Formal Synthesis of Solanoeclepin A: Enantioselective Allene Diboration and Intramolecular [2+2] Photocycloaddition for the Construction of the Tricyclic Core

Article abstract of DOI:10.1002/chem.201504894

An enantioselective synthesis of an intermediate in the Tanino total synthesis of solanoeclepin A has been developed. The key step was an intramolecular [2+2] photocycloaddition, which led to the tricyclo[5.2.1.0^{1, 6}]decane core in six steps. T

Full text of DOI:10.1002/chem.201504894

DOI: 10.1002/chem.201504894
Communication
&
Natural Product Synthesis
Formal Synthesis of Solanoeclepin A: Enantioselective Allene
Diboration and Intramolecular [2+2] Photocycloaddition for the
Construction of the Tricyclic Core
[
a]
[a]
[a]
[b]
Roel A. Kleinnijenhuis, Brian J. J. Timmer, Ginger Lutteke, Jan M. M. Smits,
[b]
[a]
[a]
RenØ de Gelder, Jan H. van Maarseveen, and Henk Hiemstra*
th
Dedicated to professor Henk Schenk on the occasion of his 76 birthday
length 1 mm), which feed on potato roots leading to major
losses to potato harvests in many countries. After the hatching
of the eggs (from cysts in the soil), the juvenile PCN penetrates
the roots and then retard potato growth.
Abstract: An enantioselective synthesis of an intermediate
in the Tanino total synthesis of solanoeclepin A has been
developed. The key step was an intramolecular [2+2] pho-
1
,6
tocycloaddition, which led to the tricyclo[5.2.1.0 ]decane
core in six steps. The first photosubstrate, prepared
through an indium-mediated Barbier-type reaction, gave
an excellent [2+2] cycloaddition, but it could not be ob-
tained in sufficient enantiopurity. The second photosub-
strate, prepared through an asymmetric allene diboryla-
tion in high enantiomeric excess, gave the [2+2] cycload-
dition product in high yield on irradiation at 365 nm on
The role of hatching agents is known for more than 80
years. The idea to use such signal substances as a method to
control PCN is attractive from an environmental point of view.
If the signal substance is applied to the soil, the eggs will
hatch. However, in the absence of potato plants the nemato-
des will starve within eight weeks, a neat way to clean the soil.
In 1940’s much research was directed at the elucidation of the
[1]
molecular structure of the hatching agent. The goal was final-
2
0 g scale in a flow system. Other important steps were
ly reached in 1992 in The Netherlands, when crystals were for-
tuitously obtained during an NMR measurement. The ensuing
X-ray diffraction study by Schenk and co-workers revealed the
the replacement of a boronate group at the quaternary
carbon by a vinyl group and diastereoselective cyclopro-
panation of an allylic alcohol.
[2]
spectacular structure of 1. The architecture of the molecule is
not unprecedented, as the hatching agent of soybean cyst
nematodes, glycinoeclepin A (2; Figure 1), determined in Japan
[
3]
Solanoeclepin A (1) is a complex terpenoid natural product
produced by young potato roots in spring (Figure 1). The com-
pound shows nanomolar activity as a hatching agent of potato
cyst nematodes (PCN). These PCN are parasitic worms (max.
in 1985, shows clear similarities albeit being less complex.
Interest from the agribusiness and the intriguing structure of
the natural product challenged us to embark on a study to-
wards a total synthesis of 1 starting in 1997. We envisaged
a convergent approach, which is briefly outlined in Scheme 1.
Figure 1. Structures of solanoeclepin A (1) and glycinoeclepin A (2).
[
a] R. A. Kleinnijenhuis, B. J. J. Timmer, Dr. G. Lutteke,
Prof. Dr. J. H. v. Maarseveen, Prof. Dr. H. Hiemstra
Van ‘t Hoff Institute for Molecular Sciences
University of Amsterdam
Scheme 1. Our retrosynthetic approach to solanoeclepin A (1).
Science Park 904, 1098 XH Amsterdam (the Netherlands)
E-mail: h.hiemstra@uva.nl
The last ring to be made was planned to be the seven-mem-
bered one through ring-closing metathesis of 3. The coupling
of structures 4 and 5 through aldol-type methodology would
lead to 3. Our early work has led to the enantioselective syn-
thesis of 4 (R=tert-butyldiphenylsilyl (TBDPS)). These studies
culminated in the synthesis of the left-hand substructure 6
(Figure 2) as the most advanced compound in the enantiopure
+
[
b] J. M. M. Smits, Dr. R. de Gelder
Institute for Molecules and Materials
Radboud University
Heyendaalseweg 135, 6525 AJ Nijmegen (the Netherlands)
[
4]
+
[
] Deceased November 22, 2013
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201504894.
Chem. Eur. J. 2016, 22, 1266 – 1269
1266
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
was best mediated by indium powder in a two-phase system
of dichloromethane and aqueous phosphate buffer of pH 6.75
[
11]
for 48 h at 208C. After weakly acidic workup, the desired a-
addition product 16 was obtained in 72% isolated yield. These
successful conditions emanated from extensive investigations
in order to prevent formation of the g-adduct and premature
acetal hydrolysis. The use of metallic tin instead of indium led
to lower yield and poor a-selectivity. The corresponding (achi-
ral) acetal from ethylene glycol was too sensitive to hydrolysis.
Unfortunately, the potential role of the chiral acetal in 15 to
induce enantioselectivity was unsatisfactory, as the enantio-
meric excess of 16 was only about 20%.
Figure 2. Advanced substructures of 1 prepared in our laboratory.
form as a mixture of four diastereomers. This mixture showed
[5]
remarkably good hatching activity.
Our efforts to access the more intricate right-hand substruc-
ture 5 have been quite laborious for a long time. From the
very start we have concentrated on employing the intramolec-
ular [2+2] photocycloaddition as a key step to prepare the tri-
In the key photocycloaddition acetate 17 (made from 16
1,6
cyclo[5.2.1.0 ]decane core, which is the central part contain-
using Ac O/DMAP in pyridine, 82%) was subjected to irradia-
2
[6]
ing the 4-, 5-, and 6-membered ring. We report herein, our
recent successful efforts in this endeavor that resulted in an ef-
ficient enantioselective synthesis of compound 7, which con-
tains the key features of 5 (Figure 2).
tion with 300 nm UV-light in a 9:1 acetonitrile/acetone mixture
using a Rayonet photoreactor (Scheme 3). To our delight, the
Meanwhile, a number of other research groups have pub-
[7]
lished their work on the total synthesis of 1. In 2011, Tanino
and co-workers achieved the first total synthesis of solanoecle-
[
8]
pin A. Their impressive route features more or less a linear
approach counting 52 steps. As compound 7 is the intermedi-
ate in Tanino’s total synthesis, our present results constitute
a formal total synthesis of solanoeclepin A.
Scheme 3. The key [2+2] photocycloaddition of the ester 17.
Returning to our retrosynthetic approach (Scheme 1), after
starting material was converted within 2 h into a clean but in-
separable mixture of four isomeric products in the ratio of
[6]
many years of experimentation in our laboratories, it was
eventually envisaged that 5 should be accessible through in-
tramolecular [2+2] photocycloaddition of a,b-unsaturated
ester 9 to give the tricyclic core structure 8, which contains all
functionalities required to reach the final target. The ester
group was deemed to be a suitable group to introduce the cy-
clopropane substituent at C7 and the benzyloxymethyl group
at C10, allowing elimination to the methylene. Enone 9 was ex-
pected to emanate from the addition of organometallic re-
agent 10 to aldehyde 11, followed by acidic hydrolysis.
1
about 7:1:1:1 according to H NMR. The major product in this
mixture was the desired 18, because after removal of the
acetyl group and recrystallization of the product pure racemic
alcohol 19 was obtained in 32% yield from 17. The crystals of
19 (m.p. 138–1398C) were suitable for X-ray analysis, which un-
ambiguously proved its structure (see the Supporting Informa-
tion).
The stereochemistry of the major isomer can be understood
on the basis of the most reasonable conformations, which lead
to the possible transition states of the photocycloaddition
(Scheme 4), this is in analogy to earlier discussions by Snapper
Thus, our studies commenced with the preparation of com-
pounds 13 and 15, which would serve the roles of 10 and 11,
respectively. Bromide 13 was easily prepared from the known
[9]
allylic alcohol 12 (Scheme 2). The most suitable aldehyde re-
Scheme 4. Conformational representation of the photocycloaddition.
[
12]
and co-workers. The orientation of the OR group with re-
spect to the carbonyl is important. In the case of alcohol 16,
a hydrogen bond with the enone carbonyl will favor conforma-
tion A so that the undesired diastereomer C is expected to be
formed in excess. With R being an acetyl group, conformation
B will be preferred and the desired D (18) will be the product.
The endo orientation of the benzyloxymethyl substituent in
the major isomer can also be understood from the conforma-
tional representation in Scheme 5.
Scheme 2. Initial steps and indium-mediated coupling.
action partner was the enantiopure acetal 15, which was pre-
[10]
pared from the known iodide 14 by acetalization with (R,R)-
hydrobenzoin and methyl orthoformate followed by iodine/
lithium exchange and reaction with DMF. The crucial Barbier-
type coupling reaction of allylic bromide 13 with aldehyde 15
Chem. Eur. J. 2016, 22, 1266 – 1269
www.chemeurj.org
1267
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
acetyl with a TBS group. Unfortunately, the acetyl group in
25a could not be removed without affecting the BPin group.
Furthermore, irradiation of silyl protected 25b at 300 nm gave
a low yield of cycloadduct 26b. The solution to this dilemma
was a change in the wavelength of the UV light. When the cy-
cloaddition of 25b was carried out at 365 nm in acetonitrile,
the desired product 26b was obtained in an excellent yield
[
17]
albeit in a rather slow reaction. In order to adapt this reac-
tion on larger scale, it was carried out in a flow system (see
the Supporting Information). In this fashion 25b was convert-
ed cleanly into 26b on 20 g scale in 87% yield. Thus, the tricy-
Scheme 5. Enantioselective synthesis of the boronate photosubstrate.
1,6
clo[5.2.1.0 ]decane core of solanoeclepin A (26b) was now
With a reliable synthesis of racemic 19 in hand, we studied
ways to obtain the required pure enantiomer. Several possibili-
ties were screened to prepare enantiopure 16, either by resolu-
tion or asymmetric synthesis, but none appeared applicable
available in six steps from commercially starting materials,
[
8]
whereas the literature synthesis needed 15 steps.
For the replacement of the BPin moiety by a carbon sub-
stituent, we chose for the introduction of a vinyl group accord-
[13]
[18]
from a practical sense. The solution was finally sought in re-
placing the methyl ester function in 16 by a pinacolyl boro-
nate, that is 23, which should be available in high enantiopuri-
ty through the elegant methodology developed by Morken
ing to the Aggarwal protocol. This methodology required
protection of the ketone, which was converted into acetal 27
(Scheme 7). A treatment of boronate 27 with excess vinylmag-
[
14]
and co-workers. This meant a restart of the entire synthetic
route, but the gain in efficiency could be substantial.
[15]
Thus, treatment of known allene 20 with bis(pinacolato)di-
boron in the presence of [Pd (dba) ] and chiral ligand 21 ac-
2
3
[14]
cording to the Morken protocol (Scheme 5) produced dibor-
onate 22. This product was not isolated, but immediately
stirred overnight with aldehyde 23, which after weakly acidic
workup led to the desired vinyl boronate 24 in excellent yield
(
84% from 20) and enantiomeric excess (91%) on a 20 g scale.
Evidently, this excellent result provided ample motivation to
continue this new approach.
With the stereochemistry issue solved, the next question
was whether 24 would facilitate the [2+2] photocycloaddition
as well as the ester did. Favorable literature precedent was
[
16]
available, although the reaction had been rarely utilized. The
hydroxyl group was first protected as acetate (25a, Scheme 6),
Scheme 7. Completion of the formal total synthesis.
Scheme 6. The crucial boronate photocycloaddition reaction.
nesium bromide followed by the subsequent addition of
iodine and sodium methoxide led to the desired alkene 28 in
[
18]
which served well in the formation of ester-substituted 18. The
irradiation was carried out as before at 300 nm in acetonitrile/
acetone 9:1 or in pure acetone for 90 min to give a satisfactory
isolated yield (68%) of the desired product 26a in addition to
less than 20% uncharacterized stereoisomers. The similarity of
NMR spectra of 26a and 18 was convincing evidence that the
cycloaddition had proceeded as desired. Much to our delight,
the yield was even better than in the ester case.
excellent overall yield. An oxidative cleavage of the alkene
to the aldehyde 29, and chain extension using the Horner–
Wadsworth–Emmons reaction with triethyl phosphonoacetate
gave the pure (E)-unsaturated ester 30 in 67% over four steps
from 27. Reduction with DIBAL-H provided the (E)-allylic alco-
hol 31, an ideal substrate for diastereoselective cyclopropana-
[
19]
tion, using the Charette method. Thus, the treatment of 31
with diethylzinc and diiodomethane in the presence of the
enantiopure tartrate-derived butylboronate 32 gave the de-
sired diastereomer 33. The diastereoselectivity of the cyclopro-
panation appeared to be close to 100%, so that at this stage
The next task was the replacement of the BPin substituent
by the cyclopropane moiety. As the acetate function appeared
not compatible with this chemistry, we tried to replace the
Chem. Eur. J. 2016, 22, 1266 – 1269
www.chemeurj.org
1268
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
the minor enantiomer, which led to a different diastereomer,
could be removed.
[4] J. C. J. Benningshof, R. H. Blaauw, A. E. van Ginkel, F. P. J. T. Rutjes, J.
Fraanje, K. Goubitz, H. Schenk, H. Hiemstra, Chem. Commun. 2000,
1465.
To finish the formal synthesis of solanoeclepin A from this
stage, alcohol 33 was protected as pivalate 34 (93% over
three steps from 30) and the benzyl group removed by Raney
[
5] J. C. J. Benningshof, M. IJsselstijn, S. R. Wallner, A. L. Koster, R. H. Blaauw,
A. E. van Ginkel, J.-F. Bri›re, J. H. van Maarseveen, F. P. J. T. Rutjes, H.
Schenk, H. Hiemstra, J. Chem. Soc. Perkin Trans. 1 2002, 1701.
[
20]
[6] a) R. H. Blaauw, J.-F. Bri›re, R. de Jong, J. C. J. Benningshof, A. E. van Gin-
kel, F. P. J. T. Rutjes, J. Fraanje, K. Goubitz, H. Schenk, H. Hiemstra, Chem.
Commun. 2000, 1463; b) R. H. Blaauw, J.-F. Bri›re, R. de Jong, J. C. J. Ben-
ningshof, A. E. van Ginkel, J. Fraanje, K. Goubitz, H. Schenk, F. P. J. T.
Rutjes, H. Hiemstra, J. Org. Chem. 2001, 66, 233; c) J.-F. Bri›re, R. H.
Blaauw, J. C. J. Benningshof, A. E. van Ginkel, J. H. van Maarseveen, H.
Hiemstra, Eur. J. Org. Chem. 2001, 2371; d) R. H. Blaauw, J. C. J. Benning-
shof, A. E. van Ginkel, J. H. van Maarseveen, H. Hiemstra, J. Chem. Soc.
Perkin Trans. 1 2001, 2250; e) B. T. B. Hue, J. Dijkink, S. Kuiper, S. van
Schaik, J. H. van Maarseveen, H. Hiemstra, Eur. J. Org. Chem. 2006, 127;
f) G. Lutteke, R. A. Kleinnijenhuis, I. Jacobs, P. J. Wrigstedt, A. C. A. Cor-
reia, R. Nieuwenhuizen, B. T. B. Hue, K. Goubitz, R. Peschar, J. H. van
Maarseveen, H. Hiemstra, Eur. J. Org. Chem. 2011, 3146.
nickel in ethanol to give 35 (96%). The alcohol 35 was sub-
[21]
jected to the Grieco elimination by initial treatment with the
o-nitrophenylselenocyanate and tributylphosphine. Interesting-
ly, the introduction of the selenium substituent required tem-
peratures close to 1008C, whereas Tanino carried out this reac-
[8]
tion in a similar system at room temperature. Likewise, the
elimination after oxidation of the selenide needed 1008C,
much harsher conditions than those reported by Tanino. This is
a reflection of the difference between the intermediates ob-
served by us and Tanino, that is, the orientation of the alkoxy-
methyl substituent on the C10 bridge, which is endo in our
[7] a) S. Tojo, M. Isobe, Synthesis 2005, 1237; b) M. Isobe, S. Niyomchon, C.-
Y. Cheng, A. Hasakunpaisarn, Tetrahedron Lett. 2011, 52, 1847; c) K.-W.
Tsao, C.-Y. Cheng, M. Isobe, Org. Lett. 2012, 14, 5274; d) H.-Y. Chuang, M.
Isobe, Org. Lett. 2014, 16, 4166; e) Y.-T. Lin, F.-Y. Lin, M. Isobe, Org. Lett.
[8]
molecules and exo in the Tanino synthesis. Despite the ele-
vated temperatures, the elimination proceeded cleanly and
produced the alkene 36 in 95% yield from 35. Removal of the
pivaloyl group with methyllithium gave 37, followed by benzy-
lation of the primary alcohol to 38. Finally, removal of the
acetal and TBS group furnished compound 7 (57% from 36),
which appeared identical to the intermediate in the Tanino
total synthesis.
2014, 16, 5948; f) T. Komada, M. Adache, T. Nishikawa, Chem. Lett. 2012,
41, 287; g) M. Adachi, M. Torii, T. Nishikawa, Synlett 2015, 26, 965.
[8] K. Tanino, M. Takahashi, Y. Tomata, H. Tokura, T. Uehara, T. Narabu, M.
Miyashita, Nat. Chem. 2011, 3, 484.
[9] A. Kamal, T. Krishnaji, P. V. Reddy, Tetrahedron Lett. 2007, 48, 7232.
[
10] R. Benhida, P. Blanchard, J. L. Fourrey, Tetrahedron Lett. 1998, 39, 6849.
[11] K.-T. Tan, S.-S. Chng, H.-S. Cheng, T.-P. Loh, J. Am. Chem. Soc. 2003, 125,
2958.
[12] S. M. Ng, S. J. Bader, M. J. Snapper, J. Am. Chem. Soc. 2006, 128, 7315.
13] These unsuccessful attempts will be detailed in a future full paper.
14] H. E. Burks, S. Liu, J. P. Morken, J. Am. Chem. Soc. 2007, 129, 8766.
[15] J. Kuang, S. Ma, J. Org. Chem. 2009, 74, 1763.
[16] a) W. G. Hollis, Jr., W. C. Lappenbusch, K. A. Everberg, C. M. Woleben, Tet-
rahedron Lett. 1993, 34, 7517; b) S. C. Coote, T. Bach, J. Am. Chem. Soc.
In conclusion, we have synthesized 7 in 18 steps, which is
[22]
eight steps fewer than in the Tanino synthesis. More impor-
tantly, we have prepared the almost enantiopure tricyclic struc-
ture 25b in only six steps, which should allow ample opportu-
nities for a more efficient total synthesis and provide quick
access to structure–activity relationship studies. Further results
on similar structures will be reported in due course.
[
[
2013, 135, 14948.
[
[
17] a) For previous successful [2+2]-photocycloadditions at 365 nm, see R.
Brimioulle, H. Guo, T. Bach, Chem. Eur. J. 2012, 18, 7552; b) For a recent
review on [2+2] photocycloadditions, see J. P. Hehn, C. Müller, T. Bach,
in Handbook of Synthetic Photochemistry (Eds: A. Albini and M. Fagnoni),
Wiley-VCH, Weinheim (Germany), 2010, pp. 171–215.
18] a) R. P. Sonawane, V. Jheengut, C. Rabalakos, R. Larouche-Gauthier, H. K.
Scott, V. K. Aggarwal, Angew. Chem. Int. Ed. 2011, 50, 3760; Angew.
Chem. 2011, 123, 3844; b) G. Zweifel, H. Arzoumanian, C. C. Whitney, J.
Am. Chem. Soc. 1967, 89, 3652.
Acknowledgements
We thank Han Peeters and Ed Zuidinga for mass spectrometry
analyses.
[
[
19] A. B. Charette, H. Juteau, H. Lebel, C. Molinaro, J. Am. Chem. Soc. 1998,
1
20, 11943.
20] K. Horita, T. Yoshioka, T. Tanaka, Y. Oikawa, O. Yonemitsu, Tetrahedron
986, 42, 3021. Hydrogenolysis using Pd catalysis gave competing
Keywords: asymmetric synthesis · cycloaddition · indium ·
terpenoids · total synthesis
1
acetal removal; see C. Murali, M. S. Shashidhar, C. S. Gopinath, Tetrahe-
dron 2007, 63, 4149.
21] a) P. A. Grieco, S. Gilman, M. Nishizawa, J. Org. Chem. 1976, 41, 1485;
b) K. B. Sharpless, M. W. Young, J. Org. Chem. 1975, 40, 947.
22] Our 18 step synthesis of 7 is unduly long due to protective group
adaptations in order to reach the Tanino intermediate.
[
[
1] D. H. Marrian, P. B. Russell, A. R. Todd, Biochem. J. 1949, 45, 513 and the
preceding five papers in the same issue.
[
[
2] a) J. G. Mulder, P. Diepenhorst, P. Plieger, I. E. M. Bruggemann-Rotgans,
PCT Int. Appl. WO 93/02,083; [Chem. Abstr. 1993, 118, 185844z]; b) H.
Schenk, R. A. J. Driessen, R. de Gelder, K. Goubitz, H. Nieboer, I. E. M.
Brüggemann-Rotgans, P. Diepenhorst, Croat. Chem. Acta 1999, 71, 593.
3] A. Fukuzawa, A. Furusaki, M. Ikura, T. Masamune, J. Chem. Soc. Chem.
Commun. 1985, 222.
[
Received: December 4, 2015
Published online on December 21, 2015
Chem. Eur. J. 2016, 22, 1266 – 1269
www.chemeurj.org
1269
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim