller, C. R. Eur. J.
Org. Chem. 2002, 4, 630. (b) Adam, W.; Bheema Rao, P.; Degen, H.-G.; Saha-
Moller, C. R. Tetrahedron: Asymmetry 2001, 12, 121.
(12) Palitzsch, P. Ger. (East) DD 104286 19740305, 1974; Chem. Abstr. CAN
81:77745.
(13) Hashimoto, T.; Nakamura, I.; Tori, M.; Takaoka, S.; Asakawa, Y.
Phytochemistry 1995, 38, 119.
(14) Chai, Y.; McIntosh, M. C. Tetrahedron Lett. 2004, 45, 3269.
3436 J. Org. Chem. Vol. 73, No. 9, 2008

Total Synthesis of (()-Hyphodermins A and D
SCHEME 3. Synthetic Approach to Hyphodermins C (3)
and D (4)
Treatment of diacid 14 with acetic anhydride in THF at 50
°C gave unsaturated anhydride 6 (85%), as a rather insoluble
solid. Reduction of 6 with LiAlH(t-BuO)3 under conditions of
high dilution gave unsaturated lactol 15 with clean conversion
and in an isolated yield of 38% (Scheme 2). Variation of the
reaction temperature, solvent, and concentration did not increase
the yield of lactol 15. The solid-state structure of lactol 15 has
been reported.19 Treatment of 15 with dimethyldioxirane in
acetone solution gave unsaturated anhydride 6 (45%) instead
of hyphodermin A (1). This result highlighted the synthetic
challenge of installation of the epoxide moiety in the presence
of the lactol group.
Attempted epoxidation of the unsaturated anhydride 6 with
excess dimethyldioxirane gave recovery of 6 (60%) and diacid
14 (40%). Treatment of 6 with an excess amount of tert-butyl
hydroperoxide and DBU,20 or with KF/Al2O3,21 gave a complex
mixture of compounds. The difficulty of working with 6,
compounded by its poor solubility and high susceptibility to
hydrolysis, led to an examination of the more stable enone
diester 13. Treatment of 13 with tert-butyl hydroperoxide and
DBU gave epoxy diester 16 (68%). However, attempted
hydrolysis of the ester 16 with LiOH under very mild conditions
(30 min at 0 °C) or with Ba(OH)2 for 2 h at rt, followed by
bubbling CO2 through the solution,22 gave a complex mixture
of products. Epoxidation of enone diacid 14 was also attempted.
Treatment of 14 with alkaline H2O2 or with tert-butyl hydro-
peroxide and NaOH gave a complex mixture of products,
whereas treatment with m-CPBA in THF at reflux for 16 h
returned unreacted starting material 14 (46%). An alternative
synthetic route to hyphodermin A (1), involving protection of
the lactol 15, was considered.
followed by addition of Et3N was unsuccessful. Under basic
conditions a carboxylate intermediate is formed that is prefer-
entially alkylated (as occurs with benzyl bromide in the
formation of benzyl ester 18).
As the C1 carbon of lactol 15 can be considered as having
similar reactivity to the anomeric carbon of monosaccharides,
it was proposed that acid-catalyzed glycosidation chemistry24
might be applied to lactol 15. Stirring lactol 15 in methanol
containing dry HCl (generated in situ from the reaction between
methanol and acetyl chloride) in the presence of 4 Å molecular
sieves for 2 h gave the (()-methyl “acetal” 19 (90%). Epoxi-
dation of 19 was expected to give a mixture of hyphodermins
C (3) and D (4) (Scheme 3), since based on the X-ray structure
of the closely related 15, both faces of 19 would be expected
to be equally accessible (the O-methyl group is too remote to
exert a steric effect). Treatment of 19 with tert-butyl hydrop-
eroxide and DBU, gave (()-hyphodermin D (4, 29%) containing
a trace (∼2%) of the diastereoisomeric hyphodermin C (3). Use
of KF/Al2O3 and tert-butyl hydroperoxide in DCM or H2O2 and
a catalytic amount of LiOH in THF/H2O for 16 h gave a single
product, (()-hyphodermin D (4) (16 and 22%, respectively).
Only (()-hyphodermin D (4) was observed in the crude product
mixture and subsequently isolated. If (()-hyphodermin C (3)
was formed it may have been unstable under the reaction and/
or workup conditions, thus accounting for the low yields
observed.
The 1H and 13C NMR data for isolated synthetic (()-
1
hyphodermin D (4) in CDCl3 were consistent with the H and
13C NMR data reported for natural 4 isolated from Hyphoderma
radula.1 Although the trace of hyphodermin C (3) was not
isolated, the 1H NMR data of the crude product were consistent
Acetylation of lactol 15, using acetic anhydride and DMAP,
gave acetyl lactol 17 (83%). The solid state structure of acetyl
lactol 17 was determined by X-ray crystallography (see the
Supporting Information). Treatment of 17 with KF/Al2O3 and
tert-butyl hydroperoxide gave a complex mixture of unidentified
products. Attempted protection of lactol 15 by treatment with
TBDMS-Cl in the presence of 2,6-lutidine23 or with benzyl
bromide followed by addition of Et3N or with TIPS-OTf
1
with the H NMR data reported for natural 3 isolated from
Hyphoderma radula.1 In the original report,1 there is no
discussion as to how the relative stereochemical assignments
were made of the natural products 3 and 4. In an attempt to
confirm the relative stereochemistry of synthetically derived 4,
a NOE NMR experiment was run on a pure sample of 4 in
CDCl3 at 400 MHz. No NOE correlation was observed between
the H1 singlet of 4 at δ 6.74 ppm and the H8 epoxide doublet
of 4 at δ 3.77 ppm. However, approximation of the interatomic
distances [ab initio] between H1 and H8 for 4 indicated the
distance between H8 and H1 was 5.033 Å, which is on the
(19) Henderson, L. C.; Loughlin, W. A.; Jenkins, I. D.; Healy, P. C. Acta
Crystallogr., Sect. E: Struct. Rep. Online 2006, E62 (4), o1440–o1442.
(20) (a) Kuwahara, S.; Imada, S. Tetrahedron Lett. 2005, 46, 547–549. (b)
Knappwost-Gieseke, C.; Nerenz, F.; Wartchow, R.; Winterfeldt, E. Chem.-Eur.
J. 2003, 9, 3849–3858. (c) Yadav, V. K.; Kapoor, K. K. Tetrahedron 1995, 51,
8573–8584.
(21) (a) Frutos, Ó. D.; Atienza, C.; Echavarren, A. M. Eur. J. Org. Chem.
2001, 2001, 163–171. (b) Yadav, V. K.; Kapoor, K. K. Tetrahedron 1996, 52,
3659–3668. (c) Yadav, V. K.; Kapoor, K. K. Tetrahedron Lett. 1994, 35, 9481–
9484.
(23) (a) Wrona, I. E.; Gabarda, A. E.; Evano, G.; Panek, J. S. J. Am. Chem.
Soc. 2005, 127, 15026–15027. (b) Staben, S. T.; Linghu, X.; Toste, F. D. J. Am.
Chem. Soc. 2006, 128, 12658–12659.
(22) (a) Kiyota, H.; Masuda, T.; Chiba, J.; Oritani, T. Biosci., Biotechnol.,
Biochem. 1996, 60, 1076–1080. (b) Baldwin, J. E.; Chen, D.; Russell, A. T.
Chem. Commun. 1997, 24, 2389–2390. (c) Shafer, A. M.; Kalai, T.; Liu, S. Q. B.;
Hideg, K.; Voss, J. C. Biochemistry 2004, 43, 8470–8482.
(24) Gao, F.; Yan, X.; Shakya, T.; Baettig, O. M.; Ait-Mohand-Brunet, S.;
Berghuis, A. M.; Wright, G. D.; Auclair, K. J. Med. Chem. 2006, 49, 5273–
5281.
J. Org. Chem. Vol. 73, No. 9, 2008 3437

Loughlin et al.
SCHEME 4. Attempted Cleavage of Methyl “Acetal”
opposite to those given in the literature.1 Definitive proof of
the structure of synthetic hyphodermin C and D could be
obtained from X-ray crystallography; however, given synthetic
hyphodermin D was isolated as an oil this is unlikely. For the
remainder of this paper, synthetic hyphodermin D will be
referred to by the more likely structure (()-3, and synthetic
hyphodermin C will be referred to by the more likely structure
(()-4.
Treatment of synthetic hyphodermin D ((()-3) with HCl/
silica in CHCl3 gave the chlorohydrin (()-20 (99%) (Scheme
4). Presumably, attack of the chloride ion at C7 is hindered by
the geminal dimethyl group, thus preventing the nucleophile
from achieving the required trajectory to open the epoxide ring
at the ꢀ-position. The formation of an R-halo species is unusual
with only a few examples in the literature.27 Analysis of the 1H
NMR 3J couplings between H7 and H8 in 20 (∼12 Hz)
suggested that H7 and H8 were pseudo-trans-diaxially orientated
in 20, thus implying that the chlorine and hydroxyl groups were
in a pseudo-trans-diequatorial orientation. Hyphodermin D ((()-
3) could be quantitatively regenerated from 20 by treatment with
DBU. This would occur via a (less stable) conformation in which
the chlorine and hydroxyl groups adopt a pseudo-trans-diaxial
arrangement. Given the apparent stability of the methyl acetal
protecting group, a tetrahydropyranyl ether was examined next.
Treatment of lactol 15 with 3,4-dihydropyran in the presence
of ethereal HCl gave crude tetrahydropyranyl (THP) ether 21
(91%), as a mixture of diastereomers (by NMR). Purification
of 21 was complicated by rapid decomposition, so the crude
THP ether 21was reacted directly with tert-butyl hydroperoxide
and DBU to give the THP epoxide 22. Crude THP epoxide 22
was treated immediately with a solution of THF/aqueous HCl
(5% w/w) (7:3) at rt for 5 h to give synthetic (()-hyphodermin
A (1, 24%; 2 steps from THP ether 21, Scheme 5). Unfortu-
nately, 1 was also unstable and repeated chromatography
resulted in a product of ∼70% purity. Nonetheless, use of the
THP protecting group has resulted in the successful conversion
of lactol 15 to (()-hyphodermin A (1/23) in three steps.
FIGURE 1. Representations of energy-minimized structures of (top)
3 and (bottom) 4 (Jaguar 7.0, B3LYP/6-31G**). O9 · · ·H1 distance: 3,
2.625 Å; 4, 2.611 Å.
threshold of a NOE correlation distance (5.0 Å). The model
also revealed that the direct line between H1 and H8 is blocked
by the ketone moiety, which makes observing a NOE correlation
in 4 highly unlikely. A chemical approach was also considered.
Reduction of the ketone group in 4 would introduce a stereo-
center at C9, thereby providing an additional hydrogen at C9,
the proximity of which to H1 would be well within NOE
distances. Treatment of 4 with ethanolic NaBH4 under very mild
conditions (0 °C, 10 min) gave a complex mixture of products.
In the absence of X-ray structures for 3 and 4, NMR shift
1
analysis and ab initio calculations were carried out. In the H
NMR of synthetic 3 and 4, H1 of 4 (δ 6.74 ppm) is downfield
of H1 of 3 (δ 6.43 ppm), whereas the O-methyl group of 4 (δ
3.7 ppm) is upfield of the O-methyl of 3 (δ 3.82 ppm).25 This
is explained if the deshielding zone of the ketone carbonyl
group26 is orientated such that H1 is in a more deshielded zone
(therefore further downfield) and the O-methyl group is in a less
deshielded zone, (therefore further upfield) than in 3. In support
of these predictions, the ab initio calculations on structures 3 and
4 confirm that in 4, the ketone carbonyl group moves out of the
plane of the aromatic ring toward the O-methyl group, giving a
dihedral angle of -49.8° between the carbonyl group and the
H1-C1 bond, whereas in 3, the ketone carbonyl group moves out
of the plane of the aromatic ring toward H1, giving a dihedral
angle of -29.7° between the carbonyl and the H1-C1 bond
(Figure 1). Moreover, these calculations predict that H1 in
structure 4 has a higher NMR shielding constant (25.45) than
H1 in structure 3 (shielding constant 25.06), i.e., H1 in 4 is
further upfield than H1 in 3.
1
The H and 13C NMR spectra and HRMS data of synthetic
hyphodermin A (1) in methanol-d4 were consistent with the data
reported for (+)-1 isolated as a single diasteromer from
Hyphoderma radula.1 However, there was no justification given
in the original report1 for the assignment of an anti relative
configuration of the epoxy and hydroxyl groups (structure 1).
If 22 is the syn product (by analogy with 3), then cleavage of
the THP group from 22 should give the syn lactol 23 as the
initial product (Scheme 5). This could then undergo ring-opening
On the basis of these considerations (Figure 1), it is more
likely that hyphodermin C, with the more upfield chemical shift
for H1 (δ 6.43 ppm), has the anti structure 4, and hyphodermin
D (δ 6.74 ppm) the syn structure 3. These assignments are
(25) For comparison, H1 in 3-methoxy-1(3H)-isobenzofuranone is at δ 6.25
ppm while the O-methyl group is at δ 3.61 ppm: Sloan, K. B.; Koch, S. A. M.
J. Org. Chem. 1983, 48, 3777–3783.
(27) (a) Loken, B.; Kaufmann, S.; Osenkranz, G.; Sondheimer, F. J. Am.
Chem. Soc. 1956, 78, 1738–1741. (b) Belyaev, V. F.; Kozlyak, R. I. Vestn. Mosk.
UniV., Ser. 2: Khim. 1977, 1, 28–31.; Chem. Abstr. CAN 88:120886. (c)
Kyasimov, A. S.; Movsumzade, M. M.; Safarova, Z. A. Dokl. Akad. Nauk Azerb.
1979, 35, 31–35.; Chem. Abstr. CAN 92:41400. (d) Belyaev, V. F.; Kozlyak,
R. I. Vestsi Akad. NaVuk, Ser. Khim. NaVuk 1980, 4, 85–89.; Chem. Abstr. CAN
94:15418.
(26) (a) Silverstein, R. M.; Bassler, G. C.; Morrill, T. C. Spectrometric
Identification of Organic Compounds, 4th ed.; John Wiley & Sons: New York,
1981. (b) Narayanan, C. R.; Sarma, M. R. Tetrahedron Lett. 1968, 13, 1553–
1556. (examples of proton deshielding by a γ-carbonyl group).
3438 J. Org. Chem. Vol. 73, No. 9, 2008

Total Synthesis of (()-Hyphodermins A and D
SCHEME 5. Synthetic Approach to Hyphodermin A (1)
following additional options: symmetry ) off; Grid density ) fine;
Diffuse ) none; Accuracy level ) accurate. The final geometries
were characterized as an energy minimum on the potential energy
surface by the absence of any negative (imaginary) vibrational
frequencies at the stationary point.
7-Bromo-5,5-dimethyl-8-oxo-5,6,7,8-tetrahydronaphthalene-
1,2-dicarboxylic Acid Dimethyl Ester (12). Diester 112 (2.0 g,
6.9 mmol), copper(II) bromide (3.02 g, 13.8 mmol), and MeOH
(25 mL) were heated at reflux for 16 h. A white precipitate formed.
The suspension was filtered (celite), and the solvent removed in
vacuo. Analysis of the brown foam (2.28 g) by 1H NMR
spectroscopy showed bromodimethyl ester 12 (1.98 g, 78%) in
>95% purity. Purification by silica gel chromatography (EtOAc:
hexane; 1:1) gave bromodiester 12 (1.91 g, 75%) as a yellow
1
powder. Mp 137–140 °C. H NMR (200 MHz, CDCl3) δ 1.44 (s,
3H), 1.51 (s, 3H), 2.45–2.65 (m, 2H), 3.91 (s, 3H), 4.03 (s, 3H),
5.08 (dd, 1H, J ) 7.6, 11.3 Hz), 7.56 (d, 1H, J ) 8.4 Hz), 8.21 (d,
1H, J ) 8.4 Hz). 13C NMR (100 MHz, CDCl3) δ 29.9, 30.5, 37.4,
47.8, 49.6, 52.8, 53.0, 127.3, 127.4, 127.6, 135.4, 136.8, 156.3,
164.7, 169.0, 189.5. HRMS ESI (+ve) C16H17O579Br calcd for [M]+
368.0259, found 368.0257.
to give an aldehyde acid followed by ring-closure on the
opposite face (analogous to mutarotation) to give the anti
compound 1.
5,5-Dimethyl-8-oxo-5,8-dihydronaphthalene-1,2-dicarboxylic
Acid Dimethyl Ester (13). Bromo dimethyl ester 12 (1.93 g, 5.2
mmol), DABCO (3.5 g, 31.2 mmol), and MeOH (25 mL) were
heated at reflux for 16 h. The methanol was removed under reduced
pressure. The brown resin was dissolved in CH2Cl2 (25 mL) and
the organic phase was washed with HCl (2 M) (2 × 30 mL), water
(1 × 40 mL), and brine (3 × 25 mL) then dried (MgSO4,
anhydrous) and the solvent was removed in vacuo. The crude
As in the case of hyphodermins C and D, the chemical shift
of H1 would be expected to be more deshielded by the ketone
carbonyl in the syn structure 23 than in the anti structure 1. For
comparison, the chemical shift of H1 in 3-hydroxy-1(3H)-
isobenzofuranone is δ 6.65 ppm28 whereas in hyphodermin A
it is δ 6.97 ppm, a downfield shift of 0.32 ppm. The analogous
H1 downfield shift observed for hyphodermin D (syn, (()-3)
versus 3-methoxy-1(3H)-isobenzofuranone was 0.49 ppm, whereas
for hyphodermin C (anti, (()-4) it was 0.18 ppm. A downfield
shift of 0.32 ppm for hyphodermin A is between these two
values, so in the absence of the second “anomer”, an unequivo-
1
product was obtained as a brown resin (1.48 g). Analysis by H
NMR spectroscopy showed enone diester 13 (1.98 g, 78%) (1.20
g, 81%) in >90% purity. Purification by silica gel chromatography
(EtOAc:hexane; 1:1) gave pure 13 as a resin. 1H NMR (200 MHz,
CDCl3) δ 1.51 (s, 6H), 3.92 (s, 3H), 4.06 (s, 3H), 6.36 (d, 1H, J )
10 Hz), 6.90 (d, 1H, J ) 10 Hz), 7.68 (d, 1H, J ) 8.4 Hz), 8.24 (d,
1H, J ) 8.4 Hz). 13C NMR (100 MHz, CDCl3) δ 29.7, 38.4, 52.9,
53.0, 126.6, 126.9, 127.9, 128.4, 133.8, 136.0, 155.0, 156.5, 165.1,
169.9, 183.3. HRMS ESI (+ve) C16H16O5 calcd for [M]+ 288.0998,
found 288.0997.
1
cal assignment based on H NMR is not possible. Ab initio
calculations on structures 1 and 23 indicate that the energy
difference between them is only 5 kJ/mol, with the anti structure
1 being the more stable of the two. This provides some evidence
for the original assignment of hyphodermin A as structure 1.
In summary, NMR studies suggest that synthetic hyphodermin
C has an anti arrangement and synthetic hyphodermin D a syn
arrangement of the epoxide and O-Me groups, respectively,
which is the opposite of that reported previously. The stereo-
chemistry of synthetic hyphodermin A also remains in doubt.
X-ray structures would be required to unequivocally prove the
relative stereochemistry of synthetic 1/23, 3, and 4. The total
synthesis of synthetic hyphodermins D ((-3) and A (1/23) was
achieved from 4,4-dimethylcyclohexan-1,3-dione in 11 steps and
12 steps, respectively. With larger quantities of synthetic
hyphodermins A and D (1/23 and (()-3) now in hand, further
investigations into the biological activity of these compounds
and their simple derivatives will be carried out.
5,5-Dimethyl-8-oxo-5,8-dihydronaphthalene-1,2-dicarboxylic
Acid (14). Dimethyl ester 13 (733 mg, 2.55 mmol) in NaOH (aq,
2 M, 15 mL) was reacted and worked up as above. Recrystallization
(CH2Cl2/hexane) of the crude yellow foam (496 mg) gave enone
1
diacid 14 as a white powder (385 mg, 58%). Mp 141 °C dec. H
NMR (200 MHz, acetone-d6) δ 1.55 (s, 6H), 6.28 (d, 1H, J ) 10.4
Hz), 7.09 (d, 1H, J ) 9.8 Hz), 7.93 (d, 1H, J ) 8.4 Hz), 8.23 (d,
1H, J ) 8.6 Hz), 2 × COOH not observed. 13C NMR (100 MHz,
acetone-d6) δ 29.7, 39.1, 126.8, 128.3, 128.6, 134.4, 134.5, 137.5,
155.7, 157.6, 166.5, 169.8, 183.5. Anal. Calcd for C14H12O5.2H2O:
C 56.76; H 5.44. Found: C 56.86, H 5.27. HRMS ESI (+ve)
C14H12O5Na calcd for [M + Na]+ 283.0582, found 283.0577.
1-Hydroxy-6,6-dimethylnaphtho[1,2-c]furan-3,9-(1H,6H)-di-
one (15). (i) Enone Anhydride 6. (i) Enone Anhydride 6. Acetic
anhydride (1 M in dichloromethane, 2 mL, 3.0 mmol) was added
to a solution of dicarboxylic acid 14 (702 mg, 2.7 mmol) in THF
(20 mL) and the solution was stirred under nitrogen atmosphere at
50 °C for 16 h. The solvent was removed in vacuo to give a yellow/
brown powder. Recrystallization of the powder from THF/hexane
gave 6,6-dimethyl-6H-naphtho[1,2-c]furan-1,3,9-trione 6 (555 mg,
85%) as an insoluble yellow powder that could not be purified
further, and was used directly in the next step. The 1H NMR
Experimental Section
Density Functional Theory Calculations. All hybrid DFT
calculations were performed with Jaguar 7.029 on a dual-core SGI
Tezro running IRIX 6.5. Equilibrium geometries of compounds 3,
4, 1, and 23 and NMR shielding constants of compounds 3 and 4
were calculated at the B3LYP/6-31G** (6d) level following
preoptimization of the global minimum energy structure located
by using a Macromodel 9.530 MMFFs conformational search.
Default Jaguar optimization parameters were employed with the
1
spectrum was consistent with structure 6. H NMR (200 MHz,
acetone-d6 + methanol-d4) δ 1.62 (s, 6H), 6.38 (d, 1H, J ) 9.8
Hz), 7.16 (d, 1H, J ) 10 Hz), 7.99 (d, 1H, J ) 8.6 Hz), 8.05 (d,
1H, J ) 8.2 Hz). (ii) Lactol 15. Compound 6 (307 mg, 1.21 mmol)
was added to a solution of LiAlH(t-BuO)3 (295 mg, 1.21 mmol) in
THF (10 mL) at 0 °C. The solution was stirred at 0 °C for 4 h
under nitrogen atmosphere. Ammonium chloride (saturated, 5 mL)
and HCl (2 M, 10 mL) were added and the aqueous phase extracted
(28) Navaro, M.; De Ciovani, W. F.; Romero, J. R. Tetrahedron 1991, 47,
851–857.
(29) Jaguar, version 7.0; Schrödinger, LLC: New York, NY, 2007.
(30) Macromodel, version 9.5; Schrödinger, LLC: New York, NY, 2007.
J. Org. Chem. Vol. 73, No. 9, 2008 3439

Loughlin et al.
with CH2Cl2 (3 × 30 mL). The combined organic layers were
washed with brine (2 × 40 mL) then dried (MgSO4, anhyd.) and
solvent was removed in vacuo. The crude product was obtained as
a brown solid (222 mg). Purification by silica gel chromatography
(EtOAc:hexane; 1:1) gave lactol 15 (112 mg, 38%) as colorless
ether solution, 1 M, 5 drops). The solution was stirred at rt under
nitrogen for 16 h. The solvent, excess 3,4-dihydropyran, and HCl
were removed in vacuo to give a tacky resin (62 mg). Purification
via silica gel chromatography (ethyl acetate:hexane 1:1) gave THP
ether 21 as a resin (48 mg, 91%), which was unstable and was
used directly in the next step. The 1H NMR spectrum was consistent
1
crystals. Mp 179–181 °C. H NMR (200 MHz, CDCl3) δ 1.57 (s,
1
6H), 6.48 (d, 1H, J ) 10.2 Hz), 7.04 (s, 1H), 7.08 (d, 1H, J ) 10.2
Hz), 7.84 (d, 1H, J ) 8 Hz), 8.08 (d, 1H, J ) 8.4 Hz), OH not
observed. 13C NMR (50 MHz, CDCl3) δ 29.0, 39.0, 97.5, 126.5,
126.7, 129.1, 130.0, 132.0, 148.3, 157.0, 159.2, 167.9, 185.7. HRMS
ESI (-ve) C14H12O4 calcd for [M - H]- 243.0657, found 243.0656.
1-Methoxy-6,6-dimethylnaphtho[1,2-c]furan-3,9(1H,6H)-di-
one (19). Acetyl chloride (0.2 mL, 221 mg, 2.8 mmol) was added
to MeOH (6 mL) in the presence of a 4 Å molecular sieve. The
solution was stirred for 20 min at rt under a nitrogen atmosphere.
A solution of unsaturated lactol 15 (98 mg, 0.4 mmol) in MeOH
(3 mL) was added and the solution stirred at rt for 2 h. The
molecular sieve was removed and the solvent was removed in
with that of 21. H NMR (200 MHz, CDCl3) δ 1.55 (s, 3H), 1.57
(s, 3H), 1.59–1.91 (m, 6H), 3.6 -3.70 (m, 2H), 5.35 (m, 1H), 6.40
(d, 1H, J ) 10.2 Hz), 6.98 (d, 1H, J ) 9.2 Hz), 7.03 (s, 1H), 7.82
(d, 1H, J ) 8 Hz), 8.06 (d, 1H, J ) 7.8 Hz). (ii) (()-Hyphodermin
A ((()-1) via THP Epoxide 22. DBU (22 mg, 22 µL, 1 equiv),
tert-butyl hydroperoxide (0.22 mmol, 1.5 equiv), and THP lactol
21 (48 mg, 0.146 mmol) in CH2Cl2 (4 mL) were reacted and worked
up as above. Crude THP ether 22 was obtained as a pale yellow
oil (44 mg), which rapidly decomposed upon standing or attempted
purification by column chromatography, and was used directly. The
yellow oil (44 mg) was dissolved in THF/HCl (5% w/w) (7:3) (10
mL) and the solution was stirred at rt for 5 h. The solution was
worked up as above. The crude product was obtained as a pale
yellow oil (44 mg). Purification by repeated silica gel chromatog-
raphy (EtOAc:hexane; 1:1) gave impure (()-hyphodermin A ((()-
1), 9 mg, 24%, 2 steps from THP ether 21 in ∼70% purity). The
1H and 13C NMR spectra of (()-1 matched the reported values.1
1H NMR (400 MHz, 95:5 d4-methanol:CDCl3) δ 1.36 (s, 3H), 1.76
(s, 3H), 3.63 (d, 1H, J ) 4.4 Hz), 3.75 (d, 1H, J ) 4 Hz), 6.97 (s,
1
vacuo. Analysis of the crude yellow powder by H NMR spec-
troscopy showed methyl “acetal” 19 (93 mg, 90%) in >97% purity.
1
Mp 156 °C. H NMR (400 MHz, CDCl3) δ 1.54 (s, 6H), 3.82 (s,
3H), 6.42 (d, 1H, J ) 10.4 Hz), 6.79 (s, 1H), 6.97 (d, 1H, J ) 10
Hz), 7.82 (d, 1H, J ) 8.4 Hz), 8.03 (d, 1H, J ) 8.4 Hz). 13C NMR
(100 MHz, CDCl3) δ 29.5, 30.0, 38.7, 59.1, 105.1, 126.7, 126.9,
127.2, 128.5, 130.2, 145.2, 156.6, 157.4, 168.3, 183.8. HRMS ESI
(+ve) C15H14O4Na calcd for [M + Na]+ 281.0790, found 281.0781.
Hyphodermin D ((()-3). tert-Butyl hydroperoxide (69 mg, 1.5
equiv) and DBU (0.18 mL, 183 mg, 1.5 equiv) were added to
methyl acetal 19 (75 mg, 0.3 mmol) in CH2Cl2 (5 mL), and the
solution was stirred at rt for 16 h under nitrogen. The organic phase
was washed with ice-cold HCl (2 M, 20 mL) and worked up as
above. The crude product was obtained as a yellow oil (67 mg).
Analysis of the crude oil by 1H NMR spectroscopy showed
incomplete conversion of 19. The crude oil was treated with tert-
butyl hydroperoxide (69 mg, 1.5 equiv) and DBU (0.18 mL, 183
mg, 1.5 equiv) as above and worked up as above. The crude product
was obtained as a yellow oil (70 mg). Trace peaks for hyphodermin
1
1H), 7.67 (d, 1H, J ) 8 Hz), 8.04 (d, 1H, J ) 8.4 Hz). H NMR
(600 MHz, methanol-d4) δ 1.35 (s, 3H), 1.73 (s, 3H), 3.74 (d, 1H,
J ) 4.2 Hz), 3.75 (d, 1H, J ) 4.2 Hz), 6.97 (s, 1H), 7.86 (d, 1H,
J ) 7.8 Hz), 8.04 (d, 1H, J ) 7.8 Hz). 13C NMR (150 MHz,
methanol-d4) δ 24.9, 28.7, 37.7, 55.7, 63.1, 103.0, 126.1, 126.7,
129.3, 130.3, 147.4, 155.5, 170.5, 194.9. HRMS ESI (+ve)
C14H13O5 calcd for [M + H]+ 261.0763, found 261.0759.
Acknowledgment. Financial support for this work was
provided by Griffith University as well as Natural Product
Discovery and Eskitis Institute for Cell and Molecular Therapies
at Griffith University. Award of an APAWS to L. C. Henderson.
1
C ((()-4) in the H NMR spectrum of the crude product were in
1
1
agreement with the reported values. Hyphodermin C ((()-4): H
NMR (200 MHz, CDCl3) δ 1.39 (s, 3H), 1.76 (s, 3H), 3.61 (d, 1H,
J ) 4 Hz), 3.74 (d, 1H, J ) 4.4 Hz), 3.82 (s, 3H), 6.43 (s, 1H),
7.71 (d, 1H, J ) 8.4 Hz), 8.06 (d, 1H, J ) 8.4 Hz). Purification to
95% purity by repeated silica gel chromatography (CH2Cl2:Et3N,
99:1) gave hyphodermin D ((()-3) as a colorless oil (25 mg, 29%).
The 1H and 13C NMR spectra of (()-3 were in agreement with the
Supporting Information Available: Additional experimental
procedures; MS and FTIR data for (()-1, (()-3, 6, 8–10, and
1
12–21; H NMR spectra for 8–10, 12–21, (()-1, and (()-3;
table comparing 1H NMR and 13C NMR data (where available)1
of natural and synthetic hyphodermins A, C, and D (1, 4, and
3); representation of energy minimized structure, Cartesian
coordinates, and computed total energy for 5; Cartesian
coordinates, computed total energy, and selected dihedral angles,
and NMR shielding constants for 3 and 4; representation of
energy minimized structure, Cartesian coordinates, and com-
puted total energy for 1 and 23; data collection, structure
solution, refinement, and crystal data for 9 and 17; and an
ORTEP plot of 9 and 17. This material is available free of charge
via the Internet at http://pubs.acs.org.
1
1
reported values. (()-3: H NMR (200 MHz, CDCl3) δ 1.35 (s,
3H), 1.74 (s, 3H), 3.62 (d, 1H, J ) 4 Hz), 3.70 (s, 3H), 3.77 (d,
1H, J ) 4.4 Hz), 6.74 (s, 1H), 7.66 (d, 1H, J ) 8 Hz), 8.02 (d, 1H,
J ) 8.4 Hz). 13C NMR (100 MHz, CDCl3) δ 25.7, 30.1, 37.5, 55.9,
58.9, 63.0, 104.0, 125.6, 127.1, 129.79, 129.81, 145.5, 154.0, 167.9,
193.6. HRMS ESI (+ve) C15H15O5 calcd for [M + H]+ 275.0920,
found 275.0916.
Hyphodermin A ((()-1). (i) 6,6-Dimethyl-1-(tetrahydro-2H-
pyran-2-yloxy)naphtho[2,1-c]furan-3,9(1H,6H)-dione (21). 3,4-
Dihydropyran (26 mg, 28 µmol) was added to a solution of
unsaturated lactol 15 (39 mg, 0.16 mmol) followed by HCl (diethyl
JO800227P
3440 J. Org. Chem. Vol. 73, No. 9, 2008