[2.2.2]- to [3.2.1]-Bicycle Skeletal Rearrangement Approach to the Gibberellin Family of Natural Products

Article abstract of DOI:10.1021/acs.orglett.8b01031

Synthetic studies toward the gibberellin family of natural products are reported. An oxidative dearomatization/Diels-Alder cascade assembles the carbon skeleton as a [2.2.2]-bicycle, which is then transformed to the [3.2.1]-bicyclic gibberellin core via a

Full text of DOI:10.1021/acs.orglett.8b01031

Letter
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
pubs.acs.org/OrgLett
[
2.2.2]- to [3.2.1]-Bicycle Skeletal Rearrangement Approach to the
Gibberellin Family of Natural Products
Brandon R. Smith and Jon T. Njardarson*
Department of Chemistry and Biochemistry, University of Arizona, Tucson, Arizona 85721, United States
*
S Supporting Information
ABSTRACT: Synthetic studies toward the gibberellin family of
natural products are reported. An oxidative dearomatization/
Diels−Alder cascade assembles the carbon skeleton as a [2.2.2]-
bicycle, which is then transformed to the [3.2.1]-bicyclic
gibberellin core via a novel Lewis acid catalyzed rearrangement.
Strategic synthetic handles allow for late-stage modification of
the gibberellin skeleton and provides efficient access to this
important family of natural compounds.
ibberellin research dates to late 19th century Japan, where
and the nearly four decades of work since Corey’s original
G
their essential role in plant growth and development was
first observed on rice seedlings. It was not until the 1960s that
gibberellins began to enter the minds of organic chemists; upon
structure elucidation by X-ray crystallography in 1963,
synthesis of GA3, the feat has still only been accomplished by
3
,5
6
four research groups: Corey (three times), Mander (twice),
7
8
Yamada, and DeClercq. Despite these extensive efforts, each
synthesis is long and plagued by low overall yields. These
syntheses and other efforts toward gibberellins employ a
number of common strategies. The A ring is generally
constructed by Diels−Alder cycloaddition, Birch reduction, or
aldol chemistry. Mimicking the biological synthesis, the B ring
is most frequently accessed via a ring contraction, though other
approaches such as Cope rearrangement or Friedel−Crafts
acylation have also been employed. The C ring commonly
originates from aromatic precursors or Cope approaches, while
the D ring has been synthesized via carbenoid, aldol and
reductive ring closure chemistry.
gibberellic acid (GA ) and other gibberellins became realistic
3
1
,2
targets for synthesis. The gibberellin family now includes
more than 130 members, named numerically in order of their
date of discovery. The biosynthetic pathway is now fairly well
understood; gibberellins derive from the tetracyclic diterpene
ent-kaurene skeleton, with GA -aldehyde serving as a common
intermediate to all gibberellins (Figure 1).
E. J. Corey and co-workers were the first to complete the
total synthesis of GA3. Corey reported at the time of his
1
2
3
4
original synthesis in 1978 there had been about 150 published
papers from approximately 25 different laboratories focused on
gibberellin synthesis. Despite this immense amount of effort
Our synthetic approach toward the gibberellin family pivots
on two key reactions (Scheme 1). First, leveraging our group’s
theme of utilizing oxidative dearomatization/Diels−Alder
9
chemistry to rapidly assemble complex structures, we sought
to access the fused [2.2.2]-bicycle 4 from phenol 5, which could
Scheme 1. Gibberellin Core Retrosynthesis
Received: March 31, 2018
Figure 1. Biosynthesis and GA total syntheses.
3
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b01031
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
be accessed from commercially available starting materials (6
and 7) in a few steps. Following synthesis of [2.2.2]-bicycle, the
C/D ring junction would be installed using a rearrangement
reaction. We first encountered this new [2.2.2]- to [3.2.1]-
cyclopentadiene-fused-bicycle rearrangement during our maoe-
3). gem-Dimethyl triflate 15 was synthesized in three steps
employing known literature procedures. In general, the
11
Scheme 3. Synthesis of gem-Dimethyl Containing
Rearrangement Precursor
9f
crystal V synthetic efforts. Hampered by what was then
undesired reactivity, we were eager to productively utilize this
rearrangement toward the gibberellin family. We are not the
first to approach the gibberellin C/D ring junction via a [2.2.2]-
to [3.2.1]-bicycle rearrangement as Mori, Monti, and Yamada
reported their efforts in the 1970s utilizing acid−based
10
rearrangement approaches. However, these prior rearrange-
ment approaches were unable to successfully access gibberellin
natural products.
We set out to synthesize a gibberellin skeleton to confirm the
validity of our key dearomatization/Diels−Alder and rearrange-
ment steps. Toward that end, Suzuki coupling of boronic acid
8
, synthesized in two steps from 6, and triflate 9, synthesized in
one step from 7, delivered ester 10 in excellent yield (Scheme
). Attempts to selectively reduce ester 10 to aldehyde 11 were
2
Scheme 2. Synthesis of Rearrangement Precursors
Suzuki cross-coupling, DIBAL reduction, DMP oxidation, and
allyl installation all proceeded smoothly and in superb yield to
deliver phenol 18. In contrast to the hydrogen substituted
system (5), dearomatization/Diels−Alder of phenol 18
performed less efficiently, delivering cycloadduct 19 as a
mixture of inseparable products. The primary competing
pathway to the intramolecular Diels−Alder reaction was self-
dimerization of the quinone intermediate. With cycloadduct 19
in hand, we set out to synthesize cyclopentadienes 20 and 22.
Dehydration with Burgess’s reagent delivered cyclopentadiene
2
0. Alternatively, Dess-Martin oxidation followed by enol
triflation delivered cyclopentadiene 22 in poor yield. The added
steric hindrance from the gem-dimethyl group favored isomer-
ization of 21b to the less hindered 21a, necessitating short
reaction times to achieve the desired selectivity in the triflation.
With several different cyclopentadiene-fused-[2.2.2]-bicycle
constructs on hand, we first investigated the skeletal key
rearrangement under thermal conditions (Table 1). While the
rearrangement from a [2.2.2]- to [3.2.1]-bicycle did occur for
cyclopentadienes 12 and 20, it was not nearly as clean or facile
as the rearrangement that had been witnessed in our
thwarted, so a two-step reduction/oxidation procedure was
employed. Following a one-pot sequential addition of vinyl
magnesium bromide and TBAF, phenol 5 was accessed in
excellent yield. Treatment of 5 with PhI(OAc) (PIDA) in
2
methanol induced the desired oxidative-dearomatization/
Diels−Alder cascade. The rigid cyclohexene ring enabled a
facile intramolecular Diels−Alder occurring at 0 °C, delivering
alcohols 4 in high selectivity (40:3:3 ratio). The alcohol mixture
could be dehydrated with Burgess’s reagent to deliver
cyclopentadiene 12. Alternatively, 4 could be converted into
triflate 14 in two steps. Only the major oxidation product, 13b,
could be selectively converted into triflate 14. Efforts to convert
9f
maoecrystal V synthesis. Additionally, the thermal rearrange-
ment toward the gibberellin core was hampered by
decomposition, leading to low yields and complete decom-
position of the triflate substituted cyclopentadienes 14 and 21.
Having explored the thermal rearrangement, we shifted our
focus to installation of the C-7 gibberellin carboxylate group. In
addition to supplying a carbon necessary for the natural
product, we also envisioned that this functional group might
provide the opportunity to facilitate the occurrence of the
desired skeletal rearrangement under catalytic conditions. With
triflate substituted cyclopentadienes 14 and 22 on hand, we
sought to install C-7 as a nitrile due to its lack of steric bulk and
1
3a delivered a mixture of inseparable triflate products due to
deprotonation of the more sterically accessible allylic hydrogens
within the cyclohexene ring.
We also set out to synthesize a gibberellin construct
containing the C-18 and C-19 gem-dimethyl groups (Scheme
B
DOI: 10.1021/acs.orglett.8b01031
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 1. Thermal Rearrangement Studies
Scheme 4. Proposed Thermal vs Lewis Acid Catalyzed
Rearrangement Mechanisms
substrate
R, R′
H, H
yield (%)
temp (°C)
time (h)
1
1
2
2
2
4
0
1
38
0
110
70
144
12
OTf, H
H, Me
36
0
110
70
168
12
OTf, Me
12
plethora of palladium-catalyzed options for its incorporation.
We postulated that in one pot we would be able to convert the
triflate into a nitrile, which we argued would not only stabilize
the intermediate cyclopentadiene anion but also serve as a
handle for Lewis acids to accelerate the rearrangement. Triflate
Scheme 5. Additional Rearrangement Experiments
1
4 was employed for these studies, which are summarized in
Table 2. Rearrangement cascade explorations rapidly revealed
Table 2. Pd(0) Catalyzed Rearrangement Cascade
temp
(°C)
yield
(%) ratio 3:2
Pd source
CN source additive solvent
Pd(PPh3)4
Pd(PPh3)4
NaCN
Cul
MeCN
DMF
MeCN
DMF
DMF
DMF
70
80
60
90
90
90
26
0
0.7:1.0
CuCN
LiCI
dppf
none
none
dppf
Pd (dba)3
2
KCN
51
75
0
7:1
Pd(PPh3)4
Zn(CN)2
Zn(CN)2
Zn(CN)2
>1:20
Pd (dba)3
2
Pd (dba)3
2
99
>1:20
how significant a role the cyanide counterion played. This
prompted us to switch to a metal counterion better suited for
nitrile coordination. Ultimately, zinc cyanide emerged as the
ideal reagent. It not only facilitated installation of the nitrile but
also catalyzed the desired skeletal rearrangement, cleanly
delivering [3.2.1]-bicycle 2 (CCDC 1821698) in near-
quantitative yield.
Scheme 4 illustrates the remarkable effect that the nitrile
substituent has on the skeletal rearrangement. Lewis acid
coordination to the nitrile not only facilitates fragmentation of
the [2.2.2]-bicycle by electron donation from the ketal moiety
but also stabilizes the intermediate cyclopentadienyl anion, thus
suppressing the decomposition pathways that plagued the
thermal rearrangement.
With the nitrile installation and rearrangement conditions
determined, we sought to understand how the C18/C19 gem-
dimethyl group and C12−C13 olefin impacted the metal-
catalyzed rearrangement (Scheme 5). Application of the
optimized conditions to gem-dimethyl substituted triflate 22
resulted in a mixture of four products in low yields delivering
the desired [3.2.1]-bicycle 23 and cyclobutane 24 as an
inseparable mixture, along with two unconfirmed imine
products. This result was in stark contrast to the clean
reactivity observed for the unsubstituted triflate 14, suggesting
that the added steric bulk hindered the recombination of the
carbenium intermediate. Regardless, it is important to note that
product 23 represents the completed gibberellin carbon
skeleton, except carbon-17, which could be accessed by ketone
olefination. We postulated that reduction of the C12−C13
olefin might eliminate these undesired pathways. Toward that
end, cyclopentadiene 25 was accessed in two steps from 13b.
Subjecting 25 to the optimized cross-coupling conditions
delivered [2.2.2]-bicycle 26 as the major product along with a
small amount of the rearranged [3.2.1]-bicycle 27, confirming
that the Lewis acid catalyzed skeletal rearrangement can still
proceed in the absence of the C12−C13 olefin. Furthermore,
26 could be easily converted to 27 with zinc triflate at 130 °C,
C
DOI: 10.1021/acs.orglett.8b01031
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
exemplifying the utility of zinc as a Lewis acid catalyst for the
rearrangement. Conjugated ketone 21b was also selectively
reduced and converted to the gem-dimethyl containing triflate,
which failed to react under the cross-coupling conditions. The
added steric hindrance from the reduction of the bicycle
prevented engagement with the palladium catalyst and suggests
that the struggles in converting 22 to 23 were primarily due to
decomposition of the triflate starting material. While removal of
the C12−C13 olefin suppressed this decomposition, it could
not overcome the inherent steric limitations of the gem-
dimethyl substituted substrates.
AUTHOR INFORMATION
Corresponding Author
■
*
E-mail: njardars@email.arizona.edu.
ORCID
Jon T. Njardarson: 0000-0003-2268-1479
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
Financial support was provided by The University of Arizona
and the National Science Foundation (CHE-1266365).
■
Having thoroughly explored the skeletal rearrangement, we
set out to perform postrearrangement modification of the
gibberellin core. Selective reduction of 2 with Wilkinson’s
catalyst afforded 27, demonstrating that the C12−C13 olefin
can be removed pre- or postrearrangement. Removal of the
dimethoxy-ketal moiety of 27 with samarium diiodide
unexpectedly enabled a retro-rearrangement, affording a
mixture of [2.2.2] and [3.2.1] deketalized products in a 1:1
ratio, suggesting that the nitrile-activated cyclopentadiene must
first be removed to achieve clean deketalization. Studies are
ongoing to further functionalize the gibberellin core and reduce
the cyclopentadiene prior to ketal cleavage and olefination.
Looking toward the future, we envision that adduct 2, which
is available in 9 steps and 38% overall yield from commercially
available materials, represents an ideal oxidase phase starting
REFERENCES
■
(1) Hartsuck, J. A.; Lipscomb, W. N. J. Am. Chem. Soc. 1963, 85,
3414.
2) (a) Mander, L. N. Chem. Rev. 1992, 92, 573. (b) Mander, L. N.
Nat. Prod. Rep. 2003, 20, 49.
3) MacMillan, J. Nat. Prod. Rep. 1997, 14, 221.
4) (a) Corey, E. J.; Danheiser, R. L.; Chandrasekaran, S.; Keck, G.
E.; Gopalan, B.; Larsen, S. D.; Siret, P.; Gras, J. L. J. Am. Chem. Soc.
1978, 100, 8034. (b) Corey, E. J.; Danheiser, R. L.; Chandrasekaran,
S.; Siret, P.; Keck, G. E.; Gras, J. L. J. Am. Chem. Soc. 1978, 100, 8031.
5) (a) Corey, E. J.; Munroe, J. E. J. Am. Chem. Soc. 1982, 104, 6129.
b) Corey, E. J.; Guzman-Perez, A.; Loh, T. P. J. Am. Chem. Soc. 1994,
16, 3611.
6) (a) Lombardo, L.; Mander, L. N.; Turner, J. V. J. Am. Chem. Soc.
980, 102, 6626. (b) Hook, J. M.; Mander, L. N.; Urech, R. J. Am.
(
(
(
(
(
1
(
1
1
3
point for implementation of a two-phase synthesis of the
gibberellin family of natural products. The dimethoxy ketal
moiety not only enables the IMDA and skeletal rearrangement
but also serves as a convenient blocking group; the olefin and
carbonyl groups will enable the application of various oxidation
protocols, providing access to higher order gibberellins, such as
GA3. Additionally, this will allow for the generation of
unnatural gibberellins that may possess interesting biological
properties.
In summary, we have implemented a concise synthesis of the
gibberellin core. The synthesis hinges on an oxidative
dearomatization/Diels−Alder cascade and Lewis acid catalyzed
skeletal rearrangement to forge the cyclopentane-fused [3.2.1]-
bicycle and provides efficient access to this biologically
important scaffold. This novel Lewis acid catalyzed rearrange-
ment is robust and may prove broadly useful for the synthesis
of complex natural products.
Chem. Soc. 1980, 102, 6628. (c) Hook, J. M.; Mander, L. N.; Urech, R.
J. Org. Chem. 1984, 49, 3250.
(7) Nagaoka, H.; Shimano, M.; Yamada, Y. Tetrahedron Lett. 1989,
3
(
1
(
0, 971.
8) Nuyttens, F.; Hoflack, J.; Appendino, G.; DeClercq, P. J. Synlett
995, 1995, 105.
9) (a) McGrath, N. A.; Bartlett, E. S.; Sittihan, S.; Njardarson, J. T.
Angew. Chem., Int. Ed. 2009, 48, 8543. (b) Morton, J. G. M.; Draghici,
C.; Kwon, L. D.; Njardarson, J. T. Org. Lett. 2009, 11, 4492.
(c) McGrath, N. A.; Binner, J. R.; Markopoulos, G.; Brichacek, M.;
Njardarson, J. T. Chem. Commun. 2011, 47, 209. (d) Yang, Q.;
Njardarson, J. T.; Draghici, C.; Li, F. Angew. Chem., Int. Ed. 2013, 52,
8
648. (e) Yang, Q.; Draghici, C.; Njardarson, J. T.; Li, F.; Smith, B. R.;
Das, P. Org. Biomol. Chem. 2014, 12, 330. (f) Smith, B. R.; Njardarson,
J. T. Org. Lett. 2017, 19, 5316.
(10) (a) Mori, K.; Nakahara, Y.; Matsui, M. Tetrahedron 1972, 28,
3217. (b) Yamada, Y.; Nagaoka, H.; Kimura, M. Synthesis 1977, 1977,
581. (c) Monti, S. A.; Chen, S. C.; Yang, Y. L.; Yuan, S. S.; Bourgeois,
O. P. J. Org. Chem. 1978, 43, 4062.
(
11) (a) Resek, J. E. J. Org. Chem. 2008, 73, 9792. (b) Laplace, D. R.;
Verbraeken, B.; Van Hecke, K.; Winne, J. M. Chem. - Eur. J. 2014, 20,
53. (c) Hosoya, T.; Sumi, K.; Doi, H.; Wakao, M.; Suzuki, M. Org.
Biomol. Chem. 2006, 4, 410.
12) (a) Zhang, Z.; Wang, J.; Li, J.; Yang, F.; Liu, G.; Tang, W.; He,
ASSOCIATED CONTENT
2
■
*
S
Supporting Information
(
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b01031.
W.; Fu, J. J.; Shen, Y. H.; Li, A.; Zhang, W. D. J. Am. Chem. Soc. 2017,
139, 5558. (b) Reza, A. F. G. M.; Higashibayashi, S.; Sakurai, H. Chem.
-
Asian J. 2009, 4, 1329. (c) Miyoshi, H.; Miki, M.; Hirano, S.; Shimizu,
A.; Kishi, R.; Fukuda, K.; Shiomi, D.; Sato, K.; Takui, T.; Hisaki, I.;
Nakano, M.; Tobe, Y. J. Org. Chem. 2017, 82, 1380. (d) Chandra, A.;
Johnston, J. N. Angew. Chem., Int. Ed. 2011, 50, 7641. (e) Drechsler,
U.; Hanack, M. Synlett 1998, 1998, 1207.
(13) (a) Ishihara, Y.; Baran, P. S. Synlett 2010, 2010, 1733. (b) Yuan,
C.; Jin, Y.; Wilde, N. C.; Baran, P. S. Angew. Chem., Int. Ed. 2016, 55,
Experimental procedures; spectral and X-ray crystallog-
raphy data (PDF)
Accession Codes
CCDC 1821698 contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge
via www.ccdc.cam.ac.uk/data_request/cif, or by emailing data_
request@ccdc.cam.ac.uk, or by contacting The Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2
8
280.
1EZ, UK; fax: +44 1223 336033.
D
DOI: 10.1021/acs.orglett.8b01031
Org. Lett. XXXX, XXX, XXX−XXX