Stereocontrolled Construction of the A-Ring of Nitiol Using a Pauson-Khand Cycloaddition-Ring Fragmentation Strategy

Article abstract of DOI:10.1021/ol016002l

(matrix presented) A stereocontrolled construction of the A-ring of nitiol (1) is presented. Key features in this approach are a diastereoselective Pauson-Khand cycloaddition and a Norrish type 1 fragmentation reaction.

Full text of DOI:10.1021/ol016002l

ORGANIC
LETTERS
2001
Vol. 3, No. 13
2041-2044
Stereocontrolled Construction of the
A-Ring of Nitiol Using a Pauson−Khand
Cycloaddition−Ring Fragmentation
Strategy
Michael S. Wilson and Gregory R. Dake*
Department of Chemistry, 2036 Main Mall, UniVersity of British Columbia,
VancouVer, British Columbia, Canada V6T 1Z1
gdake@chem.ubc.ca
Received April 19, 2001
ABSTRACT
A stereocontrolled construction of the A-ring of nitiol (1) is presented. Key features in this approach are a diastereoselective Pauson−Khand
cycloaddition and a Norrish type 1 fragmentation reaction.
Plants continue to be a rich source of compounds with diverse
structures and medicinal activities.1 The plant named
Gentianella (G.) nitida (Gentianaceae) (common names
“hercampuri” or “hircampure”) is a biennial medicinal plant
that grows in the Andes region and is used in traditional
folk medicine. It is used to remedy hepatitis, as a cholagogue,
and in the treatment of obesity. Extracts from this whole
plant were purified several times, which led to the isolation
of nitiol (1), a novel sesterterpenoid (Scheme 1).2 Preliminary
investigations have found that 1 acts as a potent enhancer
of interleukin-2 in human T cell lines. In addition, because
1 has a structure distinctly different from those of known
modulators of IL-2 gene expression, it represents a possible
tool to probe for new signal transduction pathways that guide
IL-2 gene transcription.
cis relationship between the isopropyl group on C18 and the
C23 methyl group.
Chemical synthesis of 1 would serve to corroborate the
structural determination made by Kawahara and co-workers.
In addition, possibilities exist through chemical synthesis to
fashion analogue structures of 1 that may exhibit enhanced
biological activity.
Scheme 1. Analysis of Nitiol (1)
The structure of 1 was determined exclusively by NMR
methods. Its structure is characterized by a tricyclic 5-12-5
ring system. A trans ring fusion was found between the A
and B rings. A serious issue for chemical synthesis is the
(1) Cordell, G. A. Phytochemistry 1995, 40, 1585. Houghton, P. J. Chem.
Ind. (London) 1999, 15. Paper, D. H. Planta Med. 1998, 64, 686.
(2) Kawahara, N.; Nozawa, M.; Kurata, A.; Hakamatsuka, T.; Sekita,
S.; Satake, M. Chem. Pharm. Bull. 1999, 47, 1344.
10.1021/ol016002l CCC: $20.00 © 2001 American Chemical Society
Published on Web 06/07/2001

Analysis of the target structure 1 suggests a bifurcative
synthetic approach (Scheme 1). Disconnection of the C1-
C2 bond provides an uncyclized precursor 2. In the synthetic
direction, a number of possibilities, including a carbonyl
addition reaction or a tandem carbometalation-SN2 displace-
ment, are expected to connect this crucial ring-forming bond.
Methylmetalation of a terminal alkyne will be used to
generate the necessary double bond geometry prior to the
ring-closing reaction. Subsequent disconnection of the C10-
C11 bond generates two halves, 3 and 4. This bond is to be
constructed using standard coupling protocols using a pal-
ladium(0)-catalyzed process. The vinyl halide of the “west-
ern” portion, 4, will result from reduction, Horner-
Wadsworth-Emmons olefination, and functional group
manipulation of 5. Cyclopentane 5, the A-ring of nitiol,
contains a key cis stereochemical relationship between the
methyl and isopropyl groups in addition to what will become
the trans ring junction of the A-B ring system.
substituent on the chain tethering the alkyne complex and
the alkene should end up on the convex face of the
bicyclooctene to minimize steric strain.5 The main question
was whether enough steric differentiation existed between a
methyl and a CH2OTBS group. If successful, this process
would set up the relative stereochemistry for the A ring of
nitiol. A benefit to this analysis was that 6 was thought to
be obtainable from Pauson-Khand products 7 and 8, which
would derive from 9 and 10, respectively. In this manner
the importance of the position of the chirality center (the
allylic or propargylic carbon) could be examined.
The construction of the enynes 9 and 10 was carried out
in a straightforward manner (Scheme 3).6 The formation of
Scheme 3a
A key issue in any construction of nitiol is the establish-
ment of the configuration of the isopropyl moiety on the
cyclopentane A ring. A classic method for setting stereo-
chemistry is to use the bias of a cyclic system, after which
the ring is broken open. A trans-hydrindane ring system may
seem like an appropriate precursor to 5. However, it is well
established that reactions generating trans-hydrindanes bear-
ing pendant methyl and isopropyl groups almost exclusively
result in a trans-relationship between the methyl and iso-
propyl groups.3 For this reason, our synthetic approach to 5
uses the conformational bias of a [3.3.0]bicyclooctane system
to generate the required stereochemistry (Scheme 2). A
a Reagents and conditions: (a) LAH, Et2O (for 11, 98%; for 12,
96%); (b) Et3N, TBSCl, DCM, rt (for 11, 89%; for 12, 94%); (c)
DMSO, (COCl)2, Et3N, DCM, -78 °C (for 11, 87%; for 12, 98%);
Scheme 2. Analysis of Fragment 5
n
(d) (i) Ph3P, CBr4, (ii) BuLi, THF, -78 °C (65%); (e) CH2Br2,
Zn, TiCl4, PbCl2 (86%).
2,2-disubstituted malonates 11 and 12 was carried out by
alkylating diethyl methylmalonate with the appropriate
tosylate using sodium hydride and catalytic sodium iodide
in dimethylformamide.7 Standard procedures (LAH reduc-
tion, TBSCl protection, oxidation, Corey-Fuchs or Nozaki
reactions) were then used to generate the requisite enyne
substrates 9 and 10.
(4) Brummond, K. M.; Kent J. L. Tetrahedron 2000, 56, 3263. Schore,
N. E. In ComprehensiVe Organic Synthesis; Trost, B. M., Fleming, I.,
Paquette, L. A., Eds.; Pergamon Press: Oxford, 1991; Vol. 5, p 1037.
Schore, N. E. In Organic Reactions; Paquette, L. A., Ed.; Wiley: New
York, 1991; Vol. 40, p 2.
(5) Mukai, C.; Uchiyama, M.; Sakamoto, S.; Hanaoka, M. Tetrahedron
Lett. 1995, 36, 5761. Mukai, C.; Kim, J. S.; Uchiyama, M.; Sakamoto, S.;
Hanaoka, M. J. Chem. Soc., Perkin Trans. 1 1998, 2903. Magnus, P.;
Principe, L. M. Tetrahedron Lett. 1985, 26, 4851.
number of methods could be used for the fragmentation of
the bicyclooctane, including a Baeyer-Villiger reaction, a
fragmentation proceeding through an oxygen-based radical,
or a Norrish type 1 process.
After the consideration of a number of options, it was
determined that ketone 6 represented a key intermediate,
whose construction would utilize the Pauson-Khand reac-
tion.4 Following several literature precedents, the larger
(6) All new compounds have been satisfactorily characterized spectro-
scopically (NMR, IR, MS).
(7) For an enantioselective construction of the A-ring, pig liver esterase
could be used to desymmetrize malonates 11 or 12. As the remainder of
the construction is stereoselective, this step would define the absolute
stereochemistry for the construction of the western half of 1. Trost, B. M.;
Li, Y. J. Am. Chem. Soc. 1996, 118, 6625. Bjo¨rkling, F.; Boutelje, J.;
Gatenbeck, S.; Hult, K.; Norin, T.; Szmulik, T. Tetrahedron 1985, 41, 1347.
Luyten, M.; Mu¨ller, S.; Herzog, B.; Keese, R. HelV. Chim. Acta 1987, 70,
1250. Toone, E. J.; Jones, J. B. Tetrahedron: Asymmetry 1991, 2, 1041.
(3) (a) Corey, E. J.; Desai, M. C.; Engler, T. A. J. Am. Chem. Soc. 1985,
107, 4339. (b) Snider, B. B.; Rodini, D. J.; van Straten, J. J. Am. Chem.
Soc. 1980, 102, 5872. (c) Attah-Poku, S. K.; Chau, F.; Yadav, V. K.; Fallis,
A. G. J. Org. Chem. 1985, 50, 3418.
2042
Org. Lett., Vol. 3, No. 13, 2001

These substrates were then submitted to various Pauson-
various Pauson-Khand conditions led to improved results
(Table 2).
Khand conditions (Tables 1 and 2).8 Because we were
Running the reaction in methylene chloride using 6 equiv
of N-methylmorpholine N-oxide (NMO) as a promoter gave
a modest yield (40%), but happily the diastereoselectivity
favoring 7a was quite good (7.5:1) (entry 1), especially when
considering that the stereodiscriminating event differentiates
between a methyl and a CH2OTBS group. Our focus turned
to improving the yield of this process. Other solvents and
promoters were surveyed. The use of thioanisole to promote
the reaction in dichloroethane resulted in a very poor
recovery of 7a and 7b, 20% (entry 2). The use of DMSO in
benzene reestablished the overall yield of this process to 40%,
with high stereoselectivity (entry 3). The use of Sugihara’s
conditions (cyclohexylamine in 1,2-dichloroethane or 3:1
2 M ammonium hydroxide in water and dioxane) very
rapidly converted enyne 9 to the desired bicyclooctenes 7a
and 7b in good yields (57-74%) with small diminishments
in selectivity (5.7-6.3:1) (entries 4 and 5). Substoichiometric
versions of the Pauson-Khand reaction worked well at 50
mol % loading of dicobalt octacarbonyl, but a lower loading
of 10 mol % gave a poor yield (entries 6 and 7). The
“ammonium hydroxide” method of Sugihara was satisfactory
enough to continue the investigation of our planned route.
At this time no modifications of the protecting group of 9
to other more sterically hindered groups that may result in
improved stereoselection have been attempted.
Table 1. Pauson-Khand Reaction of (Propargyl-Substituted)
Enyne 10
entry solvent (and promoter) T (°C) t (h) yield dr (8a :8b)a
1
2
3
4
hexanes (none)
CH2Cl2 (NMO)
ClCH2CH2Cl (C6H5SMe)
water/dioxane (NH4OH)
110 20
25 12
13%
31%
48%
1:1
1:1
1:1
1:1
83
2
100
0.5 41%
a Ratios determined by integration of the 1H NMR spectrum of the crude
reaction mixture.
determining the viability of the overall approach, catalytic
versions of the Pauson-Khand reaction have not been
extensively examined, although these will be the subject of
future investigations. Unfortunately enyne 10, when treated
with dicobalt octacarbonyl followed by a promoter, generated
an equimolar mixture of diastereomers 8a and 8b in very
low yields (Table 1). Investigations using 10 were halted as
a result of these disappointing findings.
The relative stereochemistries of 7a and 7b were deter-
mined by NMR methods. These compounds, which are easily
separable by standard flash column chromatography on silica
gel, were individually subjected to NOE experiments (Figure
1).
Table 2. Pauson-Khand Reaction of (Allylic-Substituted)
Enyne 9
dr
Figure 1. Selected NOE correlations of 7a and 7b.
entry solvent (and promoter) T (°C) t (h) yield (7a :7b)a
1
2
3
4
5
6
7
CH2Cl2 (NMO)
ClCH2CH2Cl (C6H5SMe)
benzene (DMSO)
ClCH2CH2Cl (C6H11NH2)
water-dioxane (NH4OH)
DMEb (C6H11NH2)
DMEc (none)
25 12
83
45 72
40%
20%
40%
7.5:1
nd
2
Cyclopentenone 7a was then subjected to conjugate
reduction using lithium tri(sec-butyl)borohydride followed
by quenching with methyl iodide to produce 13 in 83%
isolated yield. The key R-carbonyl fragmentation reaction
was performed by irradiating 12 with light from a 450-W
Hanovia medium-pressure mercury lamp (Norrish type 1).
To our delight, this procedure furnished trisubstituted cy-
clopentane 5a in 50% yield (55% brsm). At this point the
7.7:1
5.7:1
6.3:1
5.7:1
6.0:1
83
100
0.25 57%
0.25 74%
60 18
70 18
84%
25%
a Ratios determined by GC integration; nd ) not determined. b 50 mol
% Co2(CO)8 under a N2 atmosphere. c 10 mol % Co2(CO)8 under a CO
atmosphere.
(8) (a) No promoter: Exon, C.; Magnus, P. J. Am. Chem. Soc. 1983,
105, 2477. (b) NMO: Adrio, J.; Rivero, M. R.; Carretero, J. C. Angew.
Chem., Int. Ed. 2000, 39, 2906. (c) C6H5SMe: Sugihara, T.; Yamada, M.;
Yamaguchi, M.; Nishizawa, M. Synlett 1999, 771. (d) DMSO: Chung, Y.
K.; Lee, B. Y.; Jeong, N.; Hudecek, M.; Pauson, P. L. Organometallics
1993, 12, 220. (e) NH4OH: Sugihara, T.; Yamada, M.; Yamaguchi, M.;
Kaneko, C. Angew. Chem., Int. Ed. Engl. 1997, 36, 2601. (f) Krafft, M. E.;
Bonaga, L. V. R.; Hirosawa, C. J. Org. Chem. 2001, 66, 3004.
Attention was then turned to 9, which has its chirality
center at the allylic position of the enyne. It was hoped that
the proximity of the quaternary chirality center to the forming
ring junction would enhance the diastereoselectivity of the
Pauson-Khand reaction. In the event, submission of 9 to
Org. Lett., Vol. 3, No. 13, 2001
2043

(E:Z ) 14:1). Reduction of the ester to the alcohol using
disobutylaluminum hydride proceeded normally (97%).
Experimental support of the stereochemistry of the vinyl
bromide was carried out by NOE NMR experiments on 14.
Specifically, NOE enhancements were observed between the
two sets of allylic protons.
Scheme 4a
In summary, a short construction of an intermediate
representing the A-ring of nitiol has been presented. The
key strategic feature is the cleavage of a functionalized
[3.3.0]bicyclooctane ring system to introduce the necessary
stereochemical features at C18 and C23. Future work will
include an enantioselective preparation of this A-ring com-
ponent and efforts toward the natural product, nitiol.
a Reagents and conditions: (a) (i) L-Selectride, (ii) CH3I (83%)
(b) hυ (g190 nm) quartz filter (50%, 55% brsm); (c) (i) DIBAL-
H, CH2Cl2, (ii) (COCl)2, DMSO, NEt3, CH2Cl2 (86%, two steps),
(iii) (CF3CH2O)2POCHBrCO2Et (A), KOtBu, 18-crown-6, THF,
-78 °C (>97%, 14:1 E:Z), (iv) DIBAL-H, CH2Cl2 (97%).
Acknowledgment. We thank the University of British
Columbia and the National Science and Engineering Re-
search Council of Canada for the funding of this research.
G.R.D. thanks Prof. Viresh Rawal and Prof. Edward Piers
for stimulating discussions. M.S.W. thanks Prof. John
Scheffer and his group for the use of photochemistry
equipment and UBC for a McDowell Award.
ester functionality was converted to an aldehyde under stand-
ard conditions (reduction with diisobutylaluminum hydride;
Moffatt-Swern oxidation; 86% for two steps). Treatment
of the aldehyde with Kogen’s Horner-Wadsworth-
Emmons-type reagent A9 generated the trisubstituted vinyl
bromide with excellent yield (>97%) and stereoselectivity
Supporting Information Available: Spectroscopic data
for 9 and 10 and experimental procedures and spectroscopic
data for compounds 7, 8, 13, and 5a-14. This material is
available free of charge via the Internet at http://pubs.acs.org.
(9) Tago, K.; Kogen, H. Org. Lett. 2000, 2, 1975.
OL016002L
2044
Org. Lett., Vol. 3, No. 13, 2001