Total synthesis and elucidation of the absolute configuration of the diterpene tonantzitlolone

Article abstract of DOI:10.1021/ol047559e

(Chemical Equation Presented) The first enantioselective total synthesis of tonantzitlolone, a novel 15-membered macrocyclic diterpene, utilized a Julia olefination, a highly selective, potassium enolate-based anti-Felkin aldol reaction, and an E-selectiv

Full text of DOI:10.1021/ol047559e

ORGANIC
LETTERS
2005
Vol. 7, No. 3
479-482
Total Synthesis and Elucidation of the
Absolute Configuration of the Diterpene
Tonantzitlolone†
Christian Jasper,‡ Rudiger Wittenberg,‡ Monika Quitschalle,‡
1
Jasmin Jakupovic,§ and Andreas Kirschning*,‡
Institut fu¨r Organische Chemie der UniVersita¨t HannoVer, Schneiderberg 1B,
D-30167 HannoVer, Germany, and AnalytiCon DiscoVery GmbH,
Hermannswerder Haus 17, D-14473 Potsdam, Germany
andreas.kirschning@oci.uni-hannoVer.de
Received November 29, 2004
ABSTRACT
The first enantioselective total synthesis of tonantzitlolone, a novel 15-membered macrocyclic diterpene, utilized a Julia olefination, a highly
selective, potassium enolate-based anti-Felkin aldol reaction, and an E-selective ring-closing metathesis as key C
−C bond-forming steps. The
absolute configuration of tonantzitlolone is established.
In 1990, X. A. Dominguez et al. (Departamento de Quimica,
ITESM, Monterrey, Mexico) isolated a new natural com-
pound from the endemic Mexican plant Stillingia sanguino-
lenta.1,2 The structure (1, Figure 1) was elucidated in a
cooperation with J. Jakupovic, and the new diterpene of
unknown absolute configuration was named tonantzitlolone.
In 1997 in the course of a detailed reinvestigation of the
same plant, Jakupovics and Jeske2 obtained a second
derivative (OAc at C-4′ of side chain) of tonantzitlolone
along with a series of other common diterpenes.2 Tonantz-
itlolone contains a 15-membered carbocyclic ring that so far
has only been found once before in nature in the macrocycle
flexibilene 2.3 The highly oxygenated diterpene 1 contains
Figure 1.
strong internal hydrogen bonds between the hydroxy group
at C-10 and the ester oxygens of the side chain and in
particular between the lactol hydroxy group and the furan
oxygen atom.2 This network of hydrogen bonds gives the
† Dedicated to Prof. Dr. H. G. Floss on the occasion of his 70th birthday.
‡ Institut fu¨r Organische Chemie der Universita¨t Hannover.
§ AnalytiCon Discovery GmbH.
(1) Roots of these plants were used by native Mexicans, Navajos, and
Creeks in poultices of children, and the leaves were employed to treat
pulmonany ailments.
(3) Herin, M.; Colin, M.; Tursch, B. Bull. Soc. Chim. Belg. 1976, 85,
707-719. Kazlauskas, R.; Murphy, P. T.; Wells, R. J.; Scho¨nholzer, P.;
Coll, J. C. Aust. J. Chem. 1978, 31, 1817-1824.
(2) Jeske, F. Ph.D. Thesis, Technische Universita¨t Berlin, Berlin,
Germany, 1997.
10.1021/ol047559e CCC: $30.25
© 2005 American Chemical Society
Published on Web 01/13/2005

macrocyle a very rigid, almost globular overall structure. This
compactness is further enhanced by a set of five methyl
substituents that branch off the macrocycle. First biological
tests revealed moderate cytotoxic activity of IC50 ) 43 µM
on PtK2 potoroo cells.4
Recently, we described the synthesis of the complete
carbon backbone of 1 as an acyclic precursor.5 It was planned
to cyclize this precursor by ring-closing metathesis (RCM),6
thus forming the double bond at C-1 and C-2. Unfortunately,
efforts to force macrocyclization by RCM or McMurry
reaction were unsuccessful probably due to steric congestion
created by the three allylic methyl groups. Therefore, we
altered our synthetic strategy by shifting the position of
macrocyclization to the less congested site at C-4 and C-5
(Scheme 1). In our retrosynthetic analysis, we first planned
southern hemisphere D was further simplified to the methyl
ester 3 and hence to methyl geranate 4, which served as
starting material.
As the absolute configuration of tonantzitlolone 1 was
unknown and no precedence of a structurally related diter-
perne was known, we had to arbitrarily choose one target
enantiomer as depicted in Scheme 1. Thus, aldehyde 5 was
prepared from methyl geranate 4 as described in the
literature8 and was subjected to the Kiyooka aldol protocol9
using ketene acetal 6, which yielded the diester 3 in good
yield and stereoselectivity (91% ee; determined by 1H NMR
spectroscopy in the presence of Eu(hfc)3 (Scheme 2).
Scheme 2a
Scheme 1. Retrosynthetic Analysis (Total Synthesis Was
Conducted toward 1 with the Configuration Depicted in This
Scheme)a
a mCPBA ) meta-chloroperbenzoic acid. Ts ) p-toluene sul-
fonyl. DMP ) 2,2-dimethoxy propane. DET ) (-)-diethyl-D-
tartrate.
Exhaustive reduction to the triol 7, followed by protection
of the 1,3-diol to the corresponding 1,3-dioxolane, set the
stage for introducing the oxirane ring using the Sharpless-
Katsuki protocol.10 The resulting epoxy alcohol 8 was
cyclized and reprotected in one pot by treatment with
dimethoxy propane under acidic conditions. The sequence
consisted of activation of the epoxide and intramolecular
attack by a 5-exo-tet cyclization of the 1,3-dioxolane ring
onto the preferred carbenium ion, which created a new 1,2-
diol, while the acetone was released from the other terminus
of the molecule.
a PG ) protective group. Ar ) aryl.
to cleave the (E)-3-methyl-pent-2-enoic acid side chain along
with removal of the oxygen functionalities at C-4 and C-5
to create macrocycle A as a key precursor. The macro-
cyclization was planned to be achieved by ring-closing
metathesis, so an acyclic precursor was required that could
be further dissected into three fragments B,7 C, and D. While
the former two fragments could readily be prepared, the
One driving force for this thermodynamically controlled
sequence can be seen in the protection of the newly created
(4) Tests were carried out at the Gesellschaft fu¨r Biotechnologische
Forschung (GBF), Braunschweig, by F. Sasse. The activity can be enhanced
greatly when more water-soluble derivatives are prepared.
(5) Wittenberg, R.; Monenschein, H.; Dra¨ger, G.; Beier, C.; Jas, G.;
Jasper. C.; Kirschning, A. Tetrahedron Lett. 2004, 45, 4457-4460.
(6) (a) Prunet, J. Angew. Chem., Int. Ed. 2003, 42, 2826-2830. (b) Trnka,
T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18-29. (c) Fu¨rstner, A.
Angew. Chem., Int. Ed. 2000, 39, 3012-3043.
(7) (a) Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem Soc. 1982,
104, 1737-1739. (b) Evans, D. A.; Bender, S. L.; Morris, J. J. Am. Chem
Soc. 1988, 110, 2506-2526.
(8) Henrick, C. A.; Schaub, F.; Siddall, J. B. J. Am. Chem. Soc. 1972,
94, 5374-5378.
(9) Kiyooka, S.-i.; Hena, M. A. J. Org. Chem. 1999, 64, 5511-5523,
and references therein. Machajewski, T. D.; Wong, C. H. Angew. Chem.,
Int. Ed. 2000, 39, 1352-1375 and references therein.
(10) (a) Gao, Y.; Hanson, R. M.; Klunder, J. M.; Ko, S. Y.; Masamune,
H.; Sharpless, K. B. J. Am. Chem. Soc. 1987, 109, 5765-5780. (b) Marty,
M.; Stoeckli-Evans, H.; Neier, R. Tetrahedron 1996, 52, 4645-4658.
480
Org. Lett., Vol. 7, No. 3, 2005

1,2-diol as isopropylidene acetal. In fact, in the absence of
2,2-DMP a complex mixture of products was formed. TPAP-
promoted oxidation then led to the desired aldehyde (Scheme
3).12
Scheme 4a
Scheme 3a
a DIAD ) diisopropylazodicarboxylate. PT(SH) ) phenyltetra-
zole (thiol). TPAP ) tetra-n-propylammonium perruthenate. NMO
) N-methylmorpholine-N-oxide. LDA ) lithiumdiisopropylamide.
The termini of fragment 9 were sequentially elongated with
building blocks B and C, the latter being added prior to
aldehyde B by means of the Julia-Kocienski olefination.11
Sulfone 12 was prepared from (R)-methyl hydroxyisobutyrate
10, which was transformed into the phenyltetrazol thioether
under Mitsunobu conditions13 followed by reduction to yield
alcohol 11 (Scheme 3). Oxidation of the alcohol, subsequent
Wittig olefination, and finally oxidation of the thioether
yielded sulfone 12. This was coupled with the aldehyde
derived from alcohol 9 under slightly modified conditions
to the ones described before14 (lithiumdiisopropylamide in
tetrahydrofuran and warming to room temperature), which
furnished the desired (E)-olefin 13 in excellent 92% yield
(dr ) 9:1).
After three functional group manipulations (deprotection,
silylation, and oxidation), ketone 14 was obtained, which
set the stage for the key aldol reaction with aldehyde B
(Scheme 4).15 KHMDS proved to be the best choice for the
deprotonation of ketone 14 and the following aldol reaction
with aldehyde B. Remarkably, the adduct 15 was isolated
as a single diastereomer in excellent 98% yield.16 The
stereochemical outcome of the reaction could only be
a TBSCl ) tert-butyl-dimethylsilyl chloride. KHMDS ) potas-
sium hexamethyldisilazide.
elucidated after 1,3-syn-reduction to the corresponding diol,17
which then was protected as cyclic acetal 16. After desily-
lation, ring-closing olefin metathesis using the 2nd generation
Grubbs catalyst preferentially yielded the macrocyclic Z-
configured diene 17. The rigidity of the macrocycle as well
as the dioxolane ring allowed proof of the desired relative
7,8-anti-8,9-syn stereochemical relationship created during
the aldol reaction to 15 by means of observed NOE
correlations and H-H coupling constants and by analyzing
the 13C shifts with the Evans-Rychnovsky method.18,19
The synthesis was finalized starting from diol 16, which
had been obtained after 1,3-syn reduction of 15 (see Scheme
4). In fact, we had to return to this point of the synthesis, as
it is known that cis-disubstituted olefins such as 17 are
difficult to asymmetrically dihydroxylate.20 And indeed,
macrocycle 17 could not be oxidized, neither with OsO4 nor
under the Sharpless conditions.
(11) Blakemore, P. R. J. Chem. Soc., Perkin Trans. 1 2002, 2563-2585
and references therein.
(12) Ley, S. V.; Norman, J.; Griffith, W. P.; Marsden, S. P. Synthesis
1994, 639-666.
(17) Reaction did not always proceed to completion. In these cases,
recovered starting material can be resubjected to enhance material through-
put.
(18) Observed coupling constants along the C-7-C-10 chain: δ ) 1.91-
2.06 (m, 2 H; H-7, 9-OH), 3.28 (br s, 1 H, H-10), 3.69 (d, J7,8 ) 6.1 Hz,
1 H; 8-H), 3.92 (d, J9,OH ) 11.3 Hz, 1 H, H-9); no H-H coupling constants
between C-8/C-9 and C-9/C-10 are observed, which supports the suggested
conformation depicted in Scheme 4.
(19) Determination of the 8,10-syn-relationship: (a) Evans, D. A.; Rieger,
D. L.; Gage, J. R. Tetrahedron Lett. 1990, 31, 7099-7100. (b) Rychnovsky,
S. D.; Rogers, B.; Yang, G. J. Org. Chem. 1993, 58, 3511-3515.
(20) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. ReV.
1994, 94, 2483-2547.
(13) (a) Mitsunobu, O. Synthesis 1981, 1-28. (b) Hughes, D. L.; Reamer,
R. A.; Bergan, J. J.; Grabowski, E. J. J. J. Am. Chem. Soc. 1988, 110,
6487-6491.
(14) Blakemore, P. R.; Cole, W. J.; Kocienski, P. J.; Morley, A. Synlett
1998, 26-28.
(15) (a) Mengel, A.; Reiser, O. Chem. ReV. 1999, 99, 1191-1223. (b)
Marco, J. A.; Carda, M.; D´ıaz-Oltra, S.; Murga, J.; Falomir, E.; Roeper, H.
J. Org. Chem. 2003, 68, 8577-8582. (c) Evans, D. A.; Siska, S. J.; Cee,
V. J. Angew. Chem., Int. Ed. 2003, 42, 1761-1765.
(16) Detailed mechanistic discussion on this aldol reaction will be
presented in a full account.
Org. Lett., Vol. 7, No. 3, 2005
481

Removal of the TBS group in 16 yielded the corresponding
triol, which could be selectively silylated at the terminal
hydroxy groups, followed by oxidation at the central position
with the Dess-Martin periodinane. Deprotection to the
corresponding dihydroxy ketone and ring-closing metathesis
then furnished macrocycle 18, now with high E-selectivity
(Scheme 5).
yielded triol 19. With this macrocycle at hand, all stereogenic
centers of the natural product were set up. In the following,
no protecting group strategy tested allowed the efficient
differentiation of the three hydroxy groups. We therefore
decided to directly esterify triol 19, which furnished all three
possible regioisomers in an overall yield of 92%.21 Due to
the 1,3-syn relationship of the oxygen functionalities in 20a,
the ester side chain at C-10 tended to migrate to the C-8
position. This migration could be effected quantitatively by
addition of a catalytic amount of DMAP to a solution of
20a, thereby improving the yield to 57% in favor of 20b.
The final oxidation to the natural product was selectively
carried out with TPAP, providing the (ent)-tonantzitlolone
1 as was judged from the optical rotation (authentic natural
Scheme 5a
product, [R]20 ) +134° (c 0.25, CHCl3); found, [R]20
)
D
D
-119° (c 0.06, CHCl3)). Apart from these facts, all spec-
troscopic data (NMR, IR, MS) were in full accordance with
those of the authentic material.2
In conclusion, we achieved the first total synthesis of the
novel diterpene tonantzitlolone 1 and, importantly, elucidated
its absolute configuration. The synthesis is highly selective
and relies on the effective use of substrate-controlled
reactions that allow the introduction of the stereogenic centers
in the northern and the western region of 1. Details on the
biological properties and the biological target of tonantz-
itlolone, its (-)-enantiomer, and various derivatives will be
reported in a full account.
Acknowledgment. This work was supported by a Kekule´
scholarship for C.J. provided by the Stiftung Stipendien-
Fonds des Verbandes der Chemischen Industrie and by a
graduate fellowship provided by the state of Lower Saxony
for R.W. We are indepted to Dr. Florenz Sasse from the
Gesellschaft fu¨r Biotechnologische Forschung (GBF) (Braun-
schweig) for carrying out the biological tests. We are grateful
to Dr. Gerhard Jas (Synthacon, Frankfurt) for helpful
discussions. Finally, we thank the Fonds der Chemischen
Industrie for general support and Solvay Pharmaceuticals
GmbH (Hannover) for the donation of chemicals.
a DHQ-CLB ) O-(4-chlorobenzoyl)hydroquinine. DIC ) diiso-
propylcarbodiimide. DMAP ) (dimethylamino)pyridine.
Studies to selectively introduce the side chain prior to
oxidation of the newly formed double bond proved to be
ineffective. We therefore dihydroxylated the C-4/C-5 double
bond first without affecting the sterically encumbered double
bond at C-1/C-2. The best stereoselection (dr for 4S,5S/4R,5R
) 3:1) was achieved when the two hydroxy groups in 18
were first protected as TMS ethers. The resulting diol had
to be completely desilylated prior to lactol formation, which
Supporting Information Available: Descriptions of
experimental procedures as well as analytical data for all
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
OL047559E
(21) Regioisomeric esters were separable after the oxidation that followed.
482
Org. Lett., Vol. 7, No. 3, 2005