Base-Controlled Diastereoselective Synthesis of Either anti- or syn-β-Aminonitriles

Article abstract of DOI:10.1021/acs.orglett.7b00679

Deprotonation of secondary alkane nitriles with nBuLi and addition to aryl imines gives kinetic anti-β-aminonitriles. Use of LHMDS allows reversible protonation of the reaction intermediate to give syn-β-aminonitriles. The pure diastereosiomers can be iso

Full text of DOI:10.1021/acs.orglett.7b00679

Letter
pubs.acs.org/OrgLett
Base-Controlled Diastereoselective Synthesis of Either anti- or syn-β-
Aminonitriles
†
†
James C. Anderson,* Ian B. Campbell, Sebastien Campos, Christopher D. Rundell, Jonathan Shannon,
and Graham J. Tizzard*
,‡
Department of Chemistry, University College London, 20 Gordon Street, London WC1H 0AJ, U.K.
S Supporting Information
*
ABSTRACT: Deprotonation of secondary alkane nitriles with nBuLi and
addition to aryl imines gives kinetic anti-β-aminonitriles. Use of LHMDS
allows reversible protonation of the reaction intermediate to give syn-β-
aminonitriles. The pure diastereosiomers can be isolated in good yields, and
the mechanism was elucidated.
he nitrile functional group is ubiquitous in synthesis due
to its ease of incorporation, ability to facilitate the
complement existing stereoselective methodology. We commu-
T
nicate here a synthesis of β-aminonitriles which by judicious
choice of base can give either anti- or syn-β-aminonitriles with
alkanenitriles and offer some mechanistic understanding of this.
We recently developed conjugate addition nitro-Mannich
reactions where a nitronate anion generated by conjugate
formation of new bonds, and interconversion into a myriad of
1
other functional groups. It is present in biologically active
natural products, and its use as a pharmacophore has also been
2
recognized. The β-aminonitrile functional group is an under-
3
22
investigated subset of alkanenitriles. There are many naturally
addition to a nitroalkene underwent addition to an imine. We
4
−6
attempted an analogous reaction with cyanoalkenes. Although
conjugate reduction using Superhydride proceeded cleanly,
disappointingly, no 1,2-addition to the imine was observed with
or without Brønsted or Lewis acids in a range of solvents.
To confirm that 1,2-addition could occur, we generated the
corresponding α-cyano carbanion from 3-phenylpropanenitrile
occurring β-aminonitriles and pharmaceuticals.
The β-
7
aminonitrile is also a precursor to 1,3-diamines and β-amino
acid derivatives.
Routes to the β-aminonitrile functional group include ring
8
opening of aziridines with cyanide, conjugate addition of
9
10
amines to acrylonitriles, hydrocyanation of nitroalkenes, and
1
with a range of bases and then added imine (Scheme 1, R =
Thorpe−Ziegler reaction between two nitriles followed by
11
conjugate reduction. The most efficient method, and that
which has led to many enantioselective syntheses through
asymmetric transition metal catalysis, is the addition of
Scheme 1. 1,2-Addition of Alkane Nitriles to Aldimines
12
alkanenitriles to imines. Deprotonation of alkyl nitriles
1
3
(
CH CN, pK 31.3, PhCH CN, pK 21.9 in DMSO) and
3 a 2 a
3a
subsequent reaction with electrophiles requires strong bases,
such as LDA and proazaphosphatranes, which can trigger
undesired side reactions such as epimerization or elimina-
1
4,15
16
15
1
6
2
tion. Increasing the acidity of alkylnitriles through coordina-
tion with catalytic transition-metal complexes has allowed the
use of weaker bases for this reaction. Examples of
alkanenitriles other than acetonitrile often give poor diaster-
eoselectivities. Good enantioselectivities and in some cases
good diastereoselectivities have been obtained in asymmetric
catalyzed reactions, and the major diastereoisomer isolated
from these can be either anti or syn depending upon the type of
nitrile and catalyst used. A rare and recent example of the
addition of a lithio nitrile anion, generated by treatment with
LHMDS, to an imine as part of a complex natural product
Bn, R = 2-BrC H ). The 2-bromophenyl imine was used in
6
5
preparation for a possible Pd-catalyzed intramolecular
1
2
22c,g,j
amination.
We observed that the diastereomeric ratio of
the product β-aminonitriles was dependent on the nature of the
1
7
n
base and temperature (Table 1). With the base BuLi, the anti-
β-aminonitrile was observed with a diastereomeric ratio (dr) =
23
85:15 at −78 °C (entry 1). A repeat reaction with warming to
rt for 5 min switched the diastereoselectivity to favor the syn-β-
aminonitrile 15:85 (entry 2). In the case of LHMDS, the syn-β-
aminonitrile was observed with dr = 10:90 at −78 °C (entry 3).
Warming to room temperature as before gave no change in the
sense or extent of diastereoselectivity (entry 4). LDA gave
18
19
synthesis gave no diastereoselectivity. The same reaction
using Ellman’s auxiliary on the imine gave a diastereose-
n
20
essentially the same results as BuLi (compare entries 1 and 2
n
21
with 5 and 6). This similarity of BuLi and LDA had been
lectivity of ∼15:85 anti/syn for the product β-aminonitrile.
There is a clear need to be able to synthesize β-aminonitriles
from longer alkanenitriles than acetonitrile and to understand
how to access either anti or syn diastereoisomers to
Received: March 8, 2017
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b00679
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Table 1. Effect of Base and Temperature for R = Bn and R
Letter
1
2
either base gave only ∼7% yield and was not further optimized.
With primary alkyl nitriles, the base-controlled diastereoselec-
tivity was observed across a range of substrates. There was a
marginal gain in diastereoselectivity with aromatic imines if
there was an ortho-substituent (compare entries 1, 2, 5−11 with
entries 3 and 4). The beneficial effect of ortho-substituents was
only replicated on the nitrile partner 3-(2-chlorophenyl)-
propylnitrile for the syn-diastereoisomer generated with
LHMDS (entries 21, 90% yield). Other nitriles of simple
alkyl derivatives gave no clear trend in diastereoselectivities but
gave some examples of good selectivity (entries 13, 14, and 21)
and many others where the major isomers could be isolated
diastereomerically pure. Alkyl imines derived from cyclo-
hexanecarboxaldehyde and n-hexanal both gave the same
diastereoselection with each base favoring the anti-diaster-
eoisomer (entries 30−33). Other aromatic imines gave
moderate to good yields of pure major diastereoisomers
(entries 24−29).
a
=
2-BrC H4
6
b
c
entry
base
nBuLi
nBuLi
conditions
−78 °C, 1 h
−78 °C, 1 h; rt, 5 min
−78 °C, 1 h
−78 °C, 1 h; rt, 5 min
−78 °C, 1 h
conv (%)
anti/syn
1
2
3
4
5
6
90
80
85:15
15:85
10:90
10:90
80:20
15:85
LHMDS
LHMDS
LDA
100
80
90
LDA
−78 °C, 1 h; rt, 5 min
80
a
All reactions were carried out on a 0.5 mmol scale, nitrile (1.0 equiv),
base (1.1 equiv in hexane), imine (1.1 equiv) in THF (6 mL) at −78
b
1
c
1
°
C for 1 h. By H NMR. Determined by comparison of the H NMR
signals for the CH CHCN protons (∼2.5−3.5 ppm) of the crude
2
reaction mixture (to nearest 5).
noted before in the reaction of α-amidoalkylphenyl sulfones
(
generating the corresponding N-Boc-imine in situ) with
lithiated nitriles and gave variable yields and diastereoselectiv-
ities of the anti-β-aminonitrile.
To elucidate the mechanism for this diastereoselectivity, the
14
reaction (Scheme 1, R1 = Bn and R2 = o-BrC
H ) was
6 4
We were intrigued by the fact that the judicious choice of
base could prepare either anti- or syn-β-aminonitriles. A survey
of the reaction of different lithiated alkane nitriles (from 1),
performed at −78 °C at varying reaction times and the crude
material analyzed by H NMR (Table 3). The results suggest
that the initial 1,2-addition favors the anti-diastereoisomer in an
85:15 ratio after 1 min at −78 °C with both BuLi and
LHMDS. The initial anti-selectivity for LHMDS was quickly
1
n
n
generated with either BuLi or LHMDS at −78 °C for 1 h, with
a variety of imines 2 was performed (Scheme 1, Table 2). A
reaction with a secondary alkyl nitrile isobutyronitrile with
eroded over time (<30 min), and as the conversion increased
a
Table 2. Scope of Diastereoselectivity
a
All reactions were carried out on a 0.5 mmol scale, nitrile (1.0 equiv), base (1.1 equiv in hexane), imine (1.1 equiv) in THF (6 mL) at −78 °C for 1
b
1
1
2
1
h. Assigned by H NMR comparison with compounds anti-3 and syn-3 R = Bn, R = 2-BrC H and determined by comparison of the H NMR
signals for the CH CHCN protons (∼2.5−3.5 ppm) of the crude reaction mixture (to nearest 5). Isolated yield of pure major diastereoisomer. −
7
6
5
c
d
2
e
1
f
g
8 °C, 4.5 h. Conversion by H NMR. Yield of inseparable diastereomeric mixture. Yield of minor syn-diastereoisomer in parentheses.
B
DOI: 10.1021/acs.orglett.7b00679
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 3. Effect of Base and Temperature on
Diastereoselectivity
change in diastereoselectivity. This enables the conversion of
anti-β-aminonitriles to syn-β-aminonitriles.
a
The pK of HMDS (26) is similar to that of HCCN (22−
nBuLi
a
LHMDS
13
3
1). We propose at −78 °C that equilibration occurs by
b
c
b
c
time at −78 °C
anti/syn
conv (%)
anti/syn
conv (%)
deprotonation and reprotonation of the HCCN of 3 by HMDS
(Scheme 3). Using LHMDS as base, the original kinetic anti-3-
rich mixture is reprotonated by HMDS, and the regenerated
LHMDS equilibrates the HCCN center to the thermodynamic
1
5
3
1
3
6
min
min
0 min
h
85:15
85:15
85:15
85:15
50:50
25:75
80
90
85:15
70:30
25:75
10:90
10:90
10:90
30
50
90
90
n
90
100
100
100
mixture with syn-3 as the major diastereoisomer. With BuLi
h
>95
>95
the initial kinetic anti-3 rich mixture is stable at −78 °C for at
least 1 h. Additional proof was provided by a series of other
experiments (R1 = Bn, R2 = 2-BrC H ). Addition of an
h
a
All reactions were carried out on a 0.5 mmol scale, nitrile (1.0 equiv),
base (1.1 equiv in hexane), imine (1.1 equiv) in THF (6 mL) at −78
C for 1 h. Determined by comparison of the H NMR signals for the
CH CHCN protons (∼2.5−3.5 ppm) of the crude reaction mixture
to nearest 5). By H NMR.
6
5
n
equimolar quantity of HMDS to an BuLi deprotonation
directly before or after addition of imine favored the
thermodynamic syn-3 (25:75). Addition of an equimolar
b
1
°
2
c
1
(
n
quantity of BuLi to a LHMDS deprotonation followed by
imine favored the kinetic anti-3 (75:25). Allowing a LHMDS
experiment to stir for 30 min after addition of imine and then
adding an equimolar quantity of BuLi favored the thermody-
with time, the selectivity for the syn-diastereoisomer also
n
n
increased to a maximum 10:90 (∼60 min). For BuLi, the initial
namic syn-3 (20:80), implying that the equilibration was
and rapid anti-selectivity (85:15) was maintained and did not
n
interrupted by irreversible deprotonation with BuLi. Circum-
n
start to erode until after 1 h. After 6 h, the BuLi reaction was
stantially commercial LHMDS led to the formation of syn-3
syn-selective, but only to the extent of 25:75. These results
suggest that the anti-diastereoisomer is the kinetic product and
the syn-diastereoisomer is the thermodynamic product. The
reason why the syn-diastereoisomer is the thermodynamically
more stable product is not obvious to us.
24
more efficiently than freshly prepared LHMDS. This
explanation (Scheme 3) satisfactorily accounts for the
diastereoselectivities of all the aromatic imines investigated,
but an anomaly is the two examples of alkyl imines derived
from cyclohexane carboxaldehyde and n-hexanal (entries 30−
To probe the mechanism of isomerization a crossover
3
3). An equal amount of anti-diastereoisomer prevailed with
1
2
experiment was conducted. The initial product 3 (R = Bn, R =
-BrC H ), formed in situ from addition to PMP-protected
both bases in each case, suggesting that the initial kinetic
diastereoisomer does not undergo equilibration.
2
6
5
2
imine 2 (R = 2-BrC H ) after 1 h at −78 °C, was treated with
1
6
5
A previous literature report of lithium cyanoate additions to
benzaldehyde imines using LDA suggests some epimerization
via both retro-/readdition and deprotonation/reprotonation of
equiv of the corresponding N-p-ethoxyphenyl-protected
imine 4 and left to stir for a further 1 h at −78 °C (Scheme
1
2
). The crude H NMR did not show any incorporation of the
15
HCCN. The paper also goes on to show that addition of a
variety of alkyl halides gives good to moderate yields of
quaternary nitriles. In our case, it would not be unreasonable to
assume that a small equilibrium exists between aza-anion 3 and
second imine with either base. Repeating the crossover
experiment, but warming to rt for 5 min directly after the
addition of 1 equiv of 4, showed a statistical 50:50 mixture of
both β-aminonitriles, again for both bases. As there is no
crossover at −78 °C, we can conclude that the equilibration of
the LHMDS system at −78 °C is not due to a retro- then
readdition pathway. Crossover was observed at rt; therefore, a
retro-/readdition mechanism can take place on warming, which
accounts for the formation of the thermodynamic syn-
6
(Scheme 3) and that this is only significantly perturbed in the
presence of a suitable reversible proton donor, such as HMDS.
To show the synthetic versatility of the β-aminonitriles, syn-3
was reduced to the corresponding 1,3-diamine with LiAlH4/
H SO in 92% yield and hydrolyzed with basic hydrogen
2
4
n
peroxide to the β-amino acid in 65% yield, with no erosion in
diastereoselectivity (see the Supporting Information).
The judicious choice of either irreversible conditions with
diastereoisomer for BuLi as the reaction warms to rt. A key
control experiment was treatment of anti-3 with LHMDS (1.1
equiv) in the presence of 4 at −78 °C. After 1 h, there was
again no incorporation of imine 4, but the original anti-
diastereoselectivity (85:15) had epimerized to the syn-
diastereoisomer (15:85), within experimental error virtually
identical to the original LHMDS syn-diastereoseletion (10:90).
n
BuLi to give anti-diastereoselectivity or reversible conditions
with LHMDS to give syn-diastereoselectivity provides an
operationally simple method for the isolation of diastereomeri-
cally pure β-aminonitriles derived from substituted primary
acetonitriles and aryl imines. The characterization of the
thermodynamic equilibration process with HMDS offers the
opportunity for the products from enantioselective syntheses
n
Additionally treatment of anti-3 with BuLi at −78 °C for 1 h
led to no change in diastereoeslectivity, and treatment of syn-3
with either nBuLi or LHMDS at −78 °C for 1 h also led to no
Scheme 2. Crossover Experiment
C
DOI: 10.1021/acs.orglett.7b00679
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Proposed Mechanism
that are anti-diastereoselective to be converted into syn-
diastereoisomers. The methodology described provides a direct
method for the stereoselective synthesis of β-aminonitriles,
which will widen their use as chiral building blocks in target
synthesis.
(9) For example: Yadav, J. S.; Reddy, B. V. S.; Parimala, G.; Reddy, P.
V. Synthesis 2002, 2383 and references cited therein.
(
10) Anderson, J. C.; Blake, A. J.; Mills, M.; Ratcliffe, P. D. Org. Lett.
008, 10, 4141.
11) For example, see: Chhiba, V.; Bode, M. L.; Mathiba, K.; Kwezi,
W.; Brady, D. J. Mol. Catal. B: Enzym. 2012, 76, 68.
12) For a review of catalytic cyanoalkylation, see: Lop
C. Angew. Chem., Int. Ed. 2015, 54, 13170.
2
(
(
́
ez, R.; Palomo,
ASSOCIATED CONTENT
Supporting Information
■
*
S
(13) (a) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456. (b) Richard, J.
P.; Williams, G.; Gao, J. J. Am. Chem. Soc. 1999, 121, 715.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b00679.
(
14) Mecozzi, T.; Petrini, M.; Profeta, R. J. Org. Chem. 2001, 66,
264.
15) Viteva, L.; Gospodova, T.; Stefanovsky, Y.; Simova, S.
Tetrahedron 2005, 61, 5855.
16) (a) D’Sa, B. A.; Kisanga, P.; Verkade, J. G. J. Org. Chem. 1998,
3, 3961. (b) Kisanga, P.; McLeod, D.; D’Sa, B.; Verkade, J. G. J. Org.
Chem. 1999, 64, 3090.
17) (a) Poisson, T.; Gembus, V.; Oudeyer, S.; Marsais, F.; Levacher,
8
(
Experimental procedures, characterization data and
NMR spectra for all compounds; representation of X-
ray structure for syn-3 (PDF)
(
6
Crystallographic data for syn-3 (CIF)
(
V. J. Org. Chem. 2009, 74, 3516. (b) Hyodo, K.; Kondo, M.;
Funahashi, Y.; Nakamura, S. Chem. - Eur. J. 2013, 19, 4128.
AUTHOR INFORMATION
Corresponding Authors
■
*
*
(
18) (a) Aydin, J.; Conrad, C. S.; Szabo,
́
K. J. Org. Lett. 2008, 10,
5
175. (b) Hyodo, K.; Nakamura, S.; Tsuji, K.; Ogawa, T.; Funahashi,
E-mail: j.c.anderson@ucl.ac.uk.
E-mail: gjt1@soton.ac.uk for crystallographic results.
ORCID
Y.; Shibata, N. Adv. Synth. Catal. 2011, 353, 3385. (c) Zhao, J.; Liu, X.;
Luo, W.; Xie, M.; Lin, L.; Feng, X. Angew. Chem., Int. Ed. 2013, 52,
3473.
(19) Tian, M.; Yan, M.; Baran, P. S. J. Am. Chem. Soc. 2016, 138,
James C. Anderson: 0000-0001-8120-4125
1
4234.
Present Addresses
(20) Robak, M. T.; Herbage, M. A.; Ellman, J. A. Chem. Rev. 2010,
†(
110, 3600.
21) The terms anti and syn are defined in this paper according to
I.B.C., S.C.) GlaxoSmithKline, Medicines Research Centre,
(
Stevenage SG1 2NY, UK.
‡
that proposed by Masamune. Masamune, S.; Ali, S. A.; Snitman, D. L.;
Garvey, D. S. Angew. Chem., Int. Ed. Engl. 1980, 19, 557.
(G.J.T.) National Crystallography Service, School of Chem-
istry, University of Southampton, Southampton SO17 1BJ, UK.
(22) (a) Anderson, J. C.; Blake, A. J.; Koovits, P. J.; Stepney, G. J. J.
Notes
Org. Chem. 2012, 77, 4711. (b) Anderson, J. C.; Horsfall, L. R.;
Kalogirou, A. S.; Mills, M. R.; Stepney, G. R.; Tizzard, G. J. J. Org.
Chem. 2012, 77, 6186. (c) Anderson, J. C.; Noble, A.; Tocher, D. A. J.
Org. Chem. 2012, 77, 6703. (d) Anderson, J. C.; Koovits, P. J. Chem.
Sci. 2013, 4, 2897. (e) Anderson, J. C.; Kalogirou, A. S.; Porter, M. J.;
Tizzard, G. J. Beilstein J. Org. Chem. 2013, 9, 1737. (f) Anderson, J. C.;
Kalogirou, A. S.; Tizzard, G. J. Tetrahedron 2014, 70, 9337.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was supported by UCL the EPSRC and GSK (EP/
F068344/2).
■
(g) Anderson, J. C.; Campbell, I. B.; Campos, S.; Shannon, J.;
Tocher, D. A. Org. Biomol. Chem. 2015, 13, 170. (h) Anderson, J. C.;
Barham, J. P.; Rundell, C. D. Org. Lett. 2015, 17, 4090. (i) Anderson, J.
C.; Rundell, C. D. Synlett 2016, 27, 41. (j) Anderson, J. C.; Campbell,
I. B.; Campos, S.; Reid, I. H.; Rundell, C. D.; Shannon, J.; Tizzard, G.
J. Org. Biomol. Chem. 2016, 14, 8270.
REFERENCES
■
(
1) (a) Chemistry of the Cyano Group; Rappoport, Z., Patai, S., Eds.;
Wiley: London, 1970. (b) Fatiadi, A. J. In The Chemistry of Triple
Bonded Functional Groups, Part 2; Rappoport, Z., Patai, S., Eds.; Wiley:
London, 1983; Supplement C, pp 1057−1303.
(23) Unambiguous assignment of relative stereochemistry was
(
2) Fleming, F. F.; Yao, L.; Ravikumar, P. C.; Funk, L.; Shook, B. C. J.
Med. Chem. 2010, 53, 7902.
3) (a) Arseniyadis, S.; Kyler, K. S.; Watt, D. S. Org. React. 1984, 31,
. (b) Fleming, F. F. Nat. Prod. Rep. 1999, 16, 597.
4) Naruse, N.; Yamamoto, S.; Yamamoto, H.; Kondo, S.; Masuyoshi,
S.; Numata, K.-I.; Fukagawa, Y.; Oki, T. J. Antibiot. 1993, 46, 685.
5) Zeches, M.; Ravao, T.; Richard, B.; Massiot, G.; Le Men-Olivier,
L.; Verpoorte, R. J. Nat. Prod. 1987, 50, 714.
6) Mesa, R. A.; Yasothan, U.; Kirkpatrick, P. Nat. Rev. Drug Discovery
012, 11, 103.
7) For a recent review, see: Ji, X.; Huang, H. Org. Biomol. Chem.
016, 14, 10557.
8) For example: Saikia, M.; Deka, D. C.; Karmakar, S. Curr.
Microwave Chem. 2016, 3, 47 and references cited therein.
achieved through single-crystal X-ray crystallography of the prevailing
syn-diastereoisomer (syn-3) from a reaction using LHMDS. See the SI
and CCDC 1531946.
(
1
(
(24) Possible quenching of LHMDS by adventitious water over time
during use could increase the concentration of HMDS.
(
(
2
(
2
(
D
DOI: 10.1021/acs.orglett.7b00679
Org. Lett. XXXX, XXX, XXX−XXX