Cyclopropene cycloadditions with annulated furans: Total synthesis of (+)- and (-)-frondosin B and (+)-frondosin A

Article abstract of DOI:10.1021/ja413106t

The asymmetric total syntheses of the natural products (+)- and (-)-frondosin B and (+)-frondosin A are reported based on a diastereoselective cycloaddition between tetrabromocyclopropene and an annulated furan to provide a highly functionalized common bu

Full text of DOI:10.1021/ja413106t

Article
pubs.acs.org/JACS
Cyclopropene Cycloadditions with Annulated Furans: Total Synthesis
of (+)- and (−)-Frondosin B and (+)-Frondosin A
E.Zachary Oblak,† Michael D. VanHeyst,† Jin Li, Andrew J. Wiemer, and Dennis L. Wright*
Department of Pharmaceutical Sciences, University of Connecticut, 69 N. Eagleville Road, Storrs, Connecticut 06269, United States
S
* Supporting Information
ABSTRACT: The asymmetric total syntheses of the natural products
(+)- and (−)-frondosin B and (+)-frondosin A are reported based on a
diastereoselective cycloaddition between tetrabromocyclopropene and
an annulated furan to provide a highly functionalized common
building block. The bridged bicyclic intermediate could be stereo- and
chemoselectively manipulated to produce the two structurally distinct
members of the frondosins. Both syntheses feature regioselective
palladium-coupling reactions and an unprecedented phosphine-
mediated ether bridge cleavage. Surprisingly, the planned enantiose-
lective synthesis of frondosin B led to the opposite epimer of the
natural product, suggesting an unusual late stage stereoinversion at C8.
Frondosin A, but not frondosin B, was shown to have selective
antiproliferative activity against several B-cell lines.
The frondosin family1 of marine-derived meroterpenoid natural
products (Figure 1) represent an intriguing array of compounds
at this center. The initial discrepancy in assignments was later
explained by MacMillan3j who established that an inversion of
the C8-center had occurred during the Trauner synthesis. It
was suggested that an unusual stereochemical relay process had
occurred during a key palladium-catalyzed cyclization of the B-
ring to ultimately deliver the S-configuration at C8. Frondosin
A has also been the subject of synthetic studies2 including one
asymmetric2a synthesis, which established that the related C8-
center also naturally occurs as the R-configuration.
Several of the frondosins were noted for their potential to act
as antagonists toward the interleukin-8 receptor (IL-8), and
frondosin B has also been reported to be a modest inhibitor of
the serine/threonine kinase protein kinase C (PKC). The
related terpene liphagal was shown to be a relatively potent and
isozyme-selective inhibitor of the lipid kinase phosphatidylino-
sitol-3-kinase (PI-3K). However, there have been few
subsequent investigations into the biological activity of the
frondosins or their analogs.1a Herein, we present an asymmetric
total synthesis of both enantiomers of frondosin B and the
natural enantiomer of frondosin A through a unified strategy
featuring a complexity building cycloaddition between an
annulated furan and a perhalocyclopropene. The syntheses also
feature a new protocol for opening of a pivotal oxa-bridged
intermediate. Central to the ultimate design was a flexible
dibromoenone intermediate derived from a [4 + 3]
cylcoaddition reaction with an annulated furan. Interestingly,
we observed the same stereochemical inversion during our
frondosin B synthesis as noted in the Trauner system. This
appears to involve a highly unusual stereochemical inversion of
Figure 1. Meroterpenoid natural products.
possessing a bicyclo[5.4.0] undecene core with significant
structural diversity introduced through alternative annulations
with an appended hydroquinone moiety.2 The structure of
frondosin B3 is also reminiscent of the marine natural product
liphagal4 as both incorporate an appended arene system in the
form of a fused benzofuran motif. Frondosin B has been a
popular target for total synthesis which ultimately led to
assignment of the absolute stereochemistry. The first two
syntheses of (+)-frondosin B by Danishefsky3a,b and
Trauner3c,d produced conflicting assignments for the single
stereogenic center at C8. A subsequent asymmetric synthesis by
Ovaska3g provided further confirmation for the R-configuration
Received: December 24, 2013
Published: February 27, 2014
© 2014 American Chemical Society
4309
dx.doi.org/10.1021/ja413106t | J. Am. Chem. Soc. 2014, 136, 4309−4315

Journal of the American Chemical Society
Article
a carbocation intermediate. We also report the second
asymmetric total synthesis of frondosin A which proceeds
directly without the observed inversion process. Frondosin A
was also shown to have selective antiproliferative action toward
B-cell lines.
Scheme 2. TBCP Cycloaddition with Chiral Annulated
Furan
a
The retrosynthetic analyses for both natural products
targeted late stage oxabridged intermediates 1 and 2 that
contained the intact carbocyclic framework of the respective
natural products with a temporary bridging ether spanning
C7−C11 of the B-ring (Scheme 1). These intermediates were
Scheme 1. Retrosynthesis of (+)-Frondosin A and
(+)-Frondosin B
a
Reagents and conditions: (a) (S,S)-Noyori (0.01 equiv),
HCO2H:Et3N, rt, 7 days; (b) TBSCl (1.1 equiv), imidazole (1.5
equiv), DMAP (0.10 equiv), DCM, rt, 5 h, 91% (2 steps), 98% ee; (c)
(i) 5 (1.0 equiv), 1,4-dioxane, rt to 90 °C, 5 h; (ii) AgNO3 (2.0 equiv),
2:1 acetone:H2O, rt, 8 h, 77% (1 step); (d) TBSOTf (1.0 equiv), 2,6-
lutidine (1.5 equiv), DCM, −78 °C, 1 h, 98%.
tetrabromides in excellent overall yield. Pleasingly, the facial
selectivity was high with the C4-silyloxy exerting a strong
directing effect on the initial Diels−Alder reaction leading to a
syn-relationship between the protected alcohol and the oxa
bridge. The resulting crude tetrabromides could be directly
treated with aqueous silver nitrate, proceeding through a
pseudosymmetrical tribromoallyl cation 7 that can lead to the
regioisomeric enones 8a and 8b (8b shown in Supporting
Information only), depending upon which terminus of the
cation water attacks. The bridgehead substitution at C11
provided a reasonable directing effect which favored the
formation of 8a as the major regioisomer (r.r.=3.3:1). This
one-step conversion of furan 6 to bridged intermediate 8a was
highly efficient and was amenable to a gram-scale production of
the dibromoenone.
The two vinylic bromides in these systems display
substantially different levels of reactivity that allows for the
controlled, stepwise introduction of functionality. We were able
to take advantage of this versatile array to easily annulate the
CD-ring benzofuran substructure (Scheme 3). Suzuki cross-
coupling of 3 with the aryl trifluoroborate salt 99 occurred
exclusively at the β-bromide to deliver the phenols 10 as a
(∼1:1) mixture of noninterconverting atropisomers. The use of
the trifluoroborate salt was critical in the cross-coupling to this
somewhat hindered bromide. Application of more traditional
borates derived from the electron-rich phenol led to the
formation of deborinated byproducts, producing only 55% yield
of the coupled product. Exposure of the crude mixture of
phenols to stoichiometric copper(I)iodide led to a smooth
coupling of the phenolic hydroxyl group to the remaining
bromide, thus giving the annulated benzofuran 11 in excellent
overall yield.10 Frondosin B contains a single stereogenic center
at C8, typically a challenging type of center to set on a seven-
membered ring. However, the temporary ether bridge induces
rigidity into this otherwise flexible cycloheptyl ring and
effectively differentiates the two faces of the carbocycle. This
envisioned to arise from a common precursor, dibromoenone
3, the product of a diastereoselective cycloaddition between
tetrabromocyclopropene (TBCP, 5) and an annulated furan
derivative such as 4. TBCP is readily available in a one step
transformation from the commercially available tetrachlorocy-
clopropene, by reaction with boron tribromide. We have
worked considerably with the TBCP adducts of simple
substituted furans5 but had not previously attempted to extend
this formal [4 + 3] cycloaddition to annulated furans such as 4.
Although the analogous reaction between annulated furans and
oxyallyl cations is known to be difficult, with simple
electrophilic substitution products arising,6 it was hoped that
high reactivity of TBCP toward direct cycloaddition with furans
would compensate for the increased steric demands.7 More-
over, we were attracted to the notion that an asymmetric center
at C4 of the A-ring could control the diastereoselectivity of the
cycloaddition and that this information could be used to
establish the other stereogenic centers located on the seven-
membered B-rings. The commercially available ketone 4
seemed an ideal starting point, and in fact, asymmetric
reduction of 4 to the R-alcohol has been previously reported
by Noyori.8
We required the S-alcohol and found that it could be
prepared with excellent selectivity through reduction of ketone
4 with the (S,S)-Noyori transfer hydrogenation catalyst
(Scheme 2).8b The alcohol was converted to the silyl ether 6
and condensed with TBCP to give a mixture of regioisomeric
4310
dx.doi.org/10.1021/ja413106t | J. Am. Chem. Soc. 2014, 136, 4309−4315

Journal of the American Chemical Society
Article
a
Scheme 3. Total Synthesis of (−)-Frondosin B
a
Reagents and conditions: (a) 9 (1.25 equiv), Pd(PPh3)4 (0.10 equiv), Cs2CO3 (1.25 equiv), 10:1 THF:H2O, 70 °C, 4 h; (b) CuI (1.5 equiv), 1:1
MeCN:Et3N, 80 °C, 18 h, 77% (2 steps); (c) Ph3PCH3Br (3.0 equiv), n-BuLi (3.0 equiv), THF, 0 °C, 1.5 h 82%; (d) PtO2 (0.20 equiv), H2 (1 atm),
PhH, 3 h, 95%; (e) HF·pyridine (20 equiv), THF, 0 °C to rt, 10 h; (f) Dess-Martin periodinane (2.0 equiv), NaHCO3 (10 equiv), DCM, rt, 1.5 h,
89% (2 steps); (g) Bu3P (1.5 equiv), DCE, rt, 5 min, 98%; (h) Pd/C (0.10 equiv), H2 (1 atm), EtOAc, 2 h, 97%; (i) MeMgBr (3.2 equiv), CeCl3
(3.2 equiv), THF, 0 °C, 15 min; (j) TiCl4 (1.0 equiv), Me2Zn (2.0 equiv), DCM, 0 °C, 15 min, 74% (2 steps); (k) NaSEt (20 equiv), DMF, 140 °C,
5 h, 95%.
facial bias was exploited by initial Wittig condensation to give
the exo-methylene derivative that was stereoselectively hydro-
genated from the exo-face to produce 12 as a single
diastereomer, which was confirmed though a single crystal X-
ray structure determination. Deprotection of the silyl ether
preceded oxidation of the resulting allylic alcohol to give enone
1.
With completion of the carbocyclic core of frondosin B,
attention turned to opening of the ether bridge and completion
of the synthesis. The first productive ring-opening was realized
with a combination of TMSOTf and Et3N (Scheme 4). This
Lewis acid-assisted opening did produce a ring opened product,
however the reaction proceeded to give phenol 18. In an
attempt to impede the rapid aromatization of the A-ring, we
first reduced the enone olefin and then treated the saturated
ketone 19 with TMSOTf/Et3N. To our delight, a productive
ring-opening had occurred and generated alcohol 20.
Initial attempts to effect removal of the secondary alcohol of
20 with pyridine and Ac2O were hampered by a preference for
elimination reactions to yield a C8-olefin 22, thus destroying
the stereogenicity of this center (Scheme 5). Upon converting
20 to its corresponding assorted thioxoesters 23, radical
deoxygenation failed to deliver the desired product as a ring
contracted compound 24 formed preferentially, independent of
the various esters examined. The contraction is proposed to
occur though an intermediate cyclopropylcarbinyl radical S6
Scheme 5. Attempts to Deoxygenate 20
Scheme 4. Initial Ring-Opening Attempts
4311
dx.doi.org/10.1021/ja413106t | J. Am. Chem. Soc. 2014, 136, 4309−4315

Journal of the American Chemical Society
Article
produced by a transannular addition to the furan olefin. It is
likely that the driving force for this radical rearrangement is the
formation of the more stable six-membered B-ring.
Scheme 6. Isotopic Labeling Studies and a Working Model
for Stereochemical Inversion
These difficulties prompted us to develop an interesting
phosphine-mediated deoxygenation protocol that provided
complete regiocontrol over the final elimination reaction.
Addition of tributylphosphine to the enone 1 led to direct
conversion to the triene 15 in high yield. A probable
mechanism for this deoxygentation, akin to the phosphine-
induced reduction of epoxides to olefins, involves an initial
conjugate addition of the phosphine to the more accessible exo-
face to produce an initial enolate followed by ejection of the β-
disposed ether.11 The elimination yields a transient betaine 13
that collapses through the intermediacy of the oxaphosphatane
14 to deliver the desired olefin.
Regioselective hydrogenation of the disubstituted alkene
produced enone 16, a late stage intermediate in the Trauner
synthesis.4 Although we were pleased that 16 matched the
c,d
reported spectral data, it was surprising that our intermediate
also matched Trauner’s reported optical rotation, as that
compound was later shown to possess the incorrect S-
configuration at C8, ultimately leading to the antipode of the
natural product. Compound 16 was advanced to frondosin B
according to Trauner’s protocol and, as reported, possessed the
opposite rotation as that reported for (+)-frondosin B. Our
initial concern was that the assignment of silyl alcohol 6 was
incorrect as it was based on NMR shift analysis of the
corresponding Mosher esters. Fortunately, a single crystal
structure determination of dibromoenone 8a allowed for the
unambiguous assignment of the absolute stereochemistry
through analysis of the Flack parameter (Scheme 2).12 The
structure confirmed that the original assignment of 8a, which in
consideration of the X-ray structure of 12 (Scheme 3),
unambiguosly established that the C8 center possessed the
correct R-configuration in 12 and was somehow inverted to the
incorrect S-configuration during the final steps of the synthesis.
We first set out to determine if the configuration at C8 had
somehow been inverted in the processing of 12 to the Trauner
intermediate 16, perhaps through equilibration of the methyl
group to the more stable exo-orientation in one of the bridged
intermediates. Strong NOE enhancements between the C8
methyl group and the C6-vinylic protons in the intermediates
leading to 1 confirmed that the methyl group had remained in
the endo-position.
intercepted by chloride). The strong propensity to contract
the B-ring as seen during the attempted deoxygenation of 20 to
24 provides support for this pathway. Critically, this migration
generates a new stereogenic center at C9 and acts to preserve
the stereochemical information in the molecule. In compound
30, the planarized C8 carbon should adopt a conformation that
orients the methyl group outside the ring system such that
when the final nucleophilic attack at C4 triggers reformation of
the C7−C8 bond, the migration is stereoselective. In this
manner, the initial C8 configuration controls the stereo-
chemistry produced at C9 of 30, which in turn dictates the
configuration when the C8 center is re-established. The
intermediacy of a C8 cation allows for the process to occur
with overall stereochemical inversion. Based on this hypothesis,
it appeared straightforward to prepare the natural configuration
using the antipode of silyoxy 6. Utilizing the (R,R)-Noyori
catalyst for the reduction of ketone 4 produced the
enantiomeric compound (+)-6 (after protection as its silyl
ether) which could be taken through the same sequence to
deliver (+)-frondosin B.
To further probe the configuration of this center during the
final stages of the synthesis, a series of deuterium labeling
experiments were conducted (Scheme 6). Reduction of the
dieneone 15 under a deuterium atmosphere only showed
incorporation at C6−7 and no incorporation at the C8 position
of 25. Likewise, reduction of the exomethylene derived from 11
with deuterium gas gave labeled intermediate 26 that was
converted to the Trauner-like intermediate 27 without
exchange of the methine deuterium for hydrogen. As this
surprising inversion of configuration occurred in both our
synthesis and Trauner’s, it seems most likely that it is related to
a common process, namely the conversion of the C4 ketone 16
to O-methyl frondosin B through an intermediate carbonium
ion. Interestingly, there is precedent for an inversion of the
analogous C8-center in liphigal under cationic reaction
conditions.13 Our working hypothesis centers around the
initial, highly delocalized carbocation 29 that may undergo a
stereoselective, reversible alkyl shift producing an intermediate
ring contracted cation 30 (which could be reversibly
With a successful synthesis of frondosin B complete, our
attention turned to the total synthesis of the related frondosin
A from the common intermediate 3. This synthesis would also
provide an opportunity to probe the hypothesis that the
extended conjugation from the benzofuran allows a C4 cationic
center to invert stereochemistry at C8. The saturation of the
C9−C10 olefin of frondosin A should effectively eliminate this
pathway.
Again taking advantage of the differential reactivity of the two
vinylic bromides of 3, the 2,5-dimethoxy arene moiety was
introduced at the β-position of the enone 3 through Suzuki−
Miyaura cross-coupling reaction (Scheme 7). With the
protected o-phenol, it was not necessary to utilize the
trifluoroborate derivative in the coupling as was necessary in
the synthesis of frondosin B. Exploiting the curvature
embedded in the oxabicyclo[3.2.1]octadiene core, exposure of
the resulting crude material to catalytic hydrogenation
conditions, in the presence of triethylamine, produced the
saturated ketone 32 in 76% yield over the two steps and with
4312
dx.doi.org/10.1021/ja413106t | J. Am. Chem. Soc. 2014, 136, 4309−4315

Journal of the American Chemical Society
Article
a
Scheme 7. Total Synthesis of (+)-Frondosin A
a
Reagents and conditions: (a) 2,5-dimethoxyphenyl boronic acid (1.25 equiv), Pd(PPh3)4 (0.10 equiv), Cs2CO3 (1.25 equiv), 10:1 THF:H2O, 70
°C, 4 h; (b) Pd/C (0.10 equiv), Et3N (7.0 equiv), H2 (1 atm), THF, 24 h, 76% (2 steps); (c) Ph3PCH3Br (3.0 equiv), n-BuLi (3.0 equiv), THF, 0
°C, 1.5 h, 80%; (d) SeO2 (0.50 equiv), t-BuOOH (3.0 equiv), DCM, rt, 16 h, 81%; (e) [Rh(nbd)(diphos-4)]BF4 (0.20 equiv), THF (0.025 M), H2
(600 PSI), rt, 2 h, 98%; (f) TBSOTf (1.0 equiv), 2,6-lutidine (1.5 equiv), DCM, −78 °C, 1 h; (g) TBAF (1.2 equiv), THF, 0 °C to rt, 6 h; (h) Dess-
Martin periodinane (2.0 equiv), NaHCO3 (10 equiv), DCM, 0 °C to rt, 1.5 h, 86% (3 steps); (i) Ti(i-OPr)4 (3.0 equiv), Me3P (3.0 equiv), THF, 70
°C, 3 h, 86%; (j) MeMgBr (6.4 equiv), CeCl3 (3.2 equiv), THF, 0 °C, 15 min; (k) TiCl4 (1.0 equiv), Me2Zn (2.0 equiv), DCM, 0 °C, 15 min; (l)
TBAF (8.0 equiv), 65 °C, 36 h; (m) Pd/C (0.10 equiv), H2 (1 atm), MeOH, rt, 5 h, 61% (4 steps); (n) p-bromobenzoyl chloride (3.0 equiv), Et3N
(3.0 equiv), DMAP (1.0 equiv), DCM, 45 °C, 3 h, 80%; (o) Dess-Martin periodinane (2.0 equiv), NaHCO3 (10 equiv), DCM, 0 °C to rt, 1.5 h,
95%; (p) Mg (16 equiv), TiCl4 (4 equiv), DCM, 60 °C, 6 h, 55%; (q) (i) AgO (2.0 equiv), 6N HNO3 (3.0 equiv), 1,4-dioxane, rt, 15 min; (ii) Pd/C
(0.05 equiv), H2 (1 atm), DCM, 15 min, 84%.
complete diastereoselectivity.14 A subsequent Wittig condensa-
tion gave the corresponding exocyclic methylene derivative.
Further reliance on the facial bias that the temporary ether
bridge provides was illustrated when the alkene was subjected
to allylic oxidation conditions with SeO2 to provide the key
alcohol 33 with high diastereoselectivity, producing 97:3
mixture of exo:endo allylic alcohols.15
The exo-alcohol installed at C9 of intermediate 33 would not
only serve as a precursor for the exocyclic olefin in the natural
product but also employed to reinforce the expected exo-mode
reduction of the methylene group to produce the endo-methyl
group at C8. To probe the natural preference for hydro-
genation, alcohol 33 was subjected to hydrogenation in the
presence of Pd/C and PtO2. In both cases, a high-yielding
reduction occurred to give a 4:1 mixture of diastereomers
34a:34b.
To evaluate a hydroxyl-directed hydrogenation, 33 was
exposed to [Rh(nbd)(diphos-4)]BF4 catalyst at 600 psi of H2
pressure.16 Interestingly, it was found that at a higher catalyst
load (20 mol %), a 10:1 mixture of diastereomers was
produced. Further optimization revealed that the concentration
of the reaction had a dramatic effect on the diastereomeric
ratio. Reduction at 0.025 M in THF led to a completely
diastereoselective hydrogenation to give 34a exclusively.
With the appropriate stereocenters of frondosin A intact, the
stage was set for cleavage of the bridging ether. Prior results
from the total synthesis of frondosin B suggested that ether
bridge cleavage could be effectively accomplished on late stage
intermediate 2 using trialkylphosphine reagents. In efforts to
advance 34a to 2, a selective deprotection scheme was utilized
to exploit the difference in reactivity between the C4 allylic
alcohol and the C9 secondary alcohol. Protection with TBSOTf
followed by selective removal of the allylic silyl ether with
TBAF provided the requisite alcohol that oxidized with Dess-
Martin periodinane (DMP) to give the α,β-unsaturated enone
2.17
Initial attempts at ether bridge cleavage of 2 revealed that this
system was significantly less reactive toward phosphine
nucleophiles, compared to a corresponding frondosin B
intermediate. As Lewis acids have often been employed to
facilitate the opening of oxabicyclic intermediates, we
investigated whether compatible additives would catalyze the
desired transformation (Table 1).18 Pleasingly, when enone 2
was exposed to a mixture of In(III)Cl and n-Bu3P (entry 4), a
productive ring-opening/elimination reaction occurred, how-
ever the isolated yield was only 25% along with 10% of an
unidentified phosphine-containing adduct. Increasing the
nucleophilicity of the phosphine reagent greatly impacted the
isolated yield of ring-opened diene 35. A combination of Me3P
with Ti(iOPr)4 proved particularly effective (entries 8,9),
producing the formal deoxygenation product 35 in an isolated
yield of 86%.
4313
dx.doi.org/10.1021/ja413106t | J. Am. Chem. Soc. 2014, 136, 4309−4315

Journal of the American Chemical Society
Article
proliferation of a panel of leukocytes (Table 1).22 (+)-Frondo-
sin A, but not frondosin B, inhibited proliferation of
lymphocytic cell lines suggesting the necessity of the free
hydroquinone moiety for activity. This was further supported
by the finding that the dimethyl ether analog of frondosin A
(39) was also inactive toward the cell lines. The potency of
(+)-frondosin A was an order of magnitude higher for B cell
lines compared to the T-cell lines that were tested. Surprisingly,
(+)-frondosin A inhibited Ba/F3 cells23 that express oncogenic
IL-7 receptor mutations more potently than the IL-3-depend-
ent wild-type Ba/F3 cells. Taken together, these data suggest
that in addition to IL-8 in neutrophils, frondosins may have
inhibitory activity against other leukocyte cytokine receptors.
Further evaluations of the biological activity are currently
underway.
Table 1. Trends in Lewis-Acid Assisted Ether Cleavage
a
b
c
c
entry
Lewis acid
phosphine
solvent/temp yield 2
yield 35
1
2
3
4
5
6
7
8
9
−
−
InCl3
InCl3
InCl3
n-Bu3P
n-Bu3P
n-Bu3P
n-Bu3P
Me3P
DCE/23 °C
DCE/80 °C
DCE/23 °C
THF/70 °C
THF/70 °C
THF/70 °C
THF/70 °C
THF/70 °C
THF/70 °C
94%
56%
70%
−
−
−
−
−
−
−
−
−
25%
55%
62%
70%
77%
86%
Ti(OiPr)4
Ti(OiPr)4
Ti(OiPr)4
n-Bu3P
t-Bu3P
Me3P
a
d
Table 2. Inhibition of Cell Growth
Ti(OiPr)4
Me3P
a
b
6 equiv of Lewis acid used unless otherwise stated. 3 equiv of
trialkylphosphines used in all reactions. Isolated yield. Performed
72 h (GI50 values in μM)
O-methyl-
c
d
with 3 equiv of Lewis acid.
cell type
B-cell
cell line
(+)-Fro A (+)-Fro B
(+)-Fro A
Daudi
CH12
2.62
2.05
4.49
0.87
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Prior conversion of the C4 carbonyl of 35 to the geminal
dimethyl group was crucial to eliminate the cross-conjugated
dienone and allow for a selective saturation of the C6−C7
alkene. The dimethyl intermediate was difficult to isolate given
the nonpolar nature, and thus the crude material was
immediately deprotected in the presence of TBAF, at elevated
temperature, followed by traditional hydrogenation conditions
(Pd/C, MeOH, 1 atm H2) to produce intermediate 36.
The intermediacy of alcohol 36 provided an ideal
opportunity to confirm the absolute stereochemistry. Alcohol
36 was reacted with 4-bromobenzoyl chloride to generate
highly crystalline benzoate 37 that yielded thin plate crystals. X-
ray crystallographic analysis of 37 allowed for the unambiguous
conformation of the absolute stereochemistry.19
Ba/F3 WT
Ba/F3 WT mut IL7R
(88i)
Ba/F3 WT mut IL7R
(GCins)
1.58
N/A
N/A
T-cell
HuT 78
Jurkat
17.23
23.73
25.65
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
Molt-4
K562
myeloid
a
N/A - not active at 33 μM.
Overall, we have developed a synthesis of both the natural
To complete the synthesis of frondosin A, alcohol 36 was
oxidized with DMP to late stage intermediate 38 matching the
spectral data reported by the Ovaska group.2b Conversion of
the C9 carbonyl to the required exo-methylene group was
attempted with a variety of standard olefination procedures
(Wittig, Petasis) but only returned unreacted starting material.
The desired olefination of the hindered B ring carbonyl carbon
was ultimately accomplished by applying the bimetallic TiCl4-
Mg promoted methylene transfer reaction to deliver known
dimethoxy-frondosin A 39.20
and unnatural enantiomers of frondosin B and the natural
enantiomer of frondosin A utilizing a highly selective
perhalocyclopropene/furan cycloaddition reaction to access
the central core of the natural product. The versatile common
intermediate could be converted to two structurally distinct
members of this class of marine natural product. The synthesis
has also suggested the possibility of a very unusual inversion of
configuration in a carbocationic intermediate that will warrant
further investigation.
Initial attempts at final deprotection of the dimethoxy-arene
39 with CAN/Na2S2O4, as disclosed in previous total syntheses
of frondosin A, resulted in a less than effective deprotection
with only around 35% recovered product. Reports of
deprotection of dimethoxy frondosin A with BBr3 are known
to result in decomposition, while reports of deprotection with
sodium ethanethiolate only yielded monodeprotected products.
A significant improvement was made through a two-step
oxidative (AgO, HNO3)/reductive (Pd/C, H2) procedure to
provide the enantiomerically pure frondosin A in a 84% yield
over the two steps and possessing optical rotation reflective of
the natural product [α]D22 = +29.8 (c = 0.25, MeOH); lit. [α]D
= +31.5 (c = 0.25, MeOH).21 Importantly, no inversion at C8
was observed, supporting the role of extended conjugation in
the unusual results obtained in the frondosin B synthesis.
As the frondosins were suggested to exhibit inhibitory action
on signaling through IL-8 receptor in blood cells, we evaluated
both (+)-frondosin A and (+)-frondosin B for effects on
ASSOCIATED CONTENT
* Supporting Information
Experimental procedures and spectral data are provided. This
material is available free of charge via the Internet at http://
pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
dennis.wright@uconn.edu
■
Author Contributions
†These authors contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We are grateful to the National Science Foundation (CHE-
0616760) for generous support of this work.
■
4314
dx.doi.org/10.1021/ja413106t | J. Am. Chem. Soc. 2014, 136, 4309−4315

Journal of the American Chemical Society
Article
(17) (a) Nakata, T.; Fukui, M.; Oishi, T. Tetrahedron Lett. 1988, 29,
2219−2222. (b) Danishefsky, S. J.; Armistead, D. M.; Wincott, F. E.;
Selnick, H. G.; Hungate, R. J. Am. Chem. Soc. 1987, 109, 8117−8119.
(18) Chiu, P.; Lautens, M. Top. Curr. Chem. 1997, 190, 1−85.
(19) We are grateful to Dr. Victor Day, The University of Kansas and
the National Science Foundation (CHE-0923449) for X-ray crystallo-
graphic analysis of compound 14.
REFERENCES
■
(1) (a) Patil, A. D.; Freyer, A. J.; Killmer, L.; Offen, P.; Carte, B.;
Jurewicz, A. J.; Johnson, R. K. Tetrahedron 1997, 53, 5047−5060.
(b) Hallock, Y. F.; Cardellina, J. H., II; Boyd, M. R. Nat. Prod. Lett.
1998, 11, 153−160.
(2) For syntheses of frondosin A see: (a) Trost, B. M.; Hu, Y.;
Horne, D. B. J. Am. Chem. Soc. 2007, 129, 11781−11790. (b) Li, X.;
Keon, A. E.; Sullivan, J. A.; Ovaska, T. V. Org. Lett. 2008, 10, 3287−
3290. (c) Mehta, G.; Likhite, N. S. Tetrahedron Lett. 2008, 49, 7113−
7116.
(20) Yan, T.; Tsai, C.; Chien, C.; Cho, C.; Huang, P. Org. Lett. 2004,
6, 4961−4963.
(21) Snyder, C. D.; Rapoport, H. J. Am. Chem. Soc. 1972, 94, 227−
231.
(3) For syntheses of frondosin B see: (a) Inoue, M.; Frontier, A. J.;
Danishefsky, S. J. Angew. Chem., Int. Ed. 2000, 39, 761−764. (b) Inoue,
M.; Carson, M. W.; Frontier, A. J.; Danishefsky, S. J. J. Am. Chem. Soc.
2001, 123, 1878−1889. (c) Hughes, C. C.; Trauner, D. Angew. Chem.,
Int. Ed. 2002, 41, 1569−1572. Correction (c) Hughes, C. C.; Trauner,
D. Angew. Chem., Int. Ed. 2002, 114, 2331; Angew. Chem., Int. Ed. 2002,
41, 2227. (d) Hughes, C. C.; Trauner, D. Tetrahedron 2004, 60, 9675−
9686. (e) Kerr, D. J.; Willis, A. C.; Flynn, B. L. Org. Lett. 2004, 6, 457−
460. (g) Ovaska, T. V.; Sullivan, J. A.; Ovaska, S. I.; Winegrad, J. B.;
Fair, J. D. Org. Lett. 2009, 11, 2715−2718. (h) Olson, J. P.; Davies, H.
M. L. Org. Lett. 2008, 10, 573−576. Correction: Olson, J. P.; Davies,
H. M. L. Org. Lett. 2010, 12, 1144. (i) Mehta, G.; Likhite, N. S.
Tetrahedron Lett. 2008, 49, 7113−7116. (j) Reiter, M.; Torssell, S.;
Lee, S.; MacMillan, D. W. C. Chem. Sci. 2010, 1, 37−42.
(4) Marion, F.; Williams, D. E.; Patrick, B. O.; Hollander, I.; Mallon,
R.; Kim, S. C.; Roll, D. M.; Feldberg, L.; Van Soest, R.; Andersen, R. J.
Org. Lett. 2006, 8, 321−324.
(22) Zhang, J.; et al. Nature (London, U. K.) 2012, 481, 157−163 For
a full listing of contributing authors see the Supporting Information.
(23) Ba/F3 cell lines containing IL7R mutations were graciously
provided by C. Mullighan, St Jude Children’s Research Hospital.
(5) (a) Pelphrey, P. M.; Bolstad, D. B.; Wright, D. L. Synlett 2007,
2647−2650. (b) Pelphrey, P. M.; Orugunty, R. S.; Helmich, R. J.;
Battiste, M. A.; Wright, D. L. Eur. J. Org. Chem. 2005, 4296−4303.
(c) Pelphrey, P. M.; Abboud, K. A.; Wright, D. L. J. Org. Chem. 2004,
69, 6931−6933. (d) Batson, W. A.; Abboud, K. A.; Battiste, M. A.;
Wright, D. L. Tetrahedron Lett. 2004, 45, 2093−2096. (e) Orugunty,
R. S.; Wright, D. L.; Battiste, M. A.; Helmich, R. J.; Abboud, K. J. Org.
Chem. 2004, 69, 406−416.
(6) Rigby, J. H.; Pigge, F. C. Org. React. (N. Y., Engl. Transl.) 1997,
51, 351−478.
(7) Orugunty, R. S.; Ghiviriga, I.; Abboud, K. A.; Battiste, M. A.;
Wright, D. L. J. Org. Chem. 2004, 69, 570−572.
(8) (a) Ohkuma, T.; Hattori, T.; Ooka, H.; Inoue, T.; Noyori, R. Org.
Lett. 2004, 6, 2681−2683. For Noyori transfer hydrogenation catalyst
see: (b) Fujii, A.; Hashiguchi, S.; Uematsu, N.; Ikariya, T.; Noyori, R. J.
Am. Chem. Soc. 1996, 118, 2521−2522.
(9) For the preparation of 9 see: Zhang, Y.; Oblak, E. Z.; Bolstad, E.
S. D.; Anderson, A. C.; Jasinski, J. P.; Butcher, R. J.; Wright, D. L.
Tetrahedron Lett. 2010, 51, 6120−6122. For preparation of similar
trifluoroborate salts, see: (a) Park, Y. H.; Ahn, H. R.; Canturk, B.; Jeon,
S. I.; Lee, S.; Kang, H.; Molander, G. A.; Ham, J. Org. Lett. 2008, 10,
1215−1218. For a review of organotrifluoroborate coupling reactions,
see: (b) Molander, G. A.; Ellis, N. Acc. Chem. Res. 2007, 40, 275−286.
(10) For a similar closure, see: Martinez, E.; Martinez, L.; Estevez, J.
C.; Estevez, R. J.; Castedo, L. Tetrahedron Lett. 1998, 39, 2175−2176.
For a review of copper mediated closures, see: Ley, S. V.; Thomas, A.
W. Angew. Chem., Int. Ed. 2003, 42, 5400−5449.
(11) Bissing, D. E.; Speziale, A. J. J. Am. Chem. Soc. 1965, 87, 2683−
2690.
(12) Flack, H. D. Acta Crystallogr., Sect. A: Found. Crystallogr. 1983,
A39, 876−881.
(13) See conversion of 8a to 9b in: Pereira, A. R.; Strangman, W. K.;
Marion, F.; Feldberg, L.; Roll, D.; Mallon, R.; Hollander, I.; Andersen,
R. J. J. Med. Chem. 2010, 53, 8523−8533.
(14) Van Heyst, M. D.; Oblak, E. Z.; Wright, D. L. J. Org. Chem.
2013, 78, 10555−10559.
(15) Umbreit, M. A.; Sharpless, K. B. J. Am. Chem. Soc. 1977, 99,
5526−5528.
(16) Brown, J. M.; Chaloner, P. A.; Kent, A. G.; Murrer, B. A.;
Nicholson, P. N.; Parker, D.; Sidebottom, P. J. J. Organomet. Chem.
1981, 216, 263−276. For review see: Hoveyda, A. H.; Evans, D. A.;
Fu, G. C. Chem. Rev. (Washington, D. C.) 1993, 93, 1307−1370.
4315
dx.doi.org/10.1021/ja413106t | J. Am. Chem. Soc. 2014, 136, 4309−4315