The total synthesis of (-)-nitidasin

Article abstract of DOI:10.1002/anie.201403605

Nitidasin is a pentacyclic sesterterpenoid with a rare 5-8-6-5 carbon skeleton that was isolated from the Peruvian folk medicine " Hercampuri". It belongs to a small class of sesterterpenoids that feature an isopropyl trans-hydrindane moiety fused to a variety of other ring systems. As a first installment of our general approach toward these natural products, we report the total synthesis of the title compound. Our stereoselective, convergent route involves the addition of a complex alkenyl lithium compound to a trans-hydrindanone, followed by chemoselective epoxidation, ring-closing olefin metathesis, and redox adjustment.

Full text of DOI:10.1002/anie.201403605

Angewandte
Chemie
DOI: 10.1002/anie.201403605
Natural Product Synthesis
The Total Synthesis of (À)-Nitidasin**
Daniel T. Hog, Florian M. E. Huber, Peter Mayer, and Dirk Trauner*
Dedicated to Clayton Heathcock
Abstract: Nitidasin is a pentacyclic sesterterpenoid with a rare
5
-8-6-5 carbon skeleton that was isolated from the Peruvian
folk medicine “Hercampuri”. It belongs to a small class of
sesterterpenoids that feature an isopropyl trans-hydrindane
moiety fused to a variety of other ring systems. As a first
installment of our general approach toward these natural
products, we report the total synthesis of the title compound.
Our stereoselective, convergent route involves the addition of
a complex alkenyl lithium compound to a trans-hydrindanone,
followed by chemoselective epoxidation, ring-closing olefin
metathesis, and redox adjustment.
T
he Peruvian herbal infusion “Hercampuri”, derived from
Gentianella nitida and Gentianella alborosea, has been used
since ancient times as a remedy against hepatitis, diabetes,
Figure 1. Sesterterpenoids isolated from “Hercampuri” and related
compounds.
[
1]
and hypertension. Investigations into its chemical constitu-
ents have revealed a variety of bitter phenolic compounds
that occur together with a limited number of terpenoids
(
Figure 1). The most complex of these, the sesterterpenoid
a limited number of synthetic studies have been reported and
no total synthesis of a natural product shown in Figure 1 has
been disclosed to date.
[
2]
nitidasin, features a rare 5-8-6-5 carbon framework, which
includes ten stereogenic carbons. The relative configuration
of this intriguing natural product was established by extensive
NMR studies and X-ray diffraction. Nitiol is a simpler
congener that has only undergone partial cyclization,
whereas alborosin appears to be an oxidative degradation
[
9]
Several years ago, we set out to develop a general
synthetic approach toward isopropyl trans-hydrindane sester-
[3]
[10,11]
terpenoids.
We identified hydrindane 3 as a key building
block that could be used in the synthesis not only of nitidasin,
but also of astellatol, alborosin, and the YW compounds
(Scheme 1). Starting from ketone 1, which is readily available
[
4]
product of nitidasin.
Nitidasin can also be classified as a sesterterpenoid that
features an oxygenated trans-hydrindane moiety, the five-
membered ring of which bears an isopropyl substituent cis to
an angular methyl group. Compounds of this type include the
[
12]
through asymmetric Robinson annulation, we were able to
obtain hydrindanone 2 by capitalizing on a trans-selective
[5]
[6]
unusual fungal metabolite astellatol, as well as YW-3699
[
7]
and YW-3548, two inhibitors of mammalian GPI-anchor
biosynthesis (Figure 1). Owing to their complex ring systems
and oxygenation patterns, these natural products pose
[
8]
considerable challenges for synthesis. Accordingly, only
[
+]
[+]
[
*] Dr. D. T. Hog, Dipl.-Chem. F. M. E. Huber, Dr. P. Mayer,
Prof. Dr. D. Trauner
Department of Chemistry and Center for Integrated Protein Science
Ludwig-Maximilians-Universitꢀt Mꢁnchen
Butenandtstr. 5–13, 81377 Mꢁnchen (Germany)
E-mail: Dirk.Trauner@cup.uni-muenchen.de
Homepage: http://www.cup.uni-muenchen.de/oc/trauner/
+
[
] These authors contributed equally to this work.
[
**] The authors are indebted to the Fonds der Chemischen Industrie for
financial support (Ph.D. scholarship to D.T.H.). We are grateful to
Giulio Volpin for assistance in NMR measurements and to Dr.
David Barber for insightful discussions.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201403605.
Scheme 1. Preparation of hydrindanone 3 and a retrosynthetic analysis
of nitidasin.
Angew. Chem. Int. Ed. 2014, 53, 8513 –8517
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8513

.
Angewandte
Communications
[
13]
hydrogenation. Unsaturation of the five-membered ring,
conjugate addition, redox adjustment, and hydroboration/
oxidation delivered our key building block 3. Overall, this
route was highly stereoselective and was able to deliver gram
quantities of key ketone 3.
rearrangements proved nonviable, we eventually developed
a reliable and scalable sequence: protection of hydrindane 3
gave the corresponding SEM ether 7. This was followed by
a regio- and diastereoselective Pd-catalyzed allylation per-
[
16,17]
formed according to a protocol developed by Negishi.
By
Our retrosynthetic analysis of nitidasin, which evolved
from more obvious but unsuccessful disconnections, is also
shown in Scheme 1. Eventually, we envisaged tetrasubstituted
alkenyl lithium compound 4 and ketone 5 as suitable synthetic
precursors that would enable a highly convergent synthesis of
the natural product. After stereoselective addition of lithium
species 4 to hydrindanone 5, the resultant tertiary allylic
alcohol 6 would undergo ring-closing metathesis (RCM),
followed by hydrogenation and epoxidation to eventually
afford nitidasin.
contrast, attempts to obtain hydrindanone 8 with standard
alkylation procedures were unsuccessful.
[
18]
Exposure of ketone 8 to K-Selectride resulted in a diaste-
reoselective reduction of the carbonyl group. After Lemieux–
[
19]
Johnson cleavage of the terminal alkene and chromium-
mediated oxidation of the intermediate lactol to lactone 9, the
stage was set for the installation of the C3 methyl group. As
anticipated, reaction of the lithium enolate with MeI occurred
from the convex face of the tricyclic ring system, thereby
giving rise to a single diastereomer 10. Reduction of lactone
[14]
To put this plan into practice, we first investigated the
seemingly straightforward conversion of ketone 3 into trans-
hydrindanone 5 (Scheme 2a). In light of literature precedent,
we were aware that the installation of the two adjacent
stereogenic centers at C2 and C3 (nitidasin numbering) would
10 with LiAlH , followed by double silylation and chemo-
4
selective oxidation under carefully optimized Swern condi-
[20]
tions
provided efficient access to aldehyde 11. Finally,
a three-step protocol consisting of Wittig olefination, chemo-
selective deprotection, and subsequent oxidation gave trans-
hydrindanone 5, which bears five out of the ten stereocenters
[15]
be a challenge.
While initial attempts based on Claisen
[21]
of nitidasin.
The preparation of alkenyl iodide 17, the precursor of
lithium species 4, is outlined in Scheme 2b. Although metalla-
ene reactions were initially considered, they were eventually
ruled out owing to difficulties associated with the required cis
relationship of the two adjacent substituents on the cyclo-
pentane ring. Our successful sequence commenced with
oxidative cleavage of commercially available (À)-citronellene
[
22]
(12) to yield known aldehyde 13.
This was followed by
Corey–Fuchs homologation with in situ trapping of the
intermediary alkynyl lithium species with MeI, which
[
23]
afforded enyne 14.
Next, we employed a Zr-mediated
[
24]
cyclometalation developed by Negishi to install the five-
membered ring and the tetrasubstituted alkenyl iodide. The
reaction of enyne 14 with Cp Zr, generated in situ, and
2
[25]
subsequent quenching with excess NIS provided diiodide
5 as a mixture of diastereomers. Subsequent elimination of
1
the primary iodide with DBU and hydroboration with 9-BBN
gave alcohol 16 with the correct cis relationship of the two
adjacent substituents. At this stage, we determined the ee of
[26]
intermediate 16 through Mosher ester analysis.
This
intermediate was obtained with just 60% ee owing to the
low optical purity of commercially available (À)-citronellene
(12). Swern oxidation of primary alcohol 16 and Wittig
methenylation then yielded iodo diene 17. Epimerization at
the aldehyde stage was not observed under these conditions,
presumably owing to the considerable allylic strain that an
intermediary enolate would encounter.
With enantiomerically pure building block 5 and scalemic
1
7 in hand, we were in a position to investigate the joining of
the two fragments and the formation of the central eight-
Scheme 2. Synthesis of hydrindanone 5 and alkenyl iodide 17.
SEM=2-(tri-methylsilyl)ethoxymethyl, KHMDS=potassium bis(tri-
methylsilyl)amide, K-Selectride=potassium tri-sec-butylborohydride,
PCC=pyridinium chlorochromate, LiHMDS=lithium bis(trimethyl-
silyl)amide, CSA=camphor-10-sulfonic acid, DMP=Dess–Martin per-
iodinane, mCPBA=meta-chloroperbenzoic acid, NIS=N-iodosuccin-
imide, DBU=1,8-diazabicyclo[5.4.0]-undec-7-ene, 9-BBN=9-bora-
bicyclo[3.3.1]nonane.
membered ring. To conserve our valuable trans-hydrindanone
5, however, this was first done using model compound 18,
which was accessible from hydrindane 2 by using the strategy
outlined above (Scheme 3 and the Supporting Information).
Treating iodide 17 with tBuLi at À788C followed by the
gradual addition of ketone 18 afforded tertiary allylic alcohol
19 in good yield. This reaction is remarkable for several
8514
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 8513 –8517

Angewandte
Chemie
zene gave the tetracyclic model compound 21 within two
hours and in almost quantitative yield.
Encouraged by our model studies, we next exposed our
key trans-hydrindanone 5 to alkenyl lithium species 4
(Scheme 4). In agreement with our previous results, the
addition to the carbonyl was found to be highly diastereose-
Scheme 3. Model studies: synthesis of the tetracarbocyclic nitidasin
core 21.
reasons: First, tetrasubstituted alkenyl lithium species are
rarely used in synthesis. To our knowledge, additions of these
organometallic species to ketones, particularly sterically
[
27]
hindered ones, such as 18, are unprecedented. Second, the
addition was highly stereoselective with respect to the trans-
hydrindanone. Finally, the diastereomeric mixture of 10:1
obtained under these reaction conditions reflects a kinetic
resolution of the scalemic lithium species 4 (e.r. = 4:1).
Having successfully linked the two fragments, we next
turned to the closure of the central eight-membered ring
through RCM. Initial attempts to effect this reaction with
substrate 19 were frustrated by the unexpected participation
of the tetrasubstituted alkene resulting in the formation of
a cyclopentene. We therefore decided to change the order of
events and explore the introduction of the epoxide first.
Gratifyingly, treatment of triene 19 with mCPBA led to
a clean, chemoselective, and stereoselective reaction at the
tetrasubstituted double bond to provide epoxide 20 as a single
diastereomer. Virtually no epoxidation of the terminal double
bonds was observed under the applied reaction conditions,
even when a large excess of reagent was used.
Scheme 4. Completion of the total synthesis of nitidasin. TASF=tris-
(dimethylamino)sulfonium difluorotrimethylsilicate, brsm=based on
recovered starting material, TPAP=tetrapropylammonium perruthen-
ate, NMO=4-methyl-morpholine N-oxide.
lective concerning the formation of the newly formed
stereogenic center. The diastereomers originating from sca-
lemic intermediate 4 were readily separable by column
chromatography and yielded 7% of the undesired adduct.
The major isomer 6, however, was unstable upon concen-
tration and was thus immediately subjected to epoxidation
conditions. This two-step protocol furnished metathesis sub-
strate 22 as single diastereomer and in an excellent yield of
71%, thus suggesting that in analogy to the model system,
a kinetic resolution took place. The subsequent RCM
proceeded smoothly to yield cyclooctene 23, which features
a total of eleven stereogenic centers. Its structure was
unambiguously confirmed by X-ray diffraction (Figure 3).
In order to complete the total synthesis, we initially aimed
at hydrogenation of the double bond of alkene 23 to obtain
cyclooctane 24. The use of stoichiometric amounts of Pd/C
resulted in a fast conversion to intermediate 24, the structure
of which was again established by single-crystal X-ray analysis
Single-crystal X-ray analyses of both triene 19 and
epoxide 20 not only confirmed the correct configuration of
all stereogenic centers, but also provided insight into the
[
28]
preferred conformations of the molecules (Figure 2). Both
structures show a close proximity of their terminal double
bonds, thus requiring only a 1208 rotation around the C2ÀC3
bond for an RCM to occur. Accordingly, exposure of diene 20
to the Grubbs second-generation catalyst in refluxing ben-
Figure 2. Crystal structures of model compounds 19 and 20.
Angew. Chem. Int. Ed. 2014, 53, 8513 –8517
Figure 3. Crystal structures of intermediates 23 and 24.
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8515

.
Angewandte
Communications
(
Figure 3). Since this substrate proved to be sensitive, we
[7] Y. Wang, M. Dreyfuss, L. Oberer, C. Sꢀtterlin, H. Riezman, Helv.
Chim. Acta 1998, 81, 2031 – 2042.
decided to cleave the SEM ether in tetracarbocycle 23 first.
After some experimentation, we found that treatment with
excess TASF in HMPA in the presence of molecular sieves at
elevated temperatures cleanly unmasked the corresponding
[
8] a) For a recent review on synthetic approaches toward sester-
terpenoids: D. T. Hog, R. Webster, D. Trauner, Nat. Prod. Rep.
2012, 29, 752 – 779; b) for a recent review on progress concerning
the synthesis of trans-hydrindane containing natural products: G.
Yue, X. Huang, B. Liu, Chin. J. Org. Chem. 2013, 33, 1167 – 1185.
9] a) For studies toward YW-3699: R. Mizutani, K. Nakashima, Y.
Saito, M. Sono, M. Tori, Tetrahedron Lett. 2009, 50, 2225 – 2227;
for studies toward nitiol, see: b) M. S. Wilson, G. R. Dake, Org.
Lett. 2001, 3, 2041 – 2044; c) M. S. Wilson, J. C. S. Woo, G. R.
Dake, J. Org. Chem. 2006, 71, 4237 – 4245.
[
29]
secondary alcohol.
Subsequent hydrogenation of the
double bond under the previously established reaction
conditions gave “nitidasol” 25. This compound may well be
a biosynthetic precursor of nitidasin and could potentially be
found as a natural product. Finally, the oxidation of alcohol 25
by using TPAP/NMO furnished synthetic nitidasin.
The NMR spectral data for our synthetic nitidasin were
identical in all respects to those reported by Kawahara and
[
[30]
[
10] D. T. Hog, P. Mayer, D. Trauner, J. Org. Chem. 2012, 77, 5838 –
843.
5
[
[
[
11] D. T. Hog, Dissertation, LMU Munich, 2013.
12] Z. G. Hajos, D. R. Parrish, Org. Synth. 1985, 63, 26 – 31.
13] R. A. Micheli, Z. G. Hajos, N. Cohen, D. R. Parrish, L. A.
Portland, W. Sciamanna, M. A. Scott, P. A. Wehrli, J. Org. Chem.
1975, 40, 675 – 681.
[14] For reviews on the synthesis of cyclooctenes by RCM: a) M.
Tori, R. Mizutani, Molecules 2010, 15, 4242 – 4260; b) A.
Michaut, J. Rodriguez, Angew. Chem. 2006, 118, 5870 – 5881;
Angew. Chem. Int. Ed. 2006, 45, 5740 – 5750; for a recent
example: c) K. Tsuna, N. Noguchi, M. Nakada, Angew. Chem.
[2]
co-workers. Carefully purified acid-free CDCl was required
3
to obtain clean spectroscopic data for comparison. Otherwise,
a rapid conversion to an exo-methylene cyclooctane that
resembles the YW compounds was observed (see the
Supporting Information). The identity of our synthetic
sample of nitidasin was also confirmed by X-ray crystallog-
raphy (see the Supporting Information). The optical rotation
of our synthetic material was determined to be levorotary as
reported for the natural product (synthetic: [a] = À1028, c =
D
2011, 123, 9624 – 9627; Angew. Chem. Int. Ed. 2011, 50, 9452 –
9455; d) K. Tsuna, N. Noguchi, M. Nakada, Chem. Eur. J. 2013,
19, 5476 – 5486.
0
.10, CHCl ; natural: [a] = À41.48, c = 0.28, CHCl ). The
3
D
3
absolute configuration of nitidasin has thus been established.
In conclusion, we have developed a total synthesis of
nitidasin, which was a primary goal of our program on
isopropyl trans-hydrindane sesterterpenoids. Our convergent
synthesis is marked by several highly stereoselective trans-
formations, which allowed for the installation of the ten
stereogenic centers of the target molecule. These include
a facile, diastereoselective addition of a complex tetrasub-
stituted alkenyl lithium compound to a trans-hydrindanone.
Another key step of our synthesis is an efficient RCM to form
a highly substituted eight-membered ring, a reaction that
benefits from conformational pre-organization of the sub-
strate. In addition, our synthetic route established the
absolute configuration of the natural product. The total
synthesis of astellatol from key ketone 5 and synthetic
approaches toward the YW molecules are currently under
active investigation.
[15] For a similar challenge in the synthesis of colombiasin A: K. C.
Nicolaou, G. Vassilikogiannakis, W. Mꢁgerlein, R. Kranich,
Angew. Chem. 2001, 113, 2543 – 2547; Angew. Chem. Int. Ed.
2001, 40, 2482 – 2486.
[
16] a) E.-i. Negishi, H. Matsushita, S. Chatterjee, R. A. John, J. Org.
Chem. 1982, 47, 3188 – 3190; b) E.-i. Negishi, F.-T. Luo, A. J.
Pecora, A. Silveira, Jr., J. Org. Chem. 1983, 48, 2427 – 2430.
17] For a recent application: I. Jastrzebska, J. B. Scaglione, G. T.
DeKoster, N. P. Rath, D. F. Covey, J. Org. Chem. 2007, 72, 4837 –
[
4843.
[18] J. B. Scaglione, N. P. Rath, D. F. Covey, J. Org. Chem. 2005, 70,
1089 – 1092.
[
19] a) R. Pappo, D. S. Allen, Jr., R. U. Lemieux, W. S. Johnson, J.
Org. Chem. 1956, 21, 478 – 479; b) J. Xu, L. Trzoss, W. K. Chang,
E. A. Theodorakis, Angew. Chem. 2011, 123, 3756 – 3760;
Angew. Chem. Int. Ed. 2011, 50, 3672 – 3676.
20] a) A. J. Mancuso, D. Swern, Synthesis 1981, 165 – 181; for
selected examples of deprotective Swern oxidations employing
primary triethylsilyl ethers: b) J. Becker, K. Bergander, R.
Frçhlich, D. Hoppe, Angew. Chem. 2008, 120, 1678 – 1681;
Angew. Chem. Int. Ed. 2008, 47, 1654 – 1657; c) S. M. Canham,
D. J. France, L. E. Overman, J. Am. Chem. Soc. 2010, 132, 7876 –
7877.
[21] A shorter reaction sequence involving reduction of lactone 10 to
the corresponding lactol (DIBAL-H, À788C) followed by
olefination proved nonviable.
22] a) Y. Q. Tu, A. Hꢀbener, H. Zhang, C. J. Moore, M. T. Fletcher,
P. Hayes, K. Dettner, W. Francke, C. S. P. McErlean, W. Kitching,
Synthesis 2000, 1956 – 1978; b) T. Magauer, H. J. Martin, J.
Mulzer, Angew. Chem. 2009, 121, 6148 – 6152; Angew. Chem. Int.
Ed. 2009, 48, 6032 – 6036.
[
Received: March 24, 2014
Published online: June 24, 2014
Keywords: ring-closing olefin metathesis · sesterterpenoids ·
.
stereoselective synthesis · total synthesis
[
[
[
[
[
[
1] F. Senatore, V. D. Feo, Z. L. Zhou, Ann. Chim. 1991, 81, 269 –
274.
2] N. Kawahara, M. Nozawa, D. Flores, P. Bonilla, S. Sekita, M.
Satake, K.-i. Kawai, Chem. Pharm. Bull. 1997, 45, 1717 – 1719.
3] N. Kawahara, M. Nozawa, A. Kurata, T. Hakamatsuka, S. Sekita,
M. Satake, Chem. Pharm. Bull. 1999, 47, 1344 – 1345.
4] N. Kawahara, M. Nozawa, D. Flores, P. Bonilla, S. Sekita, M.
Satake, Phytochemistry 2000, 53, 881 – 884.
[23] The low yield of this two-step protocol is due to the high
volatility of enyne 14, complicating its isolation. The trans-
formations themselves were highly efficient.
[24] G. Agnel, Z. Owczarczyk, E.-i. Negishi, Tetrahedron Lett. 1992,
33, 1543 – 1546.
5] a) I. H. Sadler, T. J. Simpson, J. Chem. Soc. Chem. Commun.
1989, 1602 – 1604; b) I. H. Sadler, T. J. Simpson, Magn. Reson.
Chem. 1992, 30, S18 – S23.
6] Y. Wang, M. Dreyfuss, M. Ponelle, L. Oberer, H. Riezman,
Tetrahedron 1998, 54, 6415 – 6426.
[25] A. M. Garcꢂa, J. L. MascareÇas, L. Castedo, A. MouriÇo, J. Org.
Chem. 1997, 62, 6353 – 6358.
[
8516
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 8513 –8517

Angewandte
Chemie
[
26] a) J. A. Dale, D. L. Dull, H. S. Mosher, J. Org. Chem. 1969, 34,
A. Das, R.-S. Liu, J. Am. Chem. Soc. 2006, 128, 9340 – 9341;
2
9
543 – 2549; b) J. A. Dale, H. S. Mosher, J. Am. Chem. Soc. 1973,
5, 512 – 519.
g) M. E. Jung, M. Murakami, Org. Lett. 2006, 8, 5857 – 5859.
[28] CCDC 934939, 987776, 987777, 987778, 988636, and 990612
contain the supplementary crystallographic data for this paper.
These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[
27] For selected examples on the addition of tetrasubstituted alkenyl
lithium species to aldehydes: a) L. E. Overman, M. J. Sharp, J.
Am. Chem. Soc. 1988, 110, 612 – 614; b) M. J. Di Grandi, D. K.
Jung, W. J. Krol, S. J. Danishefsky, J. Org. Chem. 1993, 58, 4989 –
4
992; c) A. Melekhov, P. Forgione, S. Legoupy, A. G. Fallis, Org.
[29] J. A. Gladding, J. P. Bacci, S. A. Shaw, A. B. Smith III, Tetrahe-
dron 2011, 67, 6697 – 6706.
[30] W. P. Griffith, S. V. Ley, G. P. Whitcombe, A. D. White, J. Chem.
Soc. Chem. Commun. 1987, 1625 – 1627.
Lett. 2000, 2, 2793 – 2796; d) J. Chen, Q. Song, P. Li, H. Guan, X.
Jin, Z. Xi, Org. Lett. 2002, 4, 2269 – 2271; e) R. Nakajima, C.
Delas, Y. Takayama, F. Sato, Angew. Chem. 2002, 114, 3149 –
3151; Angew. Chem. Int. Ed. 2002, 41, 3023 – 3025; f) M.-Y. Lin,
Angew. Chem. Int. Ed. 2014, 53, 8513 –8517
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8517