One-pot synthesis of enantiomerically pure N-protected allylic amines from N-protected α-amino esters

Article abstract of DOI:10.3762/bjoc.12.94

An improved protocol for the synthesis of enantiomerically pure allylic amines is reported. N-Protected α-amino esters derived from natural amino acids were submitted to a one-pot tandem reduction-olefination process. The sequential reduction with DIBAL-H

Full text of DOI:10.3762/bjoc.12.94

One-pot synthesis of enantiomerically pure N-protected
allylic amines from N-protected α-amino esters
Gastón Silveira-Dorta, Sergio J. Álvarez-Méndez, Víctor S. Martín and José M. Padrón*
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2016, 12, 957–962.
Instituto Universitario de Bio-Orgánica “Antonio González”
(IUBO-AG), Centro de Investigaciones Biomédicas de Canarias
(CIBICAN), Universidad de La Laguna. C/ Astrofísico Francisco
Sánchez 2, 38206, La Laguna, Spain
doi:10.3762/bjoc.12.94
Received: 20 February 2016
Accepted: 25 April 2016
Published: 12 May 2016
Email:
José M. Padrón* - jmpadron@ull.es
Associate Editor: B. Stoltz
* Corresponding author
© 2016 Silveira-Dorta et al; licensee Beilstein-Institut.
License and terms: see end of document.
Keywords:
amino acids; olefination; protecting group free; synthetic methods;
Wittig reactions
Abstract
An improved protocol for the synthesis of enantiomerically pure allylic amines is reported. N-Protected α-amino esters derived
from natural amino acids were submitted to a one-pot tandem reduction–olefination process. The sequential reduction with
DIBAL-H at −78 °C and subsequent in situ addition of organophosphorus reagents yielded the corresponding allylic amines with-
out the need to isolate the intermediate aldehyde. This circumvents the problem of instability of the aldehydes. The method toler-
ates well both Wittig and Horner–Wadsworth–Emmons organophosphorus reagents. A better Z-(dia)stereoselectivity was observed
when compared to the previous one-pot method. The (dia)stereoselectivity of the process was affected neither by the reaction sol-
vent nor by the amount of DIBAL-H employed. The method is compatible with the presence of free hydroxy groups as shown with
serine and threonine derivatives.
Introduction
Allylic amines have received significant attention because they been largely used in organic chemistry because they are widely
represent a common scaffold in diverse biologically relevant accessible and they are relatively inexpensive, even on a bulk
compounds and natural products [1]. In addition, allylic amines scale.
serve as versatile structural building block units for the synthe-
sis of various functionalized organic compounds, playing an im- The most classical and conventional protocol to obtain allylic
portant role as intermediates in asymmetric synthesis [2]. amines from N-protected α-amino esters requires three chemi-
Furthermore, asymmetric allylic amines can be obtained in cal steps, i.e. reduction, oxidation and olefination (Figure 1)
enantiomerically pure form from conveniently functionalized [9,10]. When the benzyl group is used as protecting group for
commercially available amino acids [3-8]. Amino acids have the nitrogen functionality, the method represents a variation of
957

Beilstein J. Org. Chem. 2016, 12, 957–962.
Figure 1: Strategies for the synthesis of N-protected allylic amines. [Red], reduction; [Ox], oxidation; [Olef], olefination; [Pg], protecting group.
the well-known Reetz protocol [11]. In the particular case of the scope of the aforementioned one-pot procedure to the syn-
serine and threonine, additional O-protection and O-deprotec- thesis of enantiomerically pure allylic amines. It has been re-
tion steps are required (Figure 1). Thus, Garner’s aldehyde ported earlier the one-pot preparation of N-protected allylic
[12,13] or N- and O-protected amino ester derivatives have been amines via tandem DIBAL-H reduction–Wittig olefination of
required [14].
N-protected α-amino esters [17]. However, that preliminary
study was limited to the use of phosphonium ylide reagents and
All those methodologies involve long reaction sequences, in- commonly t-Boc (iBoc and Ac were used once) as N-protecting
cluding protection–deprotection and/or oxidation–reduction group. To the best of our knowledge, no further studies on the
processes that are usually detrimental for the overall yield. reaction conditions have been carried out. Instead, the method
Moreover, the classical three-step approach is highly undesir- was applied to N-Ac aspartic, and N-Ac and N-Boc glutamic
able in some cases not only because it generates more waste acid dialkyl esters, this time using a stabilized phosphonate
(more reactions are needed), but more importantly, because it ester [18]. This strategy was used later on in the synthesis of
implies the isolation of the aldehydes that could be unstable to aminopeptidase A inhibitors [19]. Similarly, N-methylproline
manipulate.
methyl ester was reacted in a similar one-pot fashion during the
synthesis of nine-membered ring lactams [20]. Finally, a one-
As part of our interest in the synthesis of nitrogen-containing pot reduction-olefination involving an α-amino β-hydroxy ester
bioactive molecules [15], we developed a simplified version of and a phosphonium salt, both bearing free hydroxy groups, has
the Reetz protocol [11] for the synthesis of enantiomerically been used in the synthesis of (−)-α-conhydrine [21].
pure anti-β-amino alcohols [16]. The process circumvents the
problem of the instability of the aldehydes. It comprises the In this work, our aim was to study in more detail this one-pot
one-pot sequential reduction to aldehyde with DIBAL-H at strategy. Similarly to our one-pot procedure to obtain enantio-
−78 °C and subsequent in situ addition of Grignard reagents. merically pure anti-β-amino alcohols, we selected N,N-dibenzyl
Remarkably, our method is friendly with serine and threonine amino esters as starting material. We also investigated whether
derivatives without the requisite to protect the β-hydroxy group. the method we developed could be applied to serine and threo-
With this tool in hand, we thought on the possibility to extend nine derivatives without protection of the hydroxy groups.
958

Beilstein J. Org. Chem. 2016, 12, 957–962.
Results and Discussion
Table 1: Screening of the reaction conditions for the one-pot tandem
reduction–olefination.
The experimental procedure for the tandem reduction–Wittig
olefination synthesis of allylic amines reported the use of tolu-
ene as solvent for the reduction step and THF as solvent to
prepare the phosphonium ylide [17]. Consequently, the Wittig
olefination takes place in a 2:1 (toluene/THF) solvent mixture.
In the one-pot synthesis of anti-β-amino alcohols, the best
results were achieved when the reactions were run in Et2O [16].
A fine tuning of the reducing agent was also necessary to obtain
the desired products. Our first goal was to study the outcome of
the reaction in terms of solvent and the amount of the reducing
agent. When considering solvents for the process, we selected
Et2O, THF and toluene. The initial amount of DIBAL-H was
chosen on the basis of our one-pot procedure for the synthesis
of anti-β-amino alcohols and set at 1.4 equiv.
Entry
Solvent
DIBAL-H (equiv) E:Z
ratioa
Yield (%)
1
2
3
4
5
6
7
Et2O
1.4
1
>20:1
>20:1
5b
21
–b
20
–b
71
40
Et2O
THF
1.4
1
THF
>20:1
toluene
toluene
toluene
1.4
1
>20:1
>20:1
0.5
As a model to study the reaction conditions we used as starting
material (S)-methyl 2-(dibenzylamino)propanoate (1) [22]. In
order to avoid the stereochemical drawback of the Wittig olefi-
nation (i.e., mixture of E and Z isomers), we selected the stabi-
lized ylide ethyl 2-(triphenylphosphoranylidene)acetate, which
aThe E/Z ratio was determined by 1H NMR analysis of the crude prod-
uct 2a. bOver reduction to alcohol.
gives the E isomer [23]. The results are summarized in Table 1. Once established the reaction conditions, we next studied the
An excess of DIBAL-H had a negative impact in the yield scope and limitations of the one-pot protocol using a small
leading to over reduction of the ester to the alcohol (Table 1, subset of assorted Wittig and Horner–Wadsworth–Emmons
entries 1, 3 and 5). When considering the solvent, toluene gave (HWE) organophosphorus reagents (Table 2). When consid-
the best results (Table 1, entry 6). In addition, only the E isomer ering phosphonium ylides (Table 2, entries 1 and 2), better
of 2a was obtained. The (dia)stereoselectivity (within limits of Z-(dia)stereoselectivity was observed when compared to the
1H NMR detection in the crude reaction mixture) of the process previous one-pot method. The olefination with the semistable
was affected neither by the reaction solvent nor by the amount ylide obtained by the addition of KN(TMS)2 to benzyltriphenyl-
of DIBAL-H employed. It is known from the literature that phosphonium bromide yielded 2b as an inseparable mixture of
some racemization of enantiomerically pure aldehydes occurs E and Z isomers in almost equal amounts. This result represents
during the DIBAL-H treatment. Although we have demon- an improvement in selectivity toward the Z isomer when com-
strated earlier that no loss of enantiomeric purity was observed pared to the 5–7:1 E:Z ratio reported [17]. The preparation of
in the synthesis of anti-β-amino alcohols [16], we submitted (E)-2b has been reported earlier [6] but this is the first time it is
commercially available DL-alanine to the aforementioned one- described the synthesis of its (dia)stereoisomer (Z)-2b. The
pot procedure to give rac-2a. Both rac-2a and 2a were analyzed olefination with the non-stabilized ylide of pentadecyltriphenyl-
by chiral HPLC. The analysis confirmed that the enantiomeric phosphonium bromide led exclusively to (Z)-2c.
purity was not affected by the process.
Likewise, the results obtained for (E)-2a, the use of stabilized
The literature procedure reported the use of 2 equiv of ylides (Table 2, entries 3 and 4) provided the expected E
DIBAL-H, whilst we found that only one equivalent is enough alkenes. In the particular case of the α-cyano phosphorane
to avoid the excess of the reducing reagent. We hypothesized reagent, complete stereoselectivity was not achieved (E:Z ratio
that compound 1 is half reduced by DIBAL-H forming the of 5:1) as it has been previously observed in similar reactions
appropriate aluminoxy acetal, which is expected to be stable [26]. Fortunately, both products were easily separated by
enough at temperatures as low as −78 °C. This is partly corrob- column chromatography.
orated by the decomposition of the aluminoxy acetal to the over
reduced alcohol when 1.4 equiv of DIBAL-H were used Finally, HWE reagent led to the preparation of (E)-2f as sole
(Table 1, entries 1, 3 and 5) [24]. The presence of an alumi- isomer in 68% yield (Table 2, entry 5), whilst the Still–Gennari
noxy acetal is also supported when THF is used, although the variant [27] gave the mixture of (E)-2g and (Z)-2g isomers
yields are low, probably due to the destabilization of the alumi- (Table 2, entry 6). At this point, we wondered if our methodolo-
noxy acetal (Table 1, entry 4) [25].
gy for the synthesis of allylic amines was compatible also
959

Beilstein J. Org. Chem. 2016, 12, 957–962.
Table 2: Influence of the organophosphorus reagent in the outcome of the one-pot tandem reduction–olefination.
Entry
1
Organophosphorus reagent
Ph3P=CHPh
Product
E:Z ratioa
Yield (%)
60
1:1.3
2b
2c
2
3
Ph3P=CH(CH2)13Me
Ph3P=CHCN
>1:20
5:1
40
72
2d
2e
4
Ph3P=C(Me)CO2Et
>20:1
71
5
6
(MeO)2P(O)CH2CO2Me
>20:1
1:1.6
68
78
2f
(CF3CH2O)2P(O)CH2CO2Et
2g
aThe E/Z ratio was determined by 1H NMR analysis of the crude product.
with free hydroxy groups present in the substrate, as it has
been shown earlier for a N-Boc protected amino hydroxy
ester [21]. To corroborate this idea, N,N-dibenzylamino benzyl
ester of L-serine (3) was submitted to the one-pot tandem reduc-
tion–olefination procedure described above. Disappointingly,
the product (E)-4a was obtained in low yield although in excel-
lent (dia)stereoselectivity. In the synthesis of anti-2-amino-1,3-
diols, we reported earlier that the addition of DIBAL-H must be
done necessarily in two portions [16]. Thus, fine-tuning of the
reduction conditions was required in order to improve the yield,
as shown in Table 3.
Table 3: Screening of the reaction conditions for the one-pot tandem
reduction–olefination of free hydroxyserine derivatives.
Entry Reduction
time (h)
DIBAL-H
(equiv)
E:Z
ratioa
Yield (%)
1
2
3
4
2
2
2
1
>20:1
>20:1
>20:1
>20:1
7
1.4
2
36
58
65
i) 1
ii) 2
i) 1.4b
ii) 0.5
The best results were obtained when the reducing agent was
added in two portions, with an interval of one hour between the
two (Table 3, entry 4). It should be taken into account that the
DIBAL-H addition must be done necessarily in two portions
(Table 3, entry 3 vs 4) in order to increase the yield of the reac-
aThe E/Z ratio was determined by 1H NMR analysis of the crude prod-
uct. bDIBAL-H was added in two portions.
tion. We speculate that the free hydroxy group coordinates to As shown in Table 4, this modification of the one-pot protocol
the DIBAL-H, making compulsory the additional amount of is compatible with the use of HWE organophosphorus reagents
reducing agent.
(Table 4, entry 1). However, the reaction with the non-stabi-
960

Beilstein J. Org. Chem. 2016, 12, 957–962.
Table 4: One-pot tandem reduction–olefination of free hydroxyserine and threonine derivatives.
Entry
1
Organophosphorus reagent
(MeO)2P(O)CH2CO2Me
Product
E:Z ratioa
Yield (%)
50
>20:1
4b
4c
2
3
Ph3P=CH(CH2)13Me
Ph3P=CHCO2Et
37:63
>20:1
60b
40
6a
4
Ph3P=C(Me)CO2Et
>20:1
39
6b
aThe E/Z ratio was determined by 1H NMR analysis of the crude product. bInseparable E/Z mixture.
lized ylide of pentadecyltriphenylphosphonium bromide application of this strategy are currently underway in our labo-
resulted in loss of stereoselectivity (Table 4, entry 2). The ratory.
method can be applied also to the N,N-dibenzylamino benzyl
ester of L-threonine (5) albeit lower yields are obtained. It is
Supporting Information
remarkable that this protocol avoids using O-protecting groups
or the Garner aldehyde which have been used extensively in the
Supporting Information File 1
synthesis of related compounds.
General procedures, analytical data and spectra of all
compounds, methods for conversion.
Conclusion
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-12-94-S1.pdf]
In summary, this simple protocol described herein enables a
rapid access to a number of useful enantiopure allylic amines
from readily available amino acids. Optically active amino ester
derivatives can be transformed into allylic amines by a tandem
reduction–olefination procedure. The process avoids the isola-
tion of the intermediate aldehyde, which makes it an attractive
option for unstable aldehydes. The selectivity in the olefination
step was considerably higher than those previously reported
using batch methods with related reagents. Our technique is also
compatible with free hydroxy groups displayed in the substrate,
allowing the synthesis of new products unreported to date.
Further investigations into the reaction mechanism, scope, and
Acknowledgements
Co-financed by the EU Research Potential (FP7-REGPOT-
2012-CT2012-31637-IMBRAIN), the European Regional De-
velopment Fund (FEDER), the Spanish MINECO (CTQ2011-
28417-C02-01 and CTQ2014-56362-C2-1-P). The authors
thank Dr. Romen Carrillo for helpful discussions. G.S.-D thanks
the EU Social Found (FSE) and the Canary Islands ACIISI for a
predoctoral grant. S.J.A.-M. thanks the Spanish Ministerio de
Educación y Ciencia (MEC) for an FPU grant.
961

Beilstein J. Org. Chem. 2016, 12, 957–962.
27.Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405–4408.
References
1. Ye, Z.; Brust, T. F.; Watts, V. J.; Dai, M. Org. Lett. 2015, 17, 892–895.
doi:10.1016/S0040-4039(00)85909-2
doi:10.1021/ol503748t
2. Trost, B. M.; Zhang, T.; Sieber, J. D. Chem. Sci. 2010, 1, 427–440.
doi:10.1039/c0sc00234h
License and Terms
3. Yang, H.; Sun, M.; Zhao, S.; Zhu, M.; Xie, Y.; Niu, C.; Li, C.
J. Org. Chem. 2013, 78, 339–346. doi:10.1021/jo302203g
4. Stanway, S. J.; Thomas, E. J. Tetrahedron 2012, 68, 5998–6009.
doi:10.1016/j.tet.2012.05.032
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/2.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
5. Hanessian, S.; Maji, D. K.; Govindan, S.; Matera, R.;
Tintelnot-Blomley, M. J. Org. Chem. 2010, 75, 2861–2876.
doi:10.1021/jo100017t
6. Concellón, J. M.; Méjica, C. Eur. J. Org. Chem. 2007, 5250–5255.
doi:10.1002/ejoc.200700515
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
7. Concellón, J. M.; Suárez, J. R.; del Solar, V. Org. Lett. 2006, 8,
349–351. doi:10.1021/ol0529602
(http://www.beilstein-journals.org/bjoc)
8. Baeza, A.; Casas, J.; Nájera, C.; Sansano, J. M. J. Org. Chem. 2006,
71, 3837–3848. doi:10.1021/jo060239j
9. Kumar, M. G.; Mali, S. M.; Gopi, H. N. Org. Biomol. Chem. 2013, 11,
803–813. doi:10.1039/C2OB27070F
The definitive version of this article is the electronic one
which can be found at:
10.Davies, S. G.; Lee, J. A.; Roberts, P. M.; Thomson, J. E.; Yin, J.
Org. Lett. 2012, 14, 218–221. doi:10.1021/ol203011u
11.Reetz, M. T.; Drewes, M. W.; Schwickardi, R. Org. Synth. 1999, 76,
110. doi:10.15227/orgsyn.076.0110
doi:10.3762/bjoc.12.94
12.Barco, A.; Benetti, S.; Spalluto, G.; Casolari, A.; Pollini, G. P.;
Zanirato, V. J. Org. Chem. 1992, 57, 6279–6286.
doi:10.1021/jo00049a040
13.Imashiro, R.; Sakurai, O.; Yamashita, T.; Horikawa, H. Tetrahedron
1998, 54, 10657–10670. doi:10.1016/S0040-4020(98)00615-2
14.Kozikowski, A. P.; Nieduzak, T. R.; Konoike, T.; Springer, J. P.
J. Am. Chem. Soc. 1987, 109, 5167–5175. doi:10.1021/ja00251a021
15.Silveira-Dorta, G.; Sousa, I. J.; Fernandes, M. X.; Martín, V. S.;
Padrón, J. M. Eur. J. Med. Chem. 2015, 96, 308–317.
doi:10.1016/j.ejmech.2015.03.046
And references cited therein.
16.Silveira-Dorta, G.; Donadel, O. J.; Martín, V. S.; Padrón, J. M.
J. Org. Chem. 2014, 79, 6775–6782. doi:10.1021/jo500481j
17.Wei, Z.-Y.; Knaus, E. E. Synthesis 1994, 1463–1466.
doi:10.1055/s-1994-25714
18.Wei, Z.-Y.; Knaus, E. E. Tetrahedron Lett. 1994, 35, 2305–2308.
doi:10.1016/0040-4039(94)85205-7
19.David, C.; Bischoff, L.; Meudal, H.; Mothé, A.; De Mota, N.;
DaNascimento, S.; Llorens-Cortes, C.; Fournié-Zaluski, M.-C.;
Roques, B. P. J. Med. Chem. 1999, 42, 5197–5211.
doi:10.1021/jm9903040
20.Diederich, M.; Nubbemeyer, U. Angew. Chem., Int. Ed. Engl. 1995, 34,
1026–1028. doi:10.1002/anie.199510261
21.Kandula, S. V.; Kumar, P. Tetrahedron Lett. 2003, 44, 1957–1958.
doi:10.1016/S0040-4039(03)00031-5
22.Rinehart, K. L.; Fregeau, N. L.; Warwick, R. A.; Garcia Gravalos, D.;
Avila, J.; Faircloth, G. T. WO Patent WO 9952521A, Oct 21, 1999.
Chem. Abstr. 1999, 131, 295576.
23.Vedejs, E.; Peterson, M. J. Top. Stereochem. 1994, 21, 1–157.
doi:10.1002/9780470147306.ch1
24.Yamazaki, T.; Kobayashi, R.; Kitazume, T.; Kubota, T. J. Org. Chem.
2006, 71, 2499–2502. doi:10.1021/jo052434o
25.Polt, R.; Peterson, M. A.; DeYoung, L. J. Org. Chem. 1992, 57,
5469–5480. doi:10.1021/jo00046a032
26.Tomioka, T.; Yabe, Y.; Takahashi, T.; Simmons, T. K. J. Org. Chem.
2011, 76, 4669–4674. doi:10.1021/jo200019j
962