Regio- and Stereocontrolled Synthesis of 3-Substituted 1,2-Diazetidines by Asymmetric Allylic Amination of Vinyl Epoxide

Article abstract of DOI:10.1021/acs.orglett.7b00653

Pd-catalyzed asymmetric allylic amination of rac-vinyl epoxide with unsymmetrical 1,2-hydrazines proceeds with excellent regio- and stereocontrol, which after further ring closure provides differentially protected 3-vinyl-1,2-diazetidines in good yields.

Full text of DOI:10.1021/acs.orglett.7b00653

Letter
pubs.acs.org/OrgLett
Regio- and Stereocontrolled Synthesis of 3‑Substituted 1,2-
Diazetidines by Asymmetric Allylic Amination of Vinyl Epoxide
Sundaram Rajkumar, Guy J. Clarkson, and Michael Shipman*
Department of Chemistry, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, United Kingdom
S
* Supporting Information
ABSTRACT: Pd-catalyzed asymmetric allylic amination of
rac-vinyl epoxide with unsymmetrical 1,2-hydrazines proceeds
with excellent regio- and stereocontrol, which after further ring
closure provides differentially protected 3-vinyl-1,2-diazeti-
dines in good yields. The chirality at C-3 exerts stereocontrol
over the nitrogen centers in the 1,2-diazetidine with all
substituents orientating themselves trans to their neighbors.
Efficient functionalization without rupture of the strained ring
is demonstrated (e.g., by cross-metathesis), establishing the
first general route to C-3-substituted 1,2-diazetidines in
enantioenriched form.
our-membered heterocycles such as oxetanes1 and
azetidines2 are important substructures in drug discovery.
the diastereoselective synthesis of 3,4-disubstituted 1,2-diazeti-
dines via Pd-catalyzed cyclization of chiral 2,3-allenyl hydrazines
with aryl halides in good yields.5 Alternatively, Iacobini et al.
made 3-methyl-1,2-diazetidines by Rh-catalyzed asymmetric
hydrogenation of 3-methylene-1,2-diazetidines.6 However, nei-
ther method is well-suited for the synthesis of 2, which crucially
requires differentiation of the two nitrogen atoms. Here, we
describe a concise approach to 2 that exploits the asymmetric
allylic amination of vinyl epoxide with differentially protected
hydrazines, followed by ring closure to produce the four-
membered ring (Scheme 1). Manipulation of 2 to produce a
variety of 3-substituted 1,2-diazetidines is further demonstrated.
F
For example, the azetidine-containing MEK inhibitor Cobime-
tinib was recently approved for the treatment of melanoma.3
Consequently, there is interest in the development of other four-
membered heterocyclic templates for drug discovery programs.4
One potentially interesting scaffold is the 1,2-diazetidine nucleus
1 that contains a hydrazine subunit within a saturated four-
membered ring (Figure 1). This framework has significant
Scheme 1. Planned Approach to 1,2-Diazetidine 2
Figure 1. 3-Substituted 1,2-diazetidines as stereochemically defined
scaffolds for medicinal chemistry.
potential for diversification by facile substitution at the two
nitrogen atoms. Moreover, by introduction of a substituent at C-
3, as well as increasing structural diversity, stereochemical control
over the nitrogen centers via pyramidal inversion might be
realized such that all three substituents orient themselves trans to
their neighbors to minimize repulsive interactions.
With these ideas in mind, we sought to develop an efficient,
stereocontrolled route to 1,2-diazetidine 2 in either enantiomeric
form, bearing a vinyl group at C-3 and orthogonal protection of
the two nitrogen atoms (Pg1 ≠ Pg2). Such a molecule would
allow ready introduction of different groups on each nitrogen and
allow diversification at C-3 by way of synthetic manipulations of
the double bond (cross-metathesis, ozonolysis/reductive amina-
tion, etc.).
At the outset, it was unclear whether it would be possible to
identify conditions/substrates that would realize the highly
challenging allylic amination step. High regio- and enantiocon-
trol is needed in the opening of the epoxide via the internal
carbon atom. Encouragingly, Mangion et al. have shown that
1,2,2-trisubstituted hydrazines open vinyl epoxide at the internal
carbon with high levels of control under Pd catalysis.9 Moreover,
the identification of suitable N-protecting groups (Pg1 and Pg2)
Currently, the synthesis of enantiomerically enriched 3-
Received: March 3, 2017
substituted 1,2-diazetidines is poorly developed.5−8 Ma reported
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b00653
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
to achieve chemoselective opening by a single nitrogen of the
hydrazine was required. To address this question, we targeted
hydrazines with appreciable differences in nitrogen pKa values by
substituting one end with a carbamate (Pg1 = Cbz, Boc) and the
other with a sulfonamide (Pg2 = Ts, Ns).
Initially, we examined the dynamic kinetic resolution of rac-
vinyl epoxide (3) with hydrazine 4 to give alcohol 5 under
palladium catalysis. Promisingly, only the sulfonamide-bearing
nitrogen acts as nucleophile in this transformation. Optimization
studies explored the impact of variation in ligand, solvent, catalyst
loading, and additives on the yield and enantioselectivity of the
conversion (Table 1). Three ligands developed by Trost for the
further improvements in both yield and enantioselectivity
(entries 5 and 6), although iodide proved less effective (entry
7). Changes in cation structure had minimal effect (entry 8).
Reduced ligand loadings led to incomplete conversion (entry 9)
and addition of Cs2CO3 led to lower yield and enantioselectivity
(entry 10). Solvent optimization suggested that the highly polar
aprotic solvent, acetonitrile, gave better yields and better
enantioselectivity than CH2Cl2, THF, or toluene (entries 11−
13). Thus, the optimized conditions involve reaction of rac-3 (1.1
equiv) with hydrazine 4 using Pd2(dba)3 (1 mol %) and 6 (5 mol
%) in acetonitrile as solvent at room temperature with 1 equiv of
either Bu4NCl or BnN(Et)3Cl as additive. Under these
conditions, hydrazine 5 was produced in excellent yield and ee.
Moreover, products derived from nucleophilic attack through the
carbamate-protected nitrogen or by opening of the epoxide at the
unsubstituted carbon were not observed. The gross structure of 5
and the stereochemistry were confirmed by X-ray diffraction
(XRD) (see Supporting Information). The formation of (S)-5
using (R,R)-6 was deduced by the small value of the Flack
parameter.
Table 1. Optimization of Asymmetric Allylic Amination
product
Four additional hydrazines 9a−d with different N-protecting
groups were subjected to these reaction conditions (Table 2).
Interestingly, no control could be achieved using symmetrical
1,2-bistosyl hydrazine 9a, with both singularly and doubly
alkylated hydrazine products seen (entry 1). This finding is
consistent with our early observations that sulfonamides are
highly reactive nucleophiles in this transformation. Thus,
somewhat counterintuitively, better results were achieved with
the more complex hydrazine substrates 10b−d, where an
additional element of chemoselectivity was required. Substrates
bearing one sulfonamide (Ts or Ns) and one carbamate (Cbz or
Boc) group on the hydrazine gave excellent conversions and
enantioselectivities in this process (entries 2−4). The assignment
of configuration of 10b−d was made by analogy with that seen
with (S)-5. In the case of 10b, recrystallization further improved
the product ee (entry 2).
Next, we explored ring closure to the 1,2-diazetidine ring. In
such cyclizations, the use of a soft leaving group is important to
suppress competitive six-membered 1,3,4-oxadiazine ring
formation via closure through the carbamate oxygen atom.8
Thus, alcohol 5 was first converted to iodide 11a prior to base-
induced closure to 12a (Scheme 2). Use of Cs2CO3 in DMF
proved most effective for this step. Three differentially protected
1,2-diazetidines were successfully synthesized using this two-step
method in high ee (see Supporting Information). Tosyl
derivatives 12a and 12b bearing Cbz and Boc groups on the
second nitrogen were produced in excellent yields, with lower
yields seen for the corresponding nosyl derivative 12c. Single
crystal XRD performed on ent-12a, made using (S,S)-6,
confirmed the gross structure of 1,2-diazetidine 12a and the
absolute configuration at C-3. Encouragingly, all three
substituents within 12a orient themselves trans to their neighbors
to minimize repulsive interactions in the solid state, in
accordance with our hypothesis.
a
b
ligand
additive
(mol %)
yield
ee
entry Pd2(dba)3 (mol%)
(mol %)
(%)
(%)
1
2
3
4
5
6
7
8
9
1
6 (5)
6 (5)
7 (5)
8 (5)
6 (5)
6 (5)
6 (5)
6 (5)
6 (2.5)
6 (2.5)
6 (5)
6 (5)
80
84
79
80
92
92
82
80
f
66
87
77
85
93
93
69
87
nd
57
85
25
73
1
Bu4NBr
Bu4NBr
Bu4NBr
1
c
e
1
d
1
Bu4NCl
1
BnN(Et)3Cl
Bu4NI
1
1
Bu4PBr
1
Bu4NBr
Bu4NBr
Bu4NBr
Bu4NBr
Bu4NBr
g
h
i
10
11
12
13
0.5
1
70
89
80
77
1
j
1
6 (5)
a
b
c
After column chromatography. By chiral HPLC. (S,S)-8 as ligand
producing (R)-5. Contains 15% Bu4NBr. (S,S)-6 as ligand producing
(R)-5. Reaction not completed after 2 days. With 5 mol % of
Cs2CO3. In CH2Cl2. In toluene. In THF, reaction took 7 days.
d
e
f
g
h
i
j
enantiodiscrimination of Pd-complexed π-allyl intermediates
were evaluated, namely, 6−8 (Figure 2).10 Using 1 mol % of Pd
and 5 mol % of (R,R)-6, alcohol (S)-5 was produced in good yield
and encouraging ee (Table 1, entry 1). Previous reports
suggested that halide additives can influence the selectivity.9,11
Gratifyingly, addition of tetrabutylammonium bromide markedly
improved the enantioselectivity (entry 2). Use of ligands 7 and 8
proved less effective (entries 3 and 4). Use of chloride salts led to
Next, we explored if these chiral building blocks could be
successfully functionalized at C-3. Initial work focused on
homologation by way of Ru-catalyzed cross-metathesis (CM)
(Scheme 3).12 Although strained 1,2-diazetidines are known to
tolerate a number of Pd-,5,6c Rh-,6a and Cu-catalyzed6c bond-
forming processes, it was uncertain whether they would make
good substrates for Ru-catalyzed cross-metathesis. Using the
Hoveyda−Grubbs catalyst, good to excellent yields of 13a−j
were obtained upon reaction with a variety of terminal alkenes (3
Figure 2. Ligands used in asymmetric allylic amination.
B
DOI: 10.1021/acs.orglett.7b00653
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 2. Use of Other 1,2-Disubstituted Hydrazines
a
b
entry
1
substrate
R1 = Ts
additive
Bu4NBr
yield (%)
ee (%)
product
9a
9b
9c
9d
c
nd
10a
10b
10c
10d
R1 = Ts
R2 = Ts
R1 = Boc
R2 = Ts
R1 = Boc
R2 = Ns
R1 = Cbz
R2 = Ns
R2 = Ts
R1 = Boc
R2 = Ts
R1 = Boc
R2 = Ns
R1 = Cbz
R2 = Ns
de
,
f
2
3
4
BnN(Et)3Cl
Bu4NCl
95
98
93
89 (96)
g
89
Bu4NCl
92
a
b
c
d
After column chromatography. By chiral HPLC. Both single and double alkylation observed. Using (S,S)-6 as ligand providing (R)-10b.
e
f
g
Reaction took 48 h. After recrystallization from hexane/CH2Cl2. Determined after conversion to diazetidine 12c.
equiv) (Scheme 3). One exception was 13a, where the second-
Scheme 2. Synthesis of 3-Vinyl-1,2-diazetidines 12a−c
generation Grubbs catalyst was used. For less reactive alkenes,
the addition of a second aliquot of catalyst and alkene proved
necessary to drive the reactions to completion. Reactive
functional groups such as esters and halogens were well-
tolerated. In every example, only the (E)-alkene was observed.
For 13a, the structure and E-configuration were confirmed by
XRD. Using 13g, we confirmed that no racemization occurred
during the CM process.
The C-3 vinyl group can be oxidatively cleaved to the
corresponding aldehyde by ozonolysis and efficiently converted
in situ to alcohol 14 by reduction or amine 15 by reductive
amination using sodium triacetoxyborohydride (Scheme 4).
Scheme 4. Functionalization of 3-Vinyl-1,2-diazetidine
Scheme 3. Cross-Metathesis Using 3-Vinyl-1,2-diazetidine
Both products were produced in high ee, indicating that the
intermediate aldehyde is not prone to racemization under these
conditions. We speculate that the four-membered ring may play a
role in inhibiting any such racemization because the formation of
the requisite enol/enolate intermediate would result in the
introduction of an energetically disfavored sp2-hybridized carbon
with the four-membered ring.
Reduction of the alkene functionality is also possible. Diimide
proved to be effective for the formation of the corresponding
saturated 1,2-diazetidines (Scheme 5).13 Thus, N-ethyl deriva-
tives 16 and 17 could be readily made without concomitant
reductive opening of the strained ring.14 Controlled cleavage of
the N−N bond to yield enantiomerically enriched 1,2-diamines
is also demonstrated. Using Raney Ni as catalyst at 50 bar
hydrogen pressure, saturated 1,2-diazetidine 17 was successfully
converted into differentially protected 1,2-diamine 18 in
excellent yield and ee (Scheme 5). The reductive cleavage of
14 was also achieved in 86% yield and 97% ee (see Supporting
a
Grubbs-II catalyst used, whose structure is provided in the
b
c
Supporting Information. 92% ee by chiral HPLC. After 48 h,
additional catalyst (5 mol %) and alkene (3 equiv) added.
C
DOI: 10.1021/acs.orglett.7b00653
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 5. Reduction of 3-Vinyl-1,2-diazetidines
ACKNOWLEDGMENTS
■
We thank The Leverhulme Trust (RPG-2014-362) for generous
financial support. Crystallographic data were collected using an
instrument that received funding from the European Research
Council under the European Union’s Horizon 2020 research and
innovation program (Grant Agreement No. 637313).
REFERENCES
■
(1) (a) Bull, J. A.; Croft, R. A.; Davis, O. A.; Doran, R.; Morgan, K. F.
Chem. Rev. 2016, 116, 12150. (b) Burkhard, J. A.; Wuitschik, G.; Rogers-
Information for details). This methodology could potentially be
used to develop a new route to chiral 1,2-diamines.15
Selective removal of the protecting groups within 17 was
possible (Scheme 6). The Boc group was removed with TFA to
give 19 in excellent yield. Separately, the tosyl group was cleaved
using Mg in MeOH to provide 20 in 75% yield.
Evans, M.; Muller, K.; Carreira, E. M. Angew. Chem., Int. Ed. 2010, 49,
9052.
(2) (a) Antermite, D.; Degennaro, L.; Luisi, R. Org. Biomol. Chem.
2017, 15, 34. (b) Bott, T. M.; West, F. G. Heterocycles 2012, 84, 223.
(c) Brandi, A.; Cicchi, S.; Cordero, F. M. Chem. Rev. 2008, 108, 3988.
(3) Garnock-Jones, K. P. Drugs 2015, 75, 1823.
̈
(4) (a) Carreira, E. M.; Fessard, T. C. Chem. Rev. 2014, 114, 8257.
(b) Pitts, C. R.; Lectka, T. Chem. Rev. 2014, 114, 7930. (c) Rousseau, G.;
Robin, S. In Modern Heterocyclic Chemistry; Alvarez-Builla, J., Vaquero, J.
J., Barluenga, J., Eds; Wiley-VCH: Weinheim, Germany, 2011; Chapter
3, pp 163−268.
Scheme 6. Selective Deprotection of 1,2-Diazetidine 17
(5) Cheng, X.; Ma, S. Angew. Chem., Int. Ed. 2008, 47, 4581.
(6) (a) Iacobini, G. P.; Porter, D. W.; Shipman, M. Chem. Commun.
2012, 48, 9852. See also (b) Xu, S.; Chen, J.; Shang, J.; Qing, Z.; Zhang,
J.; Tang, Y. Tetrahedron Lett. 2015, 56, 6456. (c) Brown, M. J.; Clarkson,
G. J.; Inglis, G. G.; Shipman, M. Org. Lett. 2011, 13, 1686.
(7) Miao, W.; Xu, W.; Zhang, Z.; Ma, R.; Chen, S.-H.; Li, G.
Tetrahedron Lett. 2006, 47, 6835. Doubt has been raised as to whether
these authors produced the reported chiral 3-substituted 1,2-
diazetidines with six-membered 1,3,4-oxadiazine formation more likely
(see ref 8).
(8) Brown, M. J.; Clarkson, G. J.; Fox, D. J.; Inglis, G. G.; Shipman, M.
Tetrahedron Lett. 2010, 51, 382.
(9) Mangion, I.; Strotman, N.; Drahl, M.; Imbriglio, J.; Guidry, E. Org.
Lett. 2009, 11, 3258.
(10) (a) Trost, B. M.; Van Vranken, D. L.; Bingel, C. J. Am. Chem. Soc.
1992, 114, 9327. (b) Trost, B. M.; Bunt, R. C.; Lemoine, R. C.; Calkins,
T. L. J. Am. Chem. Soc. 2000, 122, 5968.
(11) Trost, B. M.; Malhotra, S.; Chan, W. H. J. Am. Chem. Soc. 2011,
133, 7328 and references cited therein.
(12) (a) Connon, S. J.; Blechert, S. Angew. Chem., Int. Ed. 2003, 42,
1900. (b) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18.
(13) Pasto, D. J.; Taylor, R. T. Org. React. 1991, 40, 91.
(14) Hydrogenations using metal catalysts led to partial over-
reduction. For example, reduction [10% Pd/C, H2 (1 atm), MeOH,
rt] of 12b led to formation of 17 (27%) with quantities of TsNH−
N(Boc)(CH2)3CH3 resulting from concomitant reduction of the C(3)−
N bond of the diazetidine ring.
(15) For reviews on 1,2-diamine synthesis, see: (a) Grygorenko, O. O.;
Radchenko, D. S.; Volochnyuk, D. M.; Tolmachev, A. A.; Komarov, I. V.
Chem. Rev. 2011, 111, 5506. (b) Lucet, D.; Le Gall, T.; Mioskowski, C.
Angew. Chem., Int. Ed. 1998, 37, 2580. (c) Bennani, Y. L.; Hanessian, S.
Chem. Rev. 1997, 97, 3161.
In summary, a three-step asymmetric synthesis of enantioen-
riched 3-vinyl-1,2-diazetidines has been developed in which the
two nitrogen atoms are differentially substituted with carbamate
and sulfonamide protecting groups. In the key asymmetric allylic
amination reaction, only the sulfonamide-bearing nitrogen acts
as nucleophile. Crystallography evidence suggests that the
chirality at C-3 exerts stereocontrol over both nitrogen centers
such that all substituents orient themselves trans to their
neighbors. The chemical integrity of the strained 1,2-diazetidines
was maintained during a range of chemical transformations
including (i) Ru-based cross-metathesis; (ii) ozonolysis followed
by hydride reduction; (iii) alkene reduction using diimide; and
(iv) treatment with TFA. However, they can be reductively
cleaved to 1,2-diamines or selectively deprotected under
appropriate conditions. Taken together, these findings suggest
that 3-substituted 1,2-diazetidines may make excellent building
blocks for drug discovery and/or asymmetric catalysis, avenues of
work which are being actively explored in our laboratory.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.7b00653.
Experimental procedures and characterization data for all
new compounds, copies of 1H and 13C NMR spectra, and
chiral HPLC traces (PDF)
X-ray data for 5 (CIF)
X-ray data for 12a (CIF)
X-ray data for 13a (CIF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: m.shipman@warwick.ac.uk.
ORCID
Michael Shipman: 0000-0002-3349-5594
Notes
The authors declare no competing financial interest.
D
DOI: 10.1021/acs.orglett.7b00653
Org. Lett. XXXX, XXX, XXX−XXX