Synthesis of (?)-Hebelophyllene E: An Entry to Geminal Dimethyl-Cyclobutanes by [2+2] Cycloaddition of Alkenes and Allenoates

Article abstract of DOI:10.1002/anie.201801110

The first synthesis of hebelophyllene E is presented, along with assignment of its previously unknown relative configuration through synthesis of epi-ent-hebelophyllene E. Development of a catalytic enantioselective [2+2] cycloaddition of alkenes and alle

Full text of DOI:10.1002/anie.201801110

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Synthesis of (-)-Hebelophyllene E: An Entry to geminal
Dimethylcyclobutanes by [2+2] Cycloaddition of Alkenes and
Allenoates
Authors: Johannes Wiest, Michael Conner, and Michael Kevin Brown
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201801110
Angew. Chem. 10.1002/ange.201801110
Link to VoR: http://dx.doi.org/10.1002/anie.201801110
http://dx.doi.org/10.1002/ange.201801110

10.1002/anie.201801110
Angewandte Chemie International Edition
COMMUNICATION
Synthesis of (−)-Hebelophyllene E: An Entry to geminal Dimethyl-
cyclobutanes by [2+2] Cycloaddition of Alkenes and Allenoates
Johannes M. Wiest, Michael L. Conner, and M. Kevin Brown*
Abstract: The first synthesis of hebelophyllene E is presented,
along with assignment of its previously unknown relative
configuration through synthesis of epi-ent-hebelophyllene E.
Development of a catalytic enantioselective [2+2] cycloaddition of
alkenes and allenoates provides access to the required chiral
geminal dimethylcyclobutanes. Key to its success is the identification
However, trisubstituted alkenes were not compatible with these
conditions as low levels of selectivity were observed. Herein, we
disclose our results on overcoming these limitations by catalyst
optimization[11] with
a particular emphasis on prenylated
substrates (Scheme 1C). Since more than 600 of the ~2300
known cyclobutane containing-natural products comprise
geminal dimethyl substitution, development of such a method
does not only serve this particular synthesis, but is of general
utility.[12]
of
cycloaddition in high enantioselectivities with good functional group
tolerance (9 examples, up to 97:3 er). Thus, late-stage
a
novel oxazaborolidine catalyst, which promotes the
a
cycloaddition using a fully functionalized alkene followed by a
diastereoselective reduction allows for a concise entry to this class
of natural products.
Hebelophyllene E (1) is part of a small family of cis-fused
caryophyllene-type sesquiterpenes, which were isolated in the
late 1990s from the ectomycorrhizal fungus Hebeloma
longicaudum.[1] Although this class of natural products has
shown potential for nursery inoculation of conifer seedlings for
better growth and survival in the fields, no synthetic access has
been reported to date.[2] Additionally, the authors of the isolation
report were unable to determine the relative configuration of the
carbon C-4 stereocenter,[1c] which provided us with additional
incentive for pursuing this total synthesis. Since previous
methods to form geminal dimethylcyclobutanes in natural
[4]
product synthesis[3]
do not provide access to the key
β-cyclobutylacetate motif, development of such a method was
required to ensure a concise synthesis. We intended to form the
unusual cis-fused cyclobutane core by enantioselective alkene-
[6] [7] [8]
allenoate [2+2] cycloaddition[5]
followed by diastereo-
selective reduction (Scheme 1A).
In the 1980s, Snider and coworker reported an Al-promoted
[2+2] cycloaddition between 2-methyl-2-butene and methyl
allenoate to form dimethylcyclobutanes (Scheme 1B).[9] The high
level of regioselectivity presumably arises from an asynchronous
cycloaddition in which the sigma bond between the most
nucleophilic and the most electrophilic positions is formed first.
To the best of our knowledge, this is the only example of
successfully using a trisubstituted alkene for this type of
transformation, and was reported to form the required
dimethylcyclobutane motif. Due to lack of an enantioselective
version, this method was not a viable option for our synthesis. In
2015, our group identified a chiral oxazaborolidinium catalyst
that is effective in promoting similar [2+2] cycloadditions and
also allows for excellent enantioinduction (Scheme 1B).[10]
Scheme 1. Retrosynthetic Analysis and Work on the Key [2+2] Cycloaddition.
Tf = trifluoromethanesulfonyl.
We initiated our studies by examining the cycloaddition of
benzyl allenoate (2) and triisopropylsilyl (TIPS) protected
homoprenyl alcohol (3a) as a model alkene. In agreement with
our previous observations, oxazaborolidine catalyst 4a furnished
cyclobutane 5a in modest enantioselectivity and poor E:Z
selectivity (Table 1, entry 1). Whereas alteration of the boroaryl
substitution (entry 2-6) had a noticeable influence on the E/Z
selectivity (e.g. entry 5), enantiomeric selectivities remained
mostly unchanged. Examination of the backbone R-groups
resulted in a decrease of E/Z selectivity and enantioselectivity
(compare entry 6&7). Based on this observation the boroaryl
[a]
Johannes M. Wiest, Michael L. Conner and M. Kevin Brown
Department of Chemistry
Indiana University
800 E. Kirkwood Ave. Bloomington, IN 47405
E-mail: brownmkb@indiana.edu
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/anie.201801110
Angewandte Chemie International Edition
COMMUNICATION
substitution of catalyst 4a and the backbone R-group of catalyst
4g were merged to inverse the E/Z selectivity and favor the Z
isomer. To our delight, catalyst 4h also led to a significant
increase in enantioselectivity (entry 8). Lowering the
temperature to 4 °C while extending the reaction time allowed
for 35:65 - E:Z selectivity and excellent enantioselectivity for the
major Z-isomer (entry 9). It should be noted that pure Z-
cyclobutane Z-5a could be isolated in 53% yield by means of
flash column chromatography. Small amounts of regioisomer
and ene-product were always formed in the cycloaddition (~5%)
and coeluted with the minor E-isomer.
Table 1. [2+2] Cycloaddition: Reaction Optimization.
Scheme 2. Substrate Scope. aIsolated yield of pure Z-isomer (bold). bYield of
combined cycloaddition products determined by crude 1H NMR utilizing a
calibrated internal standard. cds = diastereomer.
When catalytically generated LnCuH was used as the
reductant (entry 3-4), higher yields and less pronounced trans-
selectivity were observed.[14] To overcome the selectivity issue,
we turned to chiral ligands on copper to override the inherent
substrate control. Indeed, racemic 2,2'-bis(diphenylphosphino)-
1,1'-binaphthyl (BINAP) showed different selectivity compared to
R-BINAP making us confident in this approach (entry 5-6). A
careful screening of ligands (see Supporting Information for
details) finally identified Josiphos-type ligand 7 as a good source
of induction switching the selectivity in favor of isomer cis-6
(entry 7).[15] To confirm the ligand control, ligand ent-7 was used
giving the expected trans-product as the major component in
good yield (entry 8). A decrease in the temperature to −78 °C,
resulted in an increase in cis-selectivity as cyclobutane cis-6
was obtained in 85% yield and 80:20 diastereomeric ratio (dr)
(entry 9).
Having identified a catalyst to perform the key reaction in our
synthesis with high enantioselectivity, we investigated a selected
substrate scope (Scheme 2). Functional groups such as silyl
ether, ester, ketone, and benzyl ether were well tolerated
(products Z-5b to Z-5e). High chemoselectivity for the
trisubstituted alkene in the presence of a terminal alkene was
observed (product Z-5f), which turned out to be essential for the
synthesis of hebelophyllene E (1). Sterically smaller groups (e.g.
product Z-5g, R = Me) resulted in lower E/Z selectivity and
therefore lower isolated yield as the pure Z-isomer, whereas
bulky groups led to increased E/Z selectivity of up to 76:24
(product Z-5e, R = C(CH3)2OBn).
Confident that the functional group tolerance of the reaction
(Scheme 2) would allow us to pursue a more concise route to
hebelophyllene
E (1), we initiated the synthesis of fully
functionalized alkenes 3h and 3i from known[16] acyloin rac-3f,
which was easily accessible on decagram scale (Scheme 3).
Wessjohann and coworkers reported[16] a kinetic resolution of
rac-3f using amano lipase PS from Burkholderia capacia, which
worked smoothly on gram-scale to provide both the starting
material and product as single enantiomers. Since the
stereochemistry at carbon C-4 is unknown, access to both
diastereomers of alkene (3h and 3i) was necessary. Hence, we
attempted to control the approach of the vinyl Grignard reagent
based on the polar Felkin-Anh model.[17] We found that benzoyl
(Bz) protection of alcohol 3g was beneficial for two reasons: 1) It
showed very high Felkin-Anh control when treated with
vinylmagnesium bromide in THF, and 2) It allowed for a direct
removal of the protecting group by simply using excess Grignard
reagent. After acetonide protection using 2,2-dimethoxypropane
(2,2-DMP) and tosylic acid (TsOH), alkene 3h could be purified
One unforeseen challenge we faced was achieving cis-
selective reduction after forming the cyclobutane ring – as we
discovered in the following investigation, trans-configured
cyclobutanes are generally the major products. We started
investigating the reduction on a model system using cyclobutane
Z-5a, which was easily synthesized on scale. Our results are
summarized in Table 2: Crabtree’s catalyst resulted in
completely selective formation of cyclobutane trans-6 (entry 1). It
is proposed that allylic strain (Table 2, indicated in the proposed
model) leads to a puckered conformation in which the R-group
preferentially adopts a pseudoaxial conformation. As a result,
favored attack occurs from the convex face (Table 2, bottom).[13]
Magnesium in methanol, which usually furnishes the
thermodynamically more stable product, also resulted in the
trans-product as the major component (entry 2).
This article is protected by copyright. All rights reserved.

10.1002/anie.201801110
Angewandte Chemie International Edition
COMMUNICATION
Table 2. Conjugate Reduction: Optimization and Model.
With the two alkenes in hand, each were subjected to the
[2+2] cycloaddition conditions. Since hebelophyllene E (1)
contains an S-configuration at C-1, catalyst ent-4h was used.
We noticed
a pronounced matched/mismatched scenario
between the catalyst and the nature of the alkene’s proximal
stereocenter: While alkene 3h and catalyst ent-4h resulted in
sluggish reaction and yields around 10%, 3i reacted without
complication in the [2+2] cycloaddition and Z-5i was isolated in
52% yield (Scheme 4) as a single diastereomer and enantiomer.
It is worth mentioning that the cycloaddition required four
equivalents of alkene, slightly lower temperature, and higher
catalyst loading[20] to ensure high conversion. However, excess
alkene can be recovered in 90-95% yield and reused in the
cycloaddition. Fortunately, the succeeding reduction using
matched ligand ent-7 afforded the cyclobutane 8 in high (89:11)
dr and excellent yield.
by flash column chromatography and was isolated as a single
diastereomer in 59% yield over three steps. The relative
configuration was proven by nuclear Overhauser effect (nOe)
correlation while the absolute configuration was confirmed by
comparing the intermediate diol to known literature data (see
Supporting Information for further details).[18]
We originally thought to form alkene 3i by protecting 3g with
an established protecting group suited for chelation. To
circumvent these stepwise protecting procedures, we
considered a more direct approach starting from enantiopure
acetate 3f, which is conveniently formed in the resolution. After
optimization, we found that the ester group itself was capable of
chelation under ether-free conditions (note that dichloromethane
was uniquely effective as a solvent) to give the Cram-chelate[19]
product in good selectivity (82:18). As before, utilizing excess
vinylmagnesium bromide allowed for a direct removal of the
acetate and alkene 3i was isolated after acetonide protection as
a single diastereomer in 59% yield over both steps.
Scheme 4. Synthesis of Hebelophyllene E (1a) and epi-ent-Hebelophyllene E
(1b).
After deprotection of cis-8 with trifluoroacetic acid (TFA), the
incorrect relative configuration from the required matched
cycloaddition, could be corrected by an oxidation-reduction
sequence providing diol
9
in 69% yield and 71:29
diastereoselectivity after three steps. Saponification followed by
lactonization using the Steglich protocol[21] delivered lactone 10
in 94% yield. After trimethylsilyl (TMS) protection of the tertiary
alcohol, lactone TMS-10 was deprotonated using lithium
hexamethyldisilamide (LiHMDS) and the respective enolate
Scheme 3. Synthesis of Both Diastereomers of the Desired Alkene.
This article is protected by copyright. All rights reserved.

10.1002/anie.201801110
Angewandte Chemie International Edition
COMMUNICATION
Lett. 2016, 18, 1614-1617; i) L. M. Chapman, J. C. Beck, L. Wu, S. E.
Reisman, J. Am. Chem. Soc. 2016, 138, 9803-9806; j) C. García-
Morales, B. Ranieri, I. Escofet, L. López-Suarez, C. Obradors, A. I.
Konovalov, A. M. Echavarren, J. Am. Chem. Soc. 2017, 139, 13628-
13631.
quenched with methyl iodide in >20:1 diastereoselectivity. Upon
work-up with HCl the TMS group was released and
hebelphyllene E (1a) isolated in 50% yield over the two steps
(the minor isomer from the Zn(BH4)2 reduction was easily
separated at this stage). All analytical data of the synthetic
sample including optical rotation were in agreement with the
reported data of the natural product.[1c]
[5]
[6]
For reviews on [2+2] cycloadditions see: a) Y. Xu, M. L. Conner, M. K.
Brown, Angew. Chem. Int. Ed. 2015, 54, 11918-11928; Angew. Chem.
2015, 127, 12086-12097; b) S. Poplata, A. Tröster, Y.-Q. Zou, T. Bach,
Chem. Rev. 2016, 116, 9748-9815.
To confirm the relative configuration of carbon C-4, we also
synthesized epi-ent-hebelophyllene E (1b) starting from alkene
3h and the respective matched catalyst 4h. The synthesis
proceeded in 12% yield over nine steps using a closely related
sequence as the one for hebelophyllene E (1a) (see Supporting
Information for details). To our delight, the spectra of epi-ent-
hebelophyllene E (1b) were in stark disagreement with those
from the natural product. We also confirmed the relative
configuration of all other stereocenters by 2D NMR analysis (see
Supporting Information for details), which allows us to
confidently propose depicted structure 1a as the structure of
hebelophyllene E.
For a selection of recent enantioselective [2+2] cycloadditions utilizing
photochemical excitation see: a) F. Mayr, R. Brimioulle, T. Bach, J. Org.
Chem. 2016, 81, 6965-6971; b) T. R. Blum, Z. D. Miller, D. M. Bates, I.
A. Guzei, T. P. Yoon, Science 2016, 354, 1391; c) A. Tröster, R. Alonso,
A. Bauer, T. Bach, J. Am. Chem. Soc. 2016, 138, 7808-7811; d) Z. D.
Miller, B. J. Lee, T. P. Yoon, Angew. Chem. Int. Ed. 2017, 56, 11891-
11895; Angew. Chem. 2017, 129, 12053-12057; e) X. Huang, T. R.
Quinn, K. Harms, R. D. Webster, L. Zhang, O. Wiest, E. Meggers, J.
Am. Chem. Soc. 2017, 139, 9120-9123; f) K. L. Skubi, J. B. Kidd, H.
Jung, I. A. Guzei, M.-H. Baik, T. P. Yoon, J. Am. Chem. Soc. 2017, 139,
17186-17192.
[7]
For a selection of recent enantioselective [2+2] cycloadditions utilizing
Lewis acid catalysis see: a) T. Kang, S. Ge, L. Lin, Y. Lu, X. Liu, X.
Feng, Angew. Chem. Int. Ed. 2016, 55, 5541-5544; b) J.-L. Hu, L.-W.
Feng, L. Wang, Z. Xie, Y. Tang, X. Li, J. Am. Chem. Soc. 2016, 138,
13151-13154.
In summary, the first synthesis of the sesquiterpene
hebelophyllene
E
was accomplished and the relative
configuration of the side chain was assigned. Moreover, access
to the cis-fused geminal dimethylcyclobutane core was achieved
through development of a [2+2] cycloaddition involving a fully
functionalized alkene and founded on the identification of a
novel oxazaborolidine catalyst. Given the number of isolated
geminal dimethylcyclobutane natural products, this method itself
should be useful for future synthetic efforts in the field of
cyclobutane natural product synthesis.
[8]
[9]
For a selection of recent enantioselective [2+2] cycloadditions utilizing
transition metal catalysis see: a) R. Kumar, E. Tamai, A. Ohnishi, A.
Nishimura, Y. Hoshimoto, M. Ohashi, S. Ogoshi, Synthesis 2016, 48,
2789-2794; b) D. Kossler, N. Cramer, Chem. Sci. 2017, 8, 1862-1866.
a) B. B. Snider, D. K. Spindell, J. Org. Chem.1980, 45, 5017-5020; b) B.
B. Snider, E. Ron, J. Org. Chem. 1986, 51, 3643-3652. For a related
example see: H. M. R. Hoffmann, Z. M. Ismail, A. Weber, Tetrahedron
Lett. 1981, 22, 1953-1956.
[10] M. L. Conner, Y. Xu, M. K. Brown, J. Am. Chem. Soc. 2015, 137, 3482-
3485.
[11] We have recently observed E/Z selectivity dependencies on the nature
of the oxazaborolidine catalyst: Y. Xu, Y. J. Hong, D. J. Tantillo, M. K.
Brown, Org. Lett. 2017, 19, 3703-3706.
Acknowledgements
We thank Indiana University and the National Institutes of Health
(R01GM110131) for financial support. The Deutsche
Forschungsgemeinschaft is acknowledged for postdoctoral
fellowship support to J.M.W.
[12] Database search performed at https://new.reaxys.com/#/search/quick,
January 13, 2018.
[13] The preferred pseudoaxial R-group arrangement is a result of the Z-
configuration of the alkene. When E-5a was subjected to the reduction
conditions using Crabtree's catalyst, 75% yield and 20:80 trans:cis
selectivity was observed.
Keywords: Natural Product Synthesis • [2+2] Cycloaddition •
[14] For reviews see: a) C. Deutsch, N. Krause, B. H. Lipshutz, Chem. Rev.
2008, 108, 2916-2927; b) B. H. Lipshutz, Synlett 2009, 509-524.
[15] a) B. H. Lipshutz, J. M. Servesko, Angew. Chem. Int. Ed. 2003, 42,
4789-4792; Angew. Chem. 2003, 115, 4937-4940; b) B. H. Lipshutz, J.
M. Servesko, B. R. Taft, J. Am. Chem. Soc. 2004, 126, 8352-8353; c) B.
H. Lipshutz, N. Tanaka, B. R. Taft, C.-T. Lee, Org. Lett. 2006, 8, 1963-
1966.
Enantioselective Catalysis
[1]
a) M. Wichlacz, W. A. Ayer, L. S. Trifonov, P. Chakravarty, D. Khasa,
Phytochemistry 1999, 51, 873-877; b) M. Wichlacz, W. A. Ayer, L. S.
Trifonov, P. Chakravarty, D. Khasa, Phytochemistry 1999, 52, 1421-
1425; c) M. Wichlacz, W. A. Ayer, L. S. Trifonov, P. Chakravarty, D.
Khasa, J. Nat. Prod. 1999, 62, 484-486.
[16] G. Scheid, W. Kuit, E. Ruijter, R. V. A. Orru, E. Henke, U. Bornscheuer,
L. A. Wessjohann, Eur. J. Org. Chem. 2004, 1063-1074.
[17] a) N. T. Anh, O. Eisenstein, Tetrahedron Lett. 1976, 17, 155-158; b) A.
Nguyen Trong, O. Eisenstein, J. M. Lefour, M. E. Tran Huu Dau, J. Am.
Chem. Soc. 1973, 95, 6146-6147.
[2]
a) C. D. Boyle, K. E. Hellenbrand, Can. J. Botany 1991, 69, 1764-1771;
b) C. D. Boyle, W. J. Robertson, P. O. Salonius, Can. J. For. Res. 1987,
17, 1480-1486.
[3]
[4]
For a review see: M. Wang, P. Lu, Org. Chem. Front. 2018, 5, 254-259.
For recent examples on geminal dimethylcyclobutane syntheses in
context of natural products see: a) Y. Kang, Y. Mei, Y. Du, Z. Jin, Org.
Lett. 2003, 5, 4481-4484; b) A. V. Kurdyumov, R. P. Hsung, K. Ihlen, J.
Wang, Org. Lett. 2003, 5, 3935-3938; c) M. P. Paduraru, P. D. Wilson,
Org. Lett. 2003, 5, 4911-4913; d) S. Dong, G. D. Parker, T. Tei, L. A.
Paquette, Org. Lett. 2006, 8, 2429-2431; e) M. Mondal, V. G. Puranik,
N. P. Argade, J. Org. Chem. 2007, 72, 2068-2076; f) M. Fleck, T. Bach,
Angew. Chem. Int. Ed. 2008, 47, 6189-6191; Angew. Chem. 2008, 120,
6284-6286; g) T. Hirokawa, S. Kuwahara, Eur. J. Org. Chem. 2013,
2780-2782; h) B. Ranieri, C. Obradors, M. Mato, A. M. Echavarren, Org.
[18] D. Ishiyama, Y. Fukushi, M. Ohnishi-Kameyama, T. Nagata, H. Mori, T.
Inakuma, Y. Ishiguro, J. Li, H. Kawagishi, Phytochemistry 1999, 50,
1053-1056.
[19] D. J. Cram, D. R. Wilson, J. Am. Chem. Soc. 1963, 85, 1245-1249.
[20] Higher catalyst loadings are tolerable, since the catalyst can be
synthesized in one step from commercially available starting materials.
[21] B. Neises, W. Steglich, Angew. Chem. Int. Ed. 1978, 17, 522-524;
Angew. Chem. 1978, 90, 556-557.
This article is protected by copyright. All rights reserved.

10.1002/anie.201801110
Angewandte Chemie International Edition
COMMUNICATION
Layout 2:
COMMUNICATION
Johannes M. Wiest, Michael L. Conner,
and M. Kevin Brown*
Page No. – Page No.
Synthesis of (−)-Hebelophyllene E:
An Entry to geminal Dimethyl-
cyclobutanes by [2+2] Cycloaddition
of Alkenes and Allenoates
First design, then assign: The first synthesis of hebelophyllene E is presented,
along with assignment of its previously unknown relative configuration through
synthesis of epi-ent-hebelophyllene E. Development of a catalytic enantioselective
[2+2] cycloaddition of alkenes and allenoates followed by diastereoselective
conjugate reduction provides access to the required chiral geminal
dimethylcyclobutane core.
This article is protected by copyright. All rights reserved.