Total synthesis and structural confirmation of (±)-cuevaene A

Article abstract of DOI:10.1016/j.tetlet.2012.07.121

The total synthesis and structural reassignment of cuevaene A have been completed. The key synthetic steps in the total synthesis included a base-promoted double conjugate addition and further elaboration to generate the tricyclic core structure, followed by construction of the trienoic acid side chain. Detailed comparison of proton and carbon NMR data with published values enabled the connectivity of the natural product, which had been debated in earlier publications, to be confirmed.

Full text of DOI:10.1016/j.tetlet.2012.07.121

Tetrahedron Letters 53 (2012) 5422–5425
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Total synthesis and structural confirmation of ( )-cuevaene A
⇑
Philip G. E. Craven, Richard J. K. Taylor
Department of Chemistry, University of York, Heslington, York, YO10 5DD, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
The total synthesis and structural reassignment of cuevaene A have been completed. The key synthetic
steps in the total synthesis included a base-promoted double conjugate addition and further elaboration
to generate the tricyclic core structure, followed by construction of the trienoic acid side chain. Detailed
comparison of proton and carbon NMR data with published values enabled the connectivity of the natural
product, which had been debated in earlier publications, to be confirmed.
Received 18 June 2012
Revised 13 July 2012
Accepted 27 July 2012
Available online 10 August 2012
Ó 2012 Elsevier Ltd. All rights reserved.
Keywords:
Cuevaene A
Total synthesis
Structure confirmation
Tandem oxidation process (TOP)
Cuevaenes A and B are triene natural products isolated from
Streptomyces sp. HKI 0180 in 2000 by Gräfe et al.1 and shown to
possess moderate antibacterial activity against Gram-positive bac-
teria (Fig. 1). Using a combination of mass spectrometry and 1D
and 2D NMR spectroscopy, Gräfe’s group assigned structures 1
and 2, respectively, to cuevaenes A and B.
In 2009, two structurally related natural products, JBIR-23 (3)
and 24 (4), were isolated from the same species of bacteria by
Shin-ya and co-workers.2 Trienes 3 and 4 show extremely promis-
ing activity against an asbestos-caused lung cancer, malignant
pleural mesothelioma.3 Natural products 1–4 all share the same
4-methoxy-6-methylheptatrienoic acid side chain, which appears
to be a previously undescribed natural product structural motif.
Given the similarity of the tricyclic core system in cuevaene A,
JBIR-23 (3) and 24 (4), and the fact that all of the compounds were
isolated from the same strain of bacteria, Shin-ya’s group
suggested that these four natural products may share the same
biosynthetic pathway. Moreover, they proposed that, if this was
indeed the case, the structures of 1 and 2, initially proposed by
Gräfe et al., may be incorrect and the alternative structures 5 and
6, in which the positioning of the side chains on the tricyclic core
system matches that seen in 3 and 4, are more likely to be correct
(Fig. 1).
To confirm the Shin-ya hypothesis, in 2010, Liu et al. synthesised
both 1 and 5 in an attempt to confirm the structure of cuevaene A
but, surprisingly, they found that the proton and carbon NMR spec-
OH
OH
OH
CO2H
CO2H
CO2H
O
OH
OH
MeO
OH
OH
MeO
MeO
O
OH
OH
H
H
OMe
OMe
O
H
O
H H
O
OH
OR
CO2H
CO2H
cuevaene B
(original structure) 2
cuevaene A
(alternative structure) 5
cuevaene B
(alternative structure) 6
3
JBIR-23, R = Me,
JBIR-24, R = H, 4
cuevaene A
(original structure) 1
Figure 1. Structures of cuevaenes and JBIR compounds.
⇑
Corresponding author. Tel.: +44 1904 322606; fax: +44 1904 324523.
E-mail address: richard.taylor@york.ac.uk (R.J.K. Taylor).
0040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2012.07.121

P. G. E. Craven, R. J. K. Taylor / Tetrahedron Letters 53 (2012) 5422–5425
5423
OH
O
O
CO2H
+
O
OTMS
MeO
8
10
9
OTBDMS
O
OH
OH
O
O
O
ref. 6
O
O
+
5
7
O
O
O
O
11
12
13
Scheme 1.
troscopic data of both their synthetic samples failed to match those
reported for the natural product.4 Having an interest in the prepa-
ration of JBIR analogues, we decided to first solve the ambiguity
concerning the structure of cuevaene A. For reasons outlined later,
we felt that structure 5 proposed by Shin-ya was most likely to be
correct and therefore embarked on its re-synthesis (Scheme 1).
Liu’s synthetic route utilised the tricyclic aldehyde 7 as the cor-
nerstone synthetic intermediate then elaborating the trienoic acid
side chain using sequential olefination reactions (Scheme 1). Alde-
hyde 7 was accessed from alkene 8, itself prepared from the
known5 Lewis acid promoted condensation between silyl enol
ether 9 and p-benzoquinone (10). Our synthetic plan utilised the
same aldehyde intermediate 7 and a similar, but more streamlined,
end-game to that described by Liu et al. However, we envisioned
that aldehyde 7 could be accessed via a homologation reaction of
known ketone 11 (Scheme 1).6 Ketone 11 is readily accessed from
commercially-available cyclohexa-1,3-dione (12) and quinone
monoketal 13 via a double conjugate addition process followed
by acidic rearrangement.6
Quinone monoketal 13 was readily prepared by dearomatisa-
tion of 4-methoxyphenol (14) using phenyliodonium trifluoroace-
tate (PIFA), by modification of a known procedure.7 The published
procedure7 uses CH2Cl2 as the solvent but, on a larger scale, signif-
icant amounts of p-benzoquinone were formed. However, we
found that switching the solvent to ethylene glycol, with a mini-
mum amount of dichloromethane to aid dissolution of phenol 14,
enabled the reaction to be carried out efficiently on a larger scale
(30 mmol) (Scheme 2). Following literature procedures,6
compound 13 was then converted into ketone 11 in excellent yield
via a BuOK-catalysed double conjugate addition followed by an
acidic rearrangement of intermediate 15 using 6 M hydrochloric
acid.
t
We next planned to homologate ketone 11 by an olefination/
hydroboration/oxidation sequence. Unfortunately, standard Wittig
chemistry (e.g. Ph3PCH3+BrÀ, nBuLi) was unsuccessful, but eventu-
ally methylenation was accomplished using the commercially
available Peterson reagent, trimethylsilylmethyllithium, albeit in
a moderate yield. It should be noted that silylation of compound
11 prior to olefination could not be achieved, whereas the analo-
gous protection of phenol 16 to give the TBDMS silyl ether 17
was carried out in good yield. Subsequent hydroboration using
BH3.THF followed by treatment with sodium perborate8 gave alco-
hol 18 in a good yield (traditional work-up conditions, such as
NaOH/H2O2, caused deprotection of the silyl ether; direct oxidation
of the intermediate borane to aldehyde 7 was unsuccessful). Subse-
quent oxidation of compound 18 using the Swern procedure gave
the desired aldehyde 7 in 39% yield over 7 steps from phenol 14.
Compound 7, prepared by a different route, was also employed
in the study by Liu and co-workers4
The completion of the synthesis employed a sequential olefin-
ation sequence similar to that developed by Liu’s group,4 although
a number of improvements were introduced (Scheme 3). Firstly,
aldehyde 7 was treated with ethyl 2-(triphenylphosphoranylid-
ene)propanoate to yield the desired unsaturated ester 19 in good
yield and exclusively as the E-isomer (the NMR data corresponded
closely to those reported by Liu and co-workers4). While Liu’s
OH
O
12, tBuOK
PIFA, CH2Cl2
O
6 M HCl
tBuOH
O
OH
O
1,4-dioxane
HO
O
5 d, rt
30 min, rt
91%
16 h
50 °C
O
O
95%
O
OMe
(over two steps)
15
14
13
OTBDMS
OH
OH
TBDMSCl
O
TMSCH2Li
imid, CH2Cl2
THF, 20 h
-78 °C to rt
55%
rt, 6 h
89%
O
O
O
O
17
11
16
(COCl)2, DMSO
Et3N, CH2Cl2
OTBDMS
OTBDMS
HO
O
BH3.THF, THF
rt, 2 h
then NaBO3.4H2O
THF:H2O, rt, 12 h
97%
-78 °C to rt, 2 h
72%
O
18
7
Scheme 2.

5424
P. G. E. Craven, R. J. K. Taylor / Tetrahedron Letters 53 (2012) 5422–5425
HO
CO2Et
OTBDMS
O
OTBDMS
OTBDMS
DIBAL-H
Ph3P=C(Me)CO2Et
CH2Cl2, rt, 18 h
82%
toluene
rt, 2 h
81%
O
O
19
O
20
7
MeO
22
CO2Me
CO2Me
O
P(O)(OMe)2
MeO
OTBDMS
MnO2
KHMDS, 18-C-6
OTBDMS
CH2Cl2, rt, 16 h
90%
THF, 0 °C to rt, 2 h
81%
O
21
O
23
CO2Me
HO
MnO2
Ph3P=CHCO2Me
MeO
LiBH4, THF
MeO
OTBDMS
OTBDMS
CH2Cl2, rt, 18 h
23
0 °C to rt, 3 h
61% (from
)
O
24
O
25
CO2H
LiOH
1,4-dioxane:H2O
rt, 2 h
MeO
OH
60%
O
5
Scheme 3.
Table 1
Comparison of NMR spectroscopic data of cuevaene A (5) (all values are in ppm with respect to TMS)
1
CO2H
3
5
21 MeO
20
OH
18
7
19
14
13
9
17
16
8
15
10
O
12
11
5
Cuevaene A1,11
Liu’s synthetic 54
Our synthetic 5
Proton
16
3
19
17
2
CD3OD (300 MHz)
7.15 (d, 8.7)
7.07 (d, 15.4)
6.69 (d, 2.4)
6.63 (dd, 2.4, 8.7)
5.99 (d, 15.4)
5.91 (s)
CDCl3 (400 MHz)
7.23 (d, 8.0)
7.12 (d, 15.2)
6.69 (s)
6.68 (d, 8.4)
6.04 (d, 15.6)
5.84 (s)
CD3OD (400 MHz)
7.15 (d, 8.7)
7.07 (d, 15.4)
6.69 (d, 2.4)
6.63 (dd, 2.4, 8.7)
5.99 (d, 15.4)
5.91 (s)
CDCl3 (400 MHz)
7.22 (d, 8.3)
7.11 (d, 15.4)
6.70 (d, 2.6)
6.68 (dd, 2.6, 8.3)
6.03 (d, 15.4)
5.83 (s)
5
7
8
21
11
20
9,10
5.80 (d, 10.0)
3.80–3.91 (m)
3.65 (s)
2.66–2.74 (m)
2.24 (s)
1.96–2.15 (m)
1.81–1.95 (m)
1.50–1.66 (m)
5.77 (d, 10.0)
3.82–3.83 (m)
3.66 (s)
2.72–2.73 (m)
2.22 (s)
1.98–2.11 (m)
1.82–1.89 (m)
1.52–1.60 (m)
5.79 (d, 10.1)
3.84 (ddd, 5.7, 7.7, 10.1)
3.65 (s)
2.67–2.74 (m)
2.24 (s)
1.96–2.14 (m)
1.81–1.93 (m)
1.50–1.63 (m)
5.76 (d, 9.9)
3.78–3.87 (m)
3.65 (s)
2.68–2.76 (m)
2.21 (s)
1.95–2.11 (m)
1.80–1.92 (m)
1.50–1.61 (m)

P. G. E. Craven, R. J. K. Taylor / Tetrahedron Letters 53 (2012) 5422–5425
5425
group used LiBH4 to carry out the reduction of ester 19, we
achieved better yields using DIBAL-H as the reducing agent.
Subsequent oxidation of the resulting alcohol 20 to aldehyde 21
was then effected, in a good yield, using MnO2. A Horner–
Wadsworth–Emmons reaction with phosphonate 229 afforded
the desired diene 23 in a good yield, again with complete stereose-
lectivity. Liu et al. found it necessary to reprotect the phenol after
this step, but we found that by using only one equivalent of base,
the need for reprotection could be avoided while maintaining the
yield. LiBH4 reduction of methyl ester 23 gave the desired alcohol
24 in essentially quantitative yield, although it proved to be unsta-
ble as a solution in chloroform. The final alkene was then installed
using a tandem oxidation process (TOP),10 using manganese
dioxide in the presence of methyl (triphenylphosphoranylid-
ene)acetate, to afford the desired triene 25 as a single stereoisomer
and in good yield. The final deprotection and hydrolysis were
carried out using Liu’s LiOH procedure (60%) giving Shin-ya’s pro-
posed structure of cuevaene A, (5). Our overall yield of cuevaene A
(5) from phenol 14 is 7% over 14 steps (as compared to Liu’s route
which gave 5% yield over 13 steps).4
H-5 and H-7, confirmed the geometries of the two trisubstituted
alkenes.
In conclusion, we have successfully synthesised Shin-ya’s pro-
posed structure of cuevaene A (5)2 from 4-methoxyphenol (14)
in 14 steps and 7% overall yield. In addition, we have confirmed
the true structure of cuevaene A (5), correcting the initial mis-
assignment1 and subsequent structural uncertainty.4 Furthermore,
by extrapolation, cuevaene B should be reassigned as structure 6,2
to match the connectivity of cuevaene A (5). We are currently
applying this methodology to prepare structurally related natural
products, JBIR-23 (3) and 24 (4).
Acknowledgements
We are grateful to the University of York and Elsevier for post-
graduate support (P.G.E.C.) and to the Society of Chemical Industry
for additional Scholarship funding.
Supplementary data
Having prepared the alternative structure 5 for cuevaene A as
proposed by Shin-ya et al.,2 we were in a position to clarify the cor-
rect structural assignment. At this point, we contacted Professor
Gräfe’s group and they kindly sent us the original data from the iso-
lated natural products.11 However, on inspection of the original
data, we realised that an error had been made in reporting the 1H
NMR data for cuevaene A in the original publication;1 the original
1H NMR data was actually collected in CD3OD, but in the publication
the solvent was reported to be CDCl3. We therefore collected the 1H
NMR data for compound 5 in both CD3OD and CDCl3 and compared
these with the data reported by Gräfe1,11 and Liu4 (Table 1).
Supplementary data (general procedures and spectral data)
associated with this article can be found, in the online version, at
http://dx.doi.org/10.1016/j.tetlet.2012.07.121.
References and notes
1. Schlegel, B.; Groth, I.; Gräfe, U. J. Antibiot. 2000, 53, 417–425.
2. Motohashi, K.; Hwang, J.-H.; Sekido, Y.; Takagi, M.; Shin-ya, K. Org. Lett. 2009,
11, 285–288.
3. Hwang, J.-H.; Takagi, M.; Murakami, H.; Sekido, Y.; Shin-ya, K. Cancer Lett. 2011,
300, 189–196.
4. Chen, Y.; Huang, J.; Liu, B. Tetrahedron Lett. 2010, 51, 4655–4657.
5. Saraswathy, V. G.; Sankararaman, S. J. J. Org. Chem. 1995, 60, 5024–5028.
6. Duthaler, R. O.; Wegmann, U. H.-U. Helv. Chim. Acta 1984, 67, 1755–1766.
7. Tran-Huu-Dau, M.-E.; Wartchow, R.; Winterfeldt, E.; Wong, Y.-S. Chem Eur. J.
2001, 7, 2349–2369.
8. Weiss, M. E.; Carreira, E. M. Angew. Chem., Int. Ed. 2011, 50, 11501–11505.
9. Lu, H.; Su, Z.; Song, L.; Mariano, P. S. J. Org. Chem. 2002, 67, 3525–3528.
10. Wei, X.; Taylor, R. J. K. Tetrahedron Lett. 1998, 39, 3815–3818; Taylor, R. J. K.;
Reid, M.; Foot, J.; Raw, S. A. Acc. Chem. Res. 2005, 38, 851–869.
11. Original unprocessed NMR spectroscopic data was supplied courtesy of Dr.
Michael Ramm, Leibniz Institute for Natural Product Research and Infection
Biology, and was processed using MestReC software, to be consistent with the
spectroscopic data for our synthetic compounds. The data shown for cuevaene
A in this paper therefore differs slightly from the published data (Ref.1).
12. The full 13C NMR data is given in the Supplementary data section.
As can be seen, the 1H NMR data of our synthetic compound 5
corresponded extremely well to those reported by Gräfe’s group
in deuterated methanol,1,11 and by Liu and co-workers in CDCl3.4
The doubt as to the correct structure of cuevaene A (5) expressed
by Liu et al. was obviously due to the original misreporting of
the NMR solvent. This conclusion is also supported by a compari-
son of the 13C NMR spectroscopic data of 5 with that of the natural
product which show a very close match, {e.g. (CD3OD, 100 MHz)
170.5 (C-1; Lit.1 170.5), 153.5 (C-4; Lit.1 153.4) and 33.6 (C-8;
Lit.1 33.5)}.12 The side-chain positioning was further confirmed
by observed HMBC interactions between C-12 and H-11 and C-13
and H-7. In addition, NOE interactions between H-3 and H-5, and