Construction of Morphan Derivatives by Nitroso–Ene Cyclization: Mechanistic Insight and Total Synthesis of (±)-Kopsone

Article abstract of DOI:10.1002/anie.201706018

A type II nitroso–ene cyclization was developed for the construction of morphan derivatives with good functional-group tolerance. DFT calculations revealed that the nitroso–ene reaction proceeds in a stepwise manner involving diradical or zwitterionic intermediates. The rate-determining step is C?N bond formation, followed by a rapid hydrogen-transfer step with a chair-conformation transition state. The current approach was also successfully applied in the first total synthesis of (±)-kopsone, a highly strained yet simple morphan-type alkaloid isolated from Kopsia macrophylla.

Full text of DOI:10.1002/anie.201706018

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Building Morphan Derivatives by Nitroso-Ene Cyclization:
Mechanistic Insights and Total Synthesis of (±)-Kopsone
Authors: Ran Hong, Li Zhai, Xuechao Tian, Chao Wang, Wenhua Li,
Zhi-Xiang Yu, Sha-Hua Huang, and Qi Cui
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.201706018
Angew. Chem. 10.1002/ange.201706018
Link to VoR: http://dx.doi.org/10.1002/anie.201706018
http://dx.doi.org/10.1002/ange.201706018

10.1002/anie.201706018
Angewandte Chemie International Edition
COMMUNICATION
Building Morphan Derivatives by Nitroso-Ene Cyclization:
Mechanistic Insights and Total Synthesis of (±)-Kopsone
Li Zhai[a], Xuechao Tian[b], Chao Wang[a], Qi Cui[c], Wenhua Li[a], Sha-Hua Huang*,[b], Zhi-Xiang Yu*,[c],
and Ran Hong*,[a]
Abstract: A type II nitroso-ene cyclization was developed for the
construction of morphan derivatives with a good functional group
tolerance. DFT calculations revealed the nitroso-ene reaction is
stepwise involving diradical or zwitterionic intermediates. The rate-
determining step is the C-N bond formation, following by a rapid
hydrogen transfer step with a chair-conformation transition state.
The current approach was also successfully applied in the first total
synthesis of (±)-kopsone, a highly strained yet simple morphan-type
alkaloid isolated from Kopsia macrophylla.
transition state (TS), the N-acyl nitroso would adopt an
electronically demanding s-cis rotamer of acylnitroso in the
possible pericyclic reaction. To prove this concept, a simple
cyclohexene-derived hydroxamic acid 1a, readily prepared on
the gram scale from 1,4-cyclohexanedione,[8] was designed for
the cyclization. After treatment of tetrapropylammonium
periodate (nPr4NIO4), the proposed transient N-acyl nitroso
species rapidly underwent the subsequent ene reaction to
deliver a cyclized product 2a, and its structure was further
unambiguously confirmed by X-ray analysis of its acetate
derivative 3a (Scheme 1C).[9]
Morphan, also named 2-azabicyclo[3.3.1]nonane, is abundant in
complex bioactive alkaloids and represents a privileged structure
in drug discovery.[1] Strychnine and morphine are arguably the
most known molecules in this category (Figure 1) because of
their broad biological profile, clinically usage for the treatment of
human diseases, and their intriguing structures that inspire novel
method developments.[2] For example, the practical synthesis of
morphine remains an intriguing objective for the synthetic
community despite more than 30 syntheses emerging since
Gates’s masterpiece in 1956.[3] Continuing efforts in this field
inaugurate state-of-the-art synthetic methods and tactics, which
underline the efficient routes to access medicinally significant
alkaloids.
Figure 1. The Morphan core, morphine, and strychnine.
The recent application of the nitroso-ene reaction motivates
us,[4] with synthetic application in mind, to further explore the
unique reactivity of N-acyl nitroso with implementation of a new
cyclic skeleton.[5] Based on Oppolzer and Snieckus’s
classification on the intramolecular ene reaction,[6] reaction at
the proximal end of the tethered alkene to the nitroso group
would result in
a
type
I
ene product, such as 2-
azabicyclo[3.2.1]octane[4] and other ring systems[5] (Scheme 1A).
We envisioned that the morphan skeleton could be constructed
via an alternative type II mode, in which the tether is attached to
the distal end of the alkene (Scheme 1B). However, this type of
intramolecular nitroso-ene reaction remains underdeveloped
since Kirby’s pioneer study in 1985.[7] As shown in the illustrative
[a]
L. Zhai, C. Wang, W. Li, Prof. Dr. R. Hong,
Key Laboratory of Synthetic Chemistry of Natural Substances
Shanghai Institute of Organic Chemistry (CAS)
345 Lingling Road, Shanghai 200032, China
University of Chinese Academy of Sciences
19A Yuquan Road, Beijing 100049, China
E-mail: rhong@sioc.ac.cn
Scheme 1. Reaction design. (A) Type I nitroso-ene cyclization. (B) Type II
nitroso-ene cyclization developed in this work. The definition of “tether” refers
to the longest unbranched carbon chain and is illustrated in bold in the
proposed transition state (TS). (C) Attempt to construct morphan via the
nitroso-ene reaction. 9,10-DMA = 9,10-dimethylanthracene; Ac = acetyl; Et3N
= triethylamine; DMAP = 4-N, N-dimethylaminopyridine.
[b]
[c]
X. Tian, Prof. Dr. S.-H. Huang
School of Chemical and Environmental Engineering
Shanghai Institute of Technology
100 Haiquan Road, Shanghai 201418, China
E-mail: shahua@sit.edu.cn
Q. Cui, Prof. Dr. Z.-X. Yu
College of Chemistry
Peking University, Beijing 100871, China
E-mail: yuzx@pku.edu.cn
The excellent regioselectivity for the nitroso-ene cyclization
promoted us to explore the functional group tolerance, which
may greatly improve the future utility of this method in organic
synthesis. As shown in Table 1, the cyclization was competent
for substrates bearing a variety of functional groups, such as
Supporting information for this article is given via a link:
http://dx.doi.org/xxx
This article is protected by copyright. All rights reserved.

10.1002/anie.201706018
Angewandte Chemie International Edition
COMMUNICATION
various alkyls and secondary amide (1b-j). The resulting
secondary hydroxamic acids were found to be highly polar and
difficult to elute from a silica gel.[10] Therefore, acetylation was
immediately performed for easy purification and characterization
of the corresponding products. Generally, the stereoselectivity of
the resulting exo-alkene was low despite introducing bulky
substituents (3d versus 3e; the structure determination was
assigned as shown in Table 1B).[8] Nevertheless, the overall
yields of the cyclized products were good to excellent. It is
known that acylnitroso compounds readily react with oxa-, thio-
or aza-nucleophiles to release nitrioxyl (HNO).[11] To our surprise,
for substrates 1k-m bearing free hydroxyl groups at the side
chain and the bridged carbon, the ene cyclization remained
kinetically favorable and delivered the morphan derivatives in
high yields.
derived hydroxamic acids 1n-t were prepared and subjected to
the optimal conditions (Table 2). The requisite cyclization
proceeded smoothly to give the corresponding bicyclic products
in good yields. The inherent stereogenic centers (C8 with R3 in
1o and 1p) and protecting groups at N (R4 in 1q-t) did not
deteriorate the reaction, which suggest the potential application
for late-stage modification in complex substrates. The current
approach also provides a novel ring construction to access
interesting lead compounds for medicinal development.
Table 2. Ene cyclization of heterocyclic substrates [a,b]
Table 1. Ene cyclization of carbocyclic substrates [a,b,c,d]
[a] Isolated yield after the two steps is shown in parentheses. [b] The acetyl
group introduced at step b is highlighted in gray. Boc = tert-butoxycarbonyl;
Ts = p-toluenesulfonyl; RT = room temperature.
The cyclopropyl group of 1c was kept intact in the ene
reaction, which was rather striking to us due to the sensitivity of
adjacent cyclopropane[13] towards radical or cationic pathways.
To gain more insights to the mechanism, a variety of solvents
were examined (Table S1 in the Supporting Information).
Although the reaction rate attenuated with a decrease of the
solvent polarity, the ene product 3c (the acetate form) remained
prominent, and the ring opening of the cyclopropane product
was not detected (Scheme 2A). Moreover, when a phenyl-
substituted hydroxamic acid 1u was executed for the ene
reaction, no requisite product 2u or 3u was formed, suggesting
the double bond could not be located within the cyclohexane
ring of the product (Scheme 2B).
[a] (A) Substrate scope and (B) stereochemistry determination of 3d.
Reaction conditions: hydroxamic acid (1 mmol), nPr4NIO4 (0.5 mmol), CHCl3
(40 mL), 0 °C, 10 min; then Ac2O (1.2 mmol), Et3N (1.3 mmol), DMAP (0.1
mmol), RT, 10~30 min. [b] The acetyl group introduced at step
b
(highlighted in gray). [c] Isolated yield for the two steps is shown in
parentheses. [d] The Z/E isomer was tentatively assigned based on the
characterization of 3d, and the ratio (shown in parentheses) was
determined by 1H NMR (400 MHz, acetone-d6). [e] See the Supporting
Information for details. [f] Isolated yield of the ene product. TBS = tert-butyl-
dimethylsilyl; TBDPS = tert-butyldiphenylsilyl; Bn = benzyl.
Scheme 2. Experimental execution of 1c and 1u.
Additional heteroatom-embedded morphan rings were also
documented as intriguing platforms for derivatization.[12] With this
information in mind, tetrahydropyran and tetrahydropyridine-
This article is protected by copyright. All rights reserved.

10.1002/anie.201706018
Angewandte Chemie International Edition
COMMUNICATION
Figure 2. Mechanistic study with DFT calculations at the PCM(CHCl3)/(U)B3LYP/6-31G(d)//(U)B3LYP/6-31G(d) level. IN = intermediate; TS = transition state; rot
= rotation.
Through DFT calculations at the (U)B3LYP/6-31G(d) level,
we found that these ene reactions are stepwise involving
zwitterionic or diradical intermediates (Figure 2).[14,15] The first
step of the ene reaction of 1d’, derived from 1d by oxidation, is
the addition of nitroso to the alkene moiety of the substrate,
generating IN1-Z via TS1-Z or IN1-E via TS1-E (Figure 2a). TS1-
Z and TS1-E are closed-shell species, with the computed
activation Gibbs free energies of about 10.6 kcal/mol in
chloroform solution. Intermediate IN1-Z is a diradical species,
but IN1-E is a zwitterionic species.[16] Intermediate IN1-E can
directly give 2d-E through hydrogen transfer transition state with
an activation free energy of 0.3 kcal/mol. Intermediate IN1-Z can
also directly give 2d-Z through hydrogen transfer process (via
TS2-Z with an activation free energy of 6.9 kcal/mol).
Interestingly, IN1-Z could undergo a C-C bond rotation via
transition state TS-rot (with a computed activation free energy of
5.3 kcal/mol) to IN1-E, which then gives 2d-E. From these
calculations, we can conclude that the ene reaction has two
parallel pathways: 1d’→TS1-Z →IN1-Z →2d-Z; 1d’→TS1-E →
IN1-E→2d-E, together with a bifurcation from IN1-Z→TS-rot→
2d-E. The above whole potential energy surface indicates that
the first step of the ene reaction is irreversible. Because TS1-Z
and TS1-E are close in energy, TS-rot and TS2-Z are also close
in energy, we can estimate that 2d-E and 2d-Z have similar
reaction yields. This is consistent with the experiment (Z/E 1.7:1
for 3d in Table 1).
conjugated substrate 1u’ (Figure 2c), derived from oxidation of
1u, DFT calculations revealed the hydrogen transfer step is very
difficult with an activation free energy of 28.0 kcal/mol.
Consequently, other side reactions could become favored and
no target ene product 2u (or 3u) would be formed. The difficulty
of hydrogen transfer is mainly due to the boat like transition
structure of the forming 6-membered ring in TS7 while for
substrate 1d’, this hydrogen transfer transition state has a
favorable chair conformation (see the Supporting Information).
Having established method to access morphan skeleton, we
embarked on the synthesis of kopsone, the simplest morphan-
type monoterpene alkaloid known.[17] The structure features
three methyl groups pointed in the same orientation and the
carbonyl functionality at the bridged carbon. Although the
biogenesis origin and bioactivity remain unknown, the intriguing
strained architecture is anticipated as an ideal testing ground for
the derivatization of 2m with a preinstalled hydroxyl group at C6.
Hydrogenation of the terminal alkene in 2m was enabled
under a high pressure of hydrogen when the reaction was run in
a 4.8-gram scale (Scheme 3). Interestingly, both the alkene and
N-hydroxyl group were reduced to deliver amide 4 in excellent
yield. It is gratifying that the requisite chirality at C7 was
unambiguously verified by X-ray analysis of 4.[9] To refute a
possible O-methylation and a late-stage introduction of a methyl
group at C4 with a proper facial selectivity, protection with a silyl
group was undertaken prior to the N-methylation of the amide to
deliver compound 6. The following methylation at C4 proved to
be challenging due to the steric hindrance. After screening
several basic conditions, tert-butyllithium[18] was found to
effectively realize methylation to the α-site of the amide albeit
with the opposite stereochemistry, which was further epimerized
under the repetitive basic conditions. Although the conversion
In the case of 1c’ bearing a cyclopropyl group, the ene
reaction is also stepwise with a singlet diradical IN2 (Figure 2b).
It is suggested that IN2 can undergo the cyclopropane ring-
opening reaction via TS5, but this is disfavored compared to the
rapid hydrogen transfer via TS4 to give the final ene product 2c
(the acetate form 3c). The calculation results also agree with the
experimental observations shown in Tables 1 and S1. For the
This article is protected by copyright. All rights reserved.

10.1002/anie.201706018
Angewandte Chemie International Edition
COMMUNICATION
Acknowledgements
Financial support from the National Natural Science Foundation
of China (21672236 and 21290184 to RH, 21402121 to SHH,
21232001 to ZXY), the Strategic Priority Research Program (B)
(XDB20000000), and the Key Research Program of Frontier
Sciences (QYZDY-SSW-SLH026) of the Chinese Academy of
Sciences are highly appreciated. RH also thanks the Shanghai
Science and Technology Commission (15JC1400400) for partial
support. We are grateful to Jie Sun (SIOC) for X-ray analysis.
Conflict of interest
The authors declare no conflict of interest.
Keywords: alkaloid • cyclization • diradical • morphan core •
nitroso-ene
[1]
a) D. C. Palmer, M. J. Strauss, Chem. Rev. 1977, 77, 1; b) J. Bosch, J.
Bonjoch, Hetereocycles 1980, 14, 505; c) J. Bonjoch, F. Diaba, B.
Bradshaw, Synthesis 2011, 993; d) B. Kang, P. Jakubec, D. J. Dixon,
Nat. Prod. Rep. 2014, 31, 550; for recent selected methods toward
morphan synthesis, see: e) B. Bradshaw, C. Parra, J. Bonjoch, Org.
Lett. 2013, 15, 2458; f) A. D. G. Yamagata, S. Datta, K. E. Jackson, L.
Stegbauer, R. S. Paton, D. J. Dixon, Angew. Chem. Int. Ed. 2015, 54,
4899; Angew. Chem. 2015, 127, 4981; g) R. Manzano, S. Datta, R. S.
Paton, D. J. Dixon, Angew. Chem. Int. Ed. 2017, 56, 5834; Angew.
Chem. 2017, 129, 5928; h) B. Li, R. Liu, R. Liang, Y. Jia, Acta Chim.
Sinica 2017, 75, 448.
Scheme 3. The complete synthesis of (±)-kopsone. Reaction conditions: a)
10% Pd/C (25 wt%), H2 (60 atm), MeOH, 40 °C, 24 h, quant. b) TBSCl (1.5
equiv), imidazole (1.5 equiv), DMAP (0.15 equiv), DMF, 0 °C to 80 °C, 2 h,
73%. c) NaH (2.5 equiv), MeI (2.5 equiv), THF, 0 °C to RT, 2 h, 98%. d) tBuLi
(1.6 M in heptane, 3 equiv), MeI (5 equiv), THF, -78 °C, 2 h, 97%. e) tBuLi (1.6
M in heptane, 6 equiv), THF, -78 °C; then NH4Cl (sat., aq.), 1 h, 95% (brsm). f)
HF (aq., 10 equiv), CH3CN, 0 °C to RT, 4 h, 85%. g) BH3·Me2S (2 M in THF,
25 equiv), THF, 0 °C to 80 °C, 4 h, 83%. h) (COCl)2 (1.3 equiv), DMSO (2.4
[2]
[3]
[4]
a) J. S. Cannon, L. E. Overman, Angew. Chem. Int. Ed. 2012, 51, 4288;
Angew. Chem. 2012, 124, 4362; b) U. Rinner, T. Hudlicky, Top. Curr.
Chem. 2012, 309, 33.
a) M. Gates, G. Tschudi, J. Am. Chem. Soc. 1956, 78, 1380; For a
recent synthesis of morphine, see: b) H. Umihara, S. Yokoshima, M.
Inoue, T. Fukuyama, Chem. Eur. J. 2017, 23, 6993.
equiv), Et3N (6 equiv), CH2Cl2, -78 °C,
2 h, 75%. DMF = N, N-
a) J. Ouyang, R. Yan, X. Mi, R. Hong, Angew. Chem. Int. Ed. 2015, 54,
10940; Angew. Chem. 2015, 127, 11090; b) J. Ouyang, X. Mi, Y. Wang,
R. Hong, Synlett 2017, 28, 762.
dimethylformamide; brsm = based on the recovered starting material; DMSO =
dimethyl sulfoxide; THF = tetrahydrofuran.
[5]
[6]
M. G. Memeo, P. Quadrelli, Chem. Rev. 2017, 117, 2018.
a) W. Oppolzer, V. Snieckus, Angew. Chem. Int. Ed. 1978, 17, 476;
Angew. Chem. 1978, 90, 506; b) K. Mikami, M. Shimizu, Chem. Rev.
1992, 92, 1021; b) B. B. Snider, In Comprehensive Organic Synthesis,
Vol. 2 (Eds.: G. A. Molander, P. Knochel), 2nd ed., Oxford, Elsevier,
2014, Chap. 2.03, pp. 148-191.
was moderate (50% on a 4-gram scale), an overall 83% isolated
yield of 8 was achieved after three runs. After removal of the silyl
group, the free hydroxyl greatly facilitated the borane reduction
of the amide to deliver tertiary amine 10, while amide 8 was
obstinate to the harsh reduction conditions (e.g., LiAlH4). The
subsequent Swern oxidation with freshly distilled oxalyl
dichloride[19] proceeded smoothly to afford (±)-kopsone, whose
spectral data were consistent with the literature report.[17]
In summary, a carbon-tethered type II nitroso-ene cyclization
was developed to construct morphan derivatives. The substrates
bearing free hydroxyl and amide groups as well as heteroatoms
embedded in the second ring were particularly of interest for
future medicinal development. The current approach was also
successfully applied to the first synthesis of (±)-kopsone.
Quantum chemical calculations indicated that the present
nitroso-ene reactions are stepwise, in which the first step of
formation of the C-N bond to generate diradical or zwitterionic
intermediates is rate determining, and the second step is a facile
hydrogen transfer reaction with a chair-like transition structure.
These mechanistic studies reveal the unique reactivity of the N-
acyl nitroso group for the ene reaction, which lays ground work
for the future development of novel reaction modes in the
context of complex alkaloid synthesis.
[7]
For sporatic entries only with C-nitrosoformate esters, see: G. W. Kirby,
H. McGuigan, D. McLean, J. Chem. Soc., Perkin Trans 1 1985, 1961.
See the Supporting Information for details.
[8]
[9]
The crystallographic data of 3a, Z-3d, 3p, 3r, 4, and 9 was summarized
in the Supporting Information and can be obtained free of charge from
The Cambridge Crystallographic Data Centre.
[10] a) Y. K. Agrawal, Russ. Chem. Rev. 1979, 48, 948; b) C. J. Marmion, J.
P. Parker, K. B. Nolan, In Comprehensive Inorganic Chemistry II, From
Elements to Applications, J. Reedijk, K. Poeppelmeier, Eds.; Elsevier,
2014, Chap 3.23, pp. 683-708.
[11] F. Doctorovich, P. J. Farmer, M. A. Marti, The Chemistry and Biology of
Nitroxyl (HNO), 1st ed., Amsterdam, Elsevier, 2017.
[12] a) M. Prakesch, S. Srivastava, D. M. Leek, P. Arya, J. Comb. Chem.
2006, 8, 762; b) K. Wiedemeyer, B. Wünsch, Carbohydr. Res. 2012,
359, 24; c) S. Hardy, S. F. Martin, Tetrahedron 2014, 70, 7142.
[13] D. Griller, K. U. Ingold, Acc. Chem. Res. 1980, 13, 317.
[14] a) A. G. Leach, K. N. Houk, J. Am. Chem. Soc. 2002, 124, 14820; b) A.
G. Leach, K. N. Houk, Org. Biomol. Chem. 2003, 1, 1389; c) Z.-X. Yu，
K. N. Houk, J. Am. Chem. Soc. 2003, 125, 13825; d) X. Lu, Org. Lett.
2004, 6, 2813; e) N. C. James, J. M. Um, A. B. Padias, H. K. Hall, Jr., K.
N. Houk, J. Org. Chem. 2013, 78, 6582.
[15] Here we did not consider the in situ generation of the nitroso compound
and its final elaboration of the ene product by acylation.
This article is protected by copyright. All rights reserved.

10.1002/anie.201706018
Angewandte Chemie International Edition
COMMUNICATION
[16] The intermediates shown in Figure 2 are either (polarized) diradical or
zwitterionic species, which form being taken could be determined by the
combined effects of substitution, conformation, and solvation (See
Supporting Information for more discussions).
[17] C. Kan-Fan, T. Sevenet, H. A. Hadi, M. Bonin, J.-C. Quirion, H.-P.
Husson, Nat. Prod. Lett. 1995, 7, 283.
[18] A. I. Meyers, K. B. Kunnen, W. C. Still, J. Am. Chem. Soc. 1987, 109,
4405.
[19] K. Omura, D. Swern, Tetrahedron 1978, 34, 1651.
This article is protected by copyright. All rights reserved.

10.1002/anie.201706018
Angewandte Chemie International Edition
COMMUNICATION
Entry for the Table of Contents (Please choose one layout)
Synthetic Methods
Li Zhai, Xuechao Tian, Chao Wang, Qi
Cui, Wenhua Li, Sha-Hua Huang*, Zhi-
Xiang Yu*, and Ran Hong*
Page No. – Page No.
Building Morphan Derivatives by
Nitroso-Ene Cyclization: Mechanistic
Insights and Total Synthesis of (±)-
Kopsone
ENErgize Morphan! A complementary type II nitroso-ene cyclization was
developed to construct morphan derivatives with good tolerance of a variety of
functional groups. DFT calculations have been used to understand the reaction
mechanism. This method was also successfully applied in the first total synthesis
and structural verification of (±)-kopsone, a unique monoterpenoid found in Nature.
This article is protected by copyright. All rights reserved.