Asymmetric Organocatalytic Stepwise [2+2] Entry to Tetra-Substituted Heterodimeric and Homochiral Cyclobutanes

Article abstract of DOI:10.1002/chem.201601842

An asymmetric synthesis of tetra-substituted cyclobutanes involving an organocatalytic, stepwise [2+2]-cycloaddition is described. The secondary-amine-catalyzed method allows for the hetero-dimerization of two different cinnamic-acid-derived sub-units, opening a novel one-step assembly to densely functionalized, head-to-tail coupled dimeric cyclobutanes in high enantiomeric excess. A series of selective synthetic interconversions in these sensitive cycloadducts is also described.

Full text of DOI:10.1002/chem.201601842

DOI: 10.1002/chem.201601842
Communication
&
Asymmetric Cycloadditions
Asymmetric Organocatalytic Stepwise [2+2] Entry to Tetra-
Substituted Heterodimeric and Homochiral Cyclobutanes
[
a]
Alex J. Nielsen, Hilary A. Jenkins, and James McNulty*
[
7b]
iridoid argenteoside A, whereas head-to-tail derivatives are
Abstract: An asymmetric synthesis of tetra-substituted cy-
clobutanes involving an organocatalytic, stepwise [2+2]-
cycloaddition is described. The secondary-amine-catalyzed
method allows for the hetero-dimerization of two different
cinnamic-acid-derived sub-units, opening a novel one-step
assembly to densely functionalized, head-to-tail coupled
dimeric cyclobutanes in high enantiomeric excess. A series
of selective synthetic interconversions in these sensitive
cycloadducts is also described.
[
7c]
[7d]
exemplified by dipiperamide A, nigramide P, dipiperamide
[
7e]
[7b]
[7]
E, and argenteoside B, among many others. The argen-
teosides have recently been identified as potent inhibitors of
heat shock protein 90 (Hsp90), a high-value therapeutic target,
whereas dipiperamides A and B have potent activity against
cytochrome P450 (CYP) 3A4.
The synthesis of cyclobutane derivatives has risen in promi-
nence over the last few years in view of an increased recogni-
tion of their importance within bioactive natural products,
[1]
with well over 200 derivatives now known. In addition, cyclo-
butanes are often used as synthetically useful, strained inter-
mediates. Classical alkylation and photochemical routes to
[
2]
achiral cyclobutanes including solid-state photochemistry
[3]
have been supplemented with novel methods involving step-
wise, thermal [2+2]-cycloaddition reactions mediated by transi-
[
4]
[5]
[6]
tion metals, organocatalysis, and one-electron oxidants. In
addition to expanding the scope of alkene participants, these
methods often allow for regio-controlled cycloaddition leading
to non-symmetrical dimers and/or asymmetric entries to chiral
cyclobutane derivatives.
Figure 1. Structures of selected biologically active natural product cyclobu-
tane-containing lignan-dimers including head-to-head (left) and head-to-tail
(right) coupled derivatives.
Although many dimeric and pseudo-dimeric (non-symmetri-
cal) naturally occurring cyclobutanes appear to be derived
from the dimerization of cinnamyl, coumaryl, or extended
Despite the number of approaches developed to access
[
7a]
[3]
diene amides, the precise enzyme responsible (“[2+2]-ase”),
chiral cyclobutanes, no process has been reported that per-
mits direct asymmetric heterodimerization of two different cin-
namyl-derived sub-units. The closest examples reported to
date involve remote dienamine-mediated couplings of 3-vinyl-
and hence mechanism of the biological dimerization is un-
known. Nonetheless, for non-meso examples, such cyclobu-
tanes may be homo-chiral, indicating the involvement of an
enzymatic biosynthesis rather than a strictly abiotic photo-
chemical or oxidative dimerization process. In addition, these
cinnamic acid derivatives are known to dimerize through
either a head-to-head or a head-to-tail alignment leading to re-
gioisomeric cyclobutanes. Examples of head-to-head coupled
[
5a]
[5c]
[5f]
oxindoles, nitroolefins, and vinylpyrroles. The develop-
ment of a direct asymmetric heterodimerization process would
be of great utility given the valuable biological activities re-
ported for these and other related non-symmetrical dimers,
[
8]
such as sceptrin, and lignans that could be accessed through
[7a]
[2a]
cyclobutanes include piperarborenine D (Figure 1) and the
cyclobutane fragmentation.
In this communication, we
report the first examples of organocatalytic heterodimerization
of two different cinnamyl-derived olefins in a regioselective
and highly enantioselective fashion to yield tetra-substituted
cyclobutanes.
[
a] A. J. Nielsen, Dr. H. A. Jenkins, Dr. J. McNulty
Department of Chemistry and Chemical-Biology
McMaster University
1
280 Main Street West, Hamilton, Ontario, L8S 4M1 (Canada)http://
A retrosynthetic analysis detailing the potential iminium-ion-
mediated cascade to access cyclobutanes is shown in
Scheme 1. It was postulated that heterodimerization might be
achieved through reaction of a cinnamaldehyde derivative
www.chemistry.mcmaster.ca/mcnulty/index.html
E-mail: jmcnult@mcmaster.ca
Supporting information for this article can be found under http://
dx.doi.org/10.1002/chem.201601842.
Chem. Eur. J. 2016, 22, 9111 – 9115
9111
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
with a cinnamyl alcohol as indicated. Activation of the olefin of
cinnamyl alcohol 1 with a 4-hydroxyl group, for example,
would direct the stepwise cycloaddition onto the iminium ion
derived from 2 in a regioselective fashion. It should be noted
that, although organocatalytic Friedel–Crafts reactions of phe-
bromocinnamaldehyde, and the resulting cycloadduct 4c con-
verted to its crystalline semicarbazone 6 (Figure 2). Single-crys-
tal X-ray diffraction analysis confirmed the structure, the crys-
tals proved to be homochiral, and the absolute stereochemis-
try 6 was defined as shown (Figure 2).
[
9]
[10]
[5f]
nols, vinylogous electron-rich anilines, and pyrroles onto
Michael acceptors are known, in no case have two phenylpro-
panoid sub-units been converted to cyclobutanes through
such a method.
Figure 2. Structure and absolute stereochemistry of the cyclobutane adduct
from the [2+2] cycloaddition of 4-bromocinnamaldehyde and isoeugenol as
determined via the semicarbazone (CCDC 1456076; see ref. [14]).
Scheme 1. Retrosynthetic analysis of a heterodimeric cyclobutane derivative
potentially available from dimerization of a cinnamaldehyde derivative 2
and cinnamyl alcohol 1 precursors.
While optimizing the conditions for the asymmetric catalysis,
we discovered that the [2+2] cycloaddition is in fact thermally
reversible and subject to an interesting dynamic kinetic resolu-
tion. When purified racemic 4a was stirred with catalyst 3d in
methanol at room temperature for several days, partial refor-
mation of starting materials was noted. Re-isolation and analy-
sis of 4a showed enantioenrichment of the product in favor of
the opposite enantiomer ent-4a (23% ee), produced using cat-
alyst 3d. This result is readily explained by the process of mi-
We began this investigation with attempts to engage isoeu-
genol 1a in reaction with cinnamaldehyde 2a in the presence
of the achiral secondary amine pyrrolidine 3a at room temper-
ature in THF (Table 1, entry 1). We identified that a [2+2]-cyclo-
adduct was being formed slowly, and isolated rac-4a in 14%
yield. A solvent screen (Table 1, entries 1–7) demonstrated
a strong preference for polar protic solvents, with methanol
being identified as the ideal solvent in a process that pro-
vided the cyclobutane rac-4a as a single diastereomer in
Table 1. Catalyst and solvent screening for the organocatalytic [2+2] cycload-
[a]
dition
8
0% isolated yield. We next explored asymmetric induction
using l-proline 3b and several versions of the common dia-
rylsilylprolinol (Jørgensen-type) catalysts (3c–e). Interesting-
ly, no product was detected using l-proline (Table 1,
entry 8) and only trace amounts of the cyclobutane were
detected using the free diphenylprolinol 3c or the second
generation Jørgensen catalyst 3e, even on extended reac-
tion times (Table 1, entries 9 and 10).
Nonetheless, to our delight, the (2S)-diarylsilylprolinol cat-
alyst 3d proved to be highly efficient, giving the desired
product in 77% isolated yield and with 82% ee (Table 1,
entry 11). The enantioselectivity of the reaction was also sig-
nificantly improved by cooling to 88C (refrigerator) for the
duration of the reaction. This process yielded essentially
a single enantiomer without any decrease in the isolated
yield (Table 1, entry 12). The ee of the cyclobutane 4a was
determined using chiral HPLC analysis of the alcohol 5a in
direct comparison to rac-5a prepared using pyrrolidine as
catalyst.
[
b]
[c]
Entry
Catalyst (x mol%)
Solvent
Time
Yield [%]
ee [%]
1
2
3
4
5
6
7
8
9
3a (20)
3a (20)
3a (20)
3a (20)
3a (20)
3a (20)
3a (20)
3b (20)
3c (10)
3e (10)
3d (10)
3d (10)
THF
DMSO
48 h
48 h
48 h
48 h
48 h
48 h
48 h
48 h
4 d
14
26
30
31
46
53
80
n.r.
Trace
Trace
77
–
–
–
–
–
–
–
–
CH
2
Cl
2
Toluene
iPrOH
EtOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
n.d.
n.d.
10
4 d
5 d
5 d
1
1
82
97
[d]
12
78
The structure of the isolated cyclobutane 4a was initially
1
deduced through 1- and 2-D H NMR analysis, through
[
2
3
a] Unless otherwise noted, reactions were performed with 1a (0.66 mmol),
a (1.00 mmol), and catalyst in 1.33 mL of solvent. Reactions using catalysts
b–e were performed on half scale. [b] Isolated yield of the cyclobutane alde-
hyde; n.r.=no reaction. [c] Enantiomeric excess (ee) was determined on the
reduced cyclobutanol employing chiral HPLC analysis; n.d.=not determined.
which the stereochemistry appeared to be all-trans. We
hoped to confirm this result as well as to ascertain the ab-
solute stereochemistry of the reaction mediated by secon-
dary amine 3d through X-ray analysis of a suitable deriva-
tive. The reaction of isoeugenol 1a was repeated using 4-
[d] Reaction performed at 88C without stirring.
Chem. Eur. J. 2016, 22, 9111 – 9115
www.chemeurj.org
9112
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
croscopic reversibility, with catalyst 3d favoring the lower
energy retro-cycloaddition pathway available. The cycloaddi-
tion is therefore forward-driven under kinetic conditions by en-
thalpic considerations (formation of two sigma bonds) but re-
versible thermodynamically at higher temperatures. The cyclo-
butane aldehyde 4a can readily be isolated and handled under
normal conditions, but should be prevented from re-establish-
ing an iminium ion mediated cycloreversion.
Table 2. Scope and selectivity of the organocatalytic [2+2] cycloaddition
reaction (isolated yield, ee).
A) Reaction of 1a with acceptor a,b-unsaturated aldehydes.
The scope of the reaction was next investigated using isoeu-
genol 1a in reaction with a variety of aromatic a,b-unsaturated
aldehydes 2 (Table 2A). The desired alkenals were readily pre-
pared using our recently described two-carbon homologation
[
11]
reagent. To avoid cycloreversion, the aldehydes were imme-
diately reduced after the cycloaddition reaction was complete.
The cyclobutane adducts were obtained in good yields (66–
8
0%) and very high ee values (91–98%) with either electron-
donating or electron withdrawing substituents on the alkenal
, which was also successful with ortho, meta, or para substitu-
ents.
We next explored the reaction of cinnamaldehyde 2a with
2
several different electron rich alkenes to probe the donor re-
quirements of the reaction (Table 2B). Whereas isoeugenol 1a
was used successfully in many examples, its corresponding
methyl ether methylisoeugenol 1b did not produce a cycload-
duct, demonstrating that a free phenol is required for the
donor to enter into the organocatalytic [2+2] cascade. Most
importantly, in view of access to the natural cinnamyl-derived
cyclobutanes, the reaction of cinnamaldehyde proved highly
successful with coniferyl alcohol 1c, giving rise to the cyclobu-
tane 4g in 77% isolated yield as essentially a single enantio-
mer. Hence, the method allows for the catalytic asymmetric
head-to-tail dimerization of a cinnamaldehyde derivative with
a cinnamyl alcohol permitting access to heterodimeric cyclobu-
tane carboxaldehydes (vide infra). The reaction was also suc-
cessful employing (E)-1-(4’-hydroxyphenyl)-1-butene 1d, yield-
ing the corresponding cycloadduct 5h (after reduction).
Overall, the results show that a conjugated free-phenolic sub-
stituent is required to activate the electron-rich donor olefin
for successful engagement in the reaction, and that the reac-
tion works well with a wide variety of cinnamaldehyde deriva-
tives 2.
B) Reaction of cinnamaldehyde with other donor alkenes.
Unless otherwise noted, reactions were performed with 1 (0.33 mmol), 2
(0.50 mmol), and catalyst 3d (0.1 equiv) in 0.66 mL of MeOH, over 5 days
at 88C. Yields of the cyclobutane aldehyde are reported; ee was deter-
mined by HPLC analysis of the reduced cyclobutanes; 4g was benzoylat-
ed prior to reduction to avoid production of a meso diol; n.r=no reac-
tion.
A selection of transformations were developed to investigate
applications while retaining chirality on the heterodimeric cy-
clobutane 4g (Scheme 2). Direct reduction of 4g would lead
to a meso-triol (not shown), however 4g could be converted
to the bis-benzoate 7, allowing reduction of the aldehyde to
yield the cyclobutamethanol 8. Chiral HPLC analysis of 8 dem-
onstrated that homochirality was maintained through this
tightrope of reactions (Scheme 2, ii and iii).
ylide derived from (ethoxycarbonylmethyl)triisobutylphospho-
nium bromide, gave the two-carbon extended ester 12, analo-
gous to natural products such as nigramide P (Figure 1).
It is important to note that the two-step sequence of i fol-
lowed by vii (Scheme 2) opens a controlled access to vinyl-cy-
[
7d]
In addition, rac-4a was readily converted to the diacetate 9,
reduction of which led quickly to the diacetoxy alcohol 10. We
also discovered that reaction of 9 or 10 with excess sodium
borohydride led to the reductive cleavage of the ortho-me-
thoxy acetate, most likely through a chelation-assisted path-
way. This reaction yields the free phenol derivative 11, an inter-
mediate that appears suitable for one-electron oxidative frag-
clobutanes, avoiding possible [4+2]-type adducts that often
co-occur with the natural cyclobutanes.
From a mechanistic perspective, we were able to successful-
ly conduct the stepwise [2+2]-reaction of 1a with 2a using
pyrrolidine 3a catalysis either in the dark or in the presence of
5 mol% 4-tertbutylcatechol indicating that neither photochem-
ical nor oxidative processes are involved. In conjunction with
the very high ee values observed, the evidence indicates that
[2a]
mentation reactions.
Finally, reaction of rac-4g with the
Chem. Eur. J. 2016, 22, 9111 – 9115
www.chemeurj.org
9113
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
allowing an overall anti-periplanar HOMO–LUMO alignment for
the first step of the cascade. The model shows that a strong
possibility exists for stabilizing face or edge-on p–p secondary
orbital interactions in this assembly. The enamine (II) now
closes the cyclobutane ring by adding on to the p-quinome-
thide intermediate, with iminium hydrolysis completing the
cascade.
Figure 4. Transition-state assembly using catalyst 3d leading to the product
of absolute stereochemistry shown in Figure 3
In conclusion, we report the discovery of novel organocata-
lytic methodology for the asymmetric synthesis of highly func-
tionalized chiral cyclobutanes in a regiospecific manner from
a,b-unsaturated aldehydes and alkenylphenols. The process
yields heterodimeric head-to-tail coupled cyclobutanes in
good yield and with excellent enantioselectivity.
Scheme 2. Selective manipulations of 4g. i: 3d, MeOH, 88C, 5d (0.1 equiv),
7%. ii: Benzoyl chloride (2.2 equiv), 4-dimethylaminopyridine (0.1 equiv),
diisopropylethylamine, CH Cl , 58%. iii: NaBH (1.0 equiv), EtOH, 95%. iv: ace-
tyl chloride (2.2 equiv), 4-dimethylaminopyridine (0.1 equiv), diisopropyle-
thylamine, CH Cl , 50%. v: NaBH (1.0 equiv), MeOH, 20 min, 90%. vi: NaBH
5 equiv), MeOH, 8 h (90%). vii: (ethoxycarbonylmethyl)triisobutylphosphoni-
um bromide (3.0 eq), KOtBu (3.0 eq), THF, 08C–rt, 1 h, 75%.
7
2
2
4
2
2
4
4
(
A selection of transformations on the cyclobutane carboxal-
dehyde 4g have been developed to showcase the potential of
this method to access chiral synthetic intermediates and ad-
ducts suitable for fragmentation reactions or conversion to
natural product-containing cyclobutanes and analogs. Finally,
the ease with which these strained tetrasubstituted cyclobu-
tanes are formed under iminium-ion catalysis opens another
the reaction proceeds through a stepwise formal [2+2] using
tandem iminium-enamine catalysis (Figure 3).
The stereochemistry of the reaction is rationalized through
the transition-state assembly depicted in Figure 4 using cata-
lyst 3d. The donor substituent (most likely the phenoxide
anion) attacks the least hindered Si-face of the iminium ion
[
7a]
consideration regarding the nature of the [2+2]-ase, which
could lead to such natural products under non-oxidative and
non-photoinduced conditions. As novel methods for the syn-
[
3]
(
Figure 4, (I)) with the diarylprolinol substituent placed distal
thesis of cyclobutanes are actively sought, the present dis-
covery highlights a widening gap in comparison to classic ap-
[
12]
proaches to “tetramethylene” carboxylates and cyclobutane
[
13]
itself, in terms of reaction yield, regio-control, and now one-
step asymmetric entry to these intermediates. Our group is ac-
tively exploring umpoled variations of the chemistry to access
head-to-head dimers as well as applications in total synthesis.
Acknowledgement
We thank NSERC for financial support of this work. A.J.N. is the
recipient of an NSERC postgraduate fellowship. We thank
Thomas Gero for invaluable suggestions on obtaining a crystal-
line derivative suitable for X-ray crystallographic analysis.
Keywords: asymmetric synthesis
dimers · organocatalysis
· cyclobutanes · lignan
Figure 3. Proposed catalytic cycle and absolute stereochemical configuration
of cyclobutane derivatives.
[1] V. M. Dembitsky, J. Nat. Med. 2007, 62, 1–33.
Chem. Eur. J. 2016, 22, 9111 – 9115
www.chemeurj.org
9114 ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
[
2] a) A. K. F. Albertson, J.-P. Lumb, Angew. Chem. Int. Ed. 2015, 54, 2204–
208; Angew. Chem. 2015, 127, 2232–2236; b) M. A. Ischay, M. E. Anzo-
vino, J. Du, T. P. Yoon, J. Am. Chem. Soc. 2008, 130, 12886–12887;
c) M. D. Cohen, G. M. J. Schmidt, J. Chem. Soc. 1964, 2000–2013.
3] For a recent, comprehensive mini-review, see: Y. Xu, M. L. Coner, M. K.
Brown, Angew. Chem. Int. Ed. 2015, 54, 11918–11928; Angew. Chem.
Bifulco, M. G. Chini, C. Pisano, N. De Tommasi, A. Braca, J. Med. Chem.
2013, 56, 1583–1595; c) M. Takahashi, M. Ichikawa, S. Aoyagi, C. Kibaya-
shi, Tetrahedron Lett. 2005, 46, 57–59; d) K. Wei, W. Li, K. Koike, Y. Chen,
T. Nikaido, J. Org. Chem. 2005, 70, 1164–1176; e) S. Tsukamoto, K.
Tomise, K. Miyakawa, B.-C. Cha, T. Abe, T. Hamada, H. Hirota, T. Ohta,
Bioorg. Med. Chem. 2002, 10, 2981–2985; f) S. Tsukamoto, B.-C. Cha, T.
Ohta, Tetrahedron 2002, 58, 1667–1671; g) R. Muharini, Z. Liu, W. Lin, P.
Proksch, Tetrahedron Lett. 2015, 56, 2521–2525.
2
[
[
2
015, 127, 12086–12097.
4] a) J. M. Hoyt, V. A. Schmidt, A. M. Tondreau, P. J. Chirik, Science 2015,
49, 960–963; b) Y.-J. Chen, T.-J. Hu, C.-G. Feng, G.-Q. Lin, Chem.
3
[8] Z. Ma, X. Wang, X. Wang, R. A. Rodriguez, C. E. Moore, S. Gao, X. Tan, Y.
Ma, A. L. Rheingold, P. S. Baran, C. Chen, Science 2014, 346, 219–224.
[9] S. Bai, X. Liu, Z. Wang, W. Cao, L. Lin, X. Feng, Adv. Synth. Catal. 2012,
354, 2096–2100.
Commun. 2015, 51, 8773–8776; c) Y. Wang, Z. Zheng, L. Zhang, Angew.
Chem. Int. Ed. 2014, 53, 9572–9576; Angew. Chem. 2014, 126, 9726–
9
730.
[
5] a) L.-W. Qi, Y. Yang, Y.-Y. Gui, Y. Zhang, F. Chen, F. Tian, L. Peng, L.-X.
Wang, Org. Lett. 2014, 16, 6436–6439; b) N. Vallavoju, S. Selvakumar, S.
Jockusch, M. P. Sibi, J. Sivaguru, Angew. Chem. Int. Ed. 2014, 53, 5604–
[10] A. R. Philipps, L. Fritze, N. Erdmann, D. Enders, Synthesis 2015, 47,
2377–2384.
[11] a) J. McNulty, C. Zepeda-Velazquez, D. McLeod, Green Chem. 2013, 15,
3146–3149; b) J. McNulty, C. Zepeda-Velazquez, Angew. Chem. Int. Ed.
2014, 53, 8450–8454; Angew. Chem. 2014, 126, 8590–8594; c) J. McNul-
ty, D. McLeod, H. A. Jenkins, Eur. J. Org. Chem. 2016, 688–692.
[12] W. H. Perkin Jun., J. Chem. Soc. 1887, 51, 1–28.
[13] R. M. Willstätter, J. Bruce, Ber. Dtsch. Chem. Ges. 1907, 40, 3979–3999.
[14] CCDC 1456076 contains the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre.
5
608; Angew. Chem. 2014, 126, 5710–5714; c) L. Albrecht, G. Dickmeiss,
F. C. Acosta, C. Rodriguez-Escrich, R. L. Davis, K. A. Jørgensen, J. Am.
Chem. Soc. 2012, 134, 2543–2546. For two recent reviews, see: d) H.
Jiang, L. Albrecht, K. A. Jørgensen, Chem. Sci. 2013, 4, 2287–2300;
e) I. D. Jurberg, I. Chatterjee, R. Tannert, P. Melchiorre, Chem. Commun.
2013, 49, 4869–4883; f) G.-J. Duan, J.-B. Ling, W.-P. Wang, Y.-C. Luo, P.-F.
Xu, Chem. Commun. 2013, 49, 4625–4627.
6] a) I. Colomer, R. C. Barcelos, T. J. Donohoe, Angew. Chem. Int. Ed. 2016,
[
[
5
5, 4748–4752; Angew. Chem. 2016, 128, 4826–4830; b) M. Riener, D. A.
Nicewicz, Chem. Sci. 2013, 4, 2625–2629.
7] a) W. R. Gutekunst, P. S. Baran, J. Am. Chem. Soc. 2011, 133, 19076–
Received: April 19, 2016
1
9079; b) F. D. Piaz, A. Vassallo, A. Temraz, R. Cotugno, M. A. Belisario, G.
Published online on May 24, 2016
Chem. Eur. J. 2016, 22, 9111 – 9115
www.chemeurj.org
9115
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim