Diastereoselective scalable synthesis of 2,6-trans-Piperidines using an aza-Michael reaction

Article abstract of DOI:10.1016/j.tetlet.2018.12.061

Highly efficient substrate and reagent controlled stereoselective synthesis of 2,6-trans-piperidine derivative (1) using an aza-Michael reaction is reported. This method was utilized to synthesize a variety of trans-piperidines on hundred-gram scales.

Full text of DOI:10.1016/j.tetlet.2018.12.061

Tetrahedron Letters xxx (xxxx) xxx
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Diastereoselective scalable synthesis of 2,6-trans-Piperidines using an
aza-Michael reaction
Somenath Chowdhury, Jianmin Mao, Jacek Martynow, He Zhao, Erin Duffy, Yusheng Wu, Vinay Thakur,
⇑
Gopal Sirasani, Yuanqing Tang, Florine Collin, Sandra Sinishtaj, Ashoke Bhattacharjee
Department of Chemistry, Melinta Therapeutics Inc., 300 George Street, New Haven, CT 06511, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Highly efficient substrate and reagent controlled stereoselective synthesis of 2,6-trans-piperidine deriva-
tive (1) using an aza-Michael reaction is reported. This method was utilized to synthesize a variety of
trans-piperidines on hundred-gram scales.
Received 31 October 2018
Revised 19 December 2018
Accepted 24 December 2018
Available online xxxx
Ó 2018 Elsevier Ltd. All rights reserved.
Keywords:
trans-piperidine
aza-Michael cyclization
Diastereoselective
Base catalyzed
tert-butanesulfinamide
Piperidines are extremely important core motif in a wide vari-
ety of natural products and active pharmaceutical ingredients [1].
Due to its intrinsic conformational preferences 2,6-disubstitued
piperidine derivatives are broadly used in chemical synthesis to
generate various compounds with therapeutic potential [2]. The
nucleophilic and basic nitrogen in the piperidine framework offers
numerous synthetic opportunities to tune SAR and explore a vari-
ety of biological and physicochemical properties. Therefore, under-
standing the combined structural and functional properties of
piperidine derivatives plays a critical role in de novo drug design
and development [3]. Although thermodynamically stable cis-2,6-
disubstituted piperidines can be prepared relatively easily, con-
structing multigram-scale chiral 2,6-trans-piperidines is a daunt-
ing challenge. Herein, we report a highly efficient substrate and
reagent controlled diastereoselective synthesis of 2,6-trans-piperi-
dine derivatives 1 from easily available starting materials through
an intramolecular aza-Michael reaction.
attention where the reaction has been catalyzed either by triflic
acid or by a Pd (II) species. For our ongoing drug discovery
program, we needed to prepare a large stock of compound 1 or an
analog that contains an easily removable protecting group on the
piperidine nitrogen to explore a variety of SAR. We planned to use
an
a,b-unsaturated ester as a starting material which eventually,
after cyclization, can be further manipulated to a variety of
functional groups to improve or tune properties of biologically
active compounds. Although
a,b-unsaturated esters are known to
be poor Michael acceptors [4e], we wanted to use this substrate so
that we can have better control in trans/cis selectivity during
piperidine construction. Our investigation started from a suitable
intermediate as shown in Scheme 1. The condensation of 4-bromo
benzaldehyde with (S)-tert-butanesulfinamide gave sulfinimine
5a, which was then treated with pent-4-enylmagnesium bromide
at low temperature. Although the overall yield of this reaction was
very good, we noticed moderate stereoselectivity where desired
6a is the major product. Repeated attempts (Scheme 2) to achieve
better selectivity failed. All the sulfinimines (5a–c) yielded similar
diastereomeric mixtures of 6/7.
At this point 6a was purified from its diastereomer 7a by tradi-
tional chromatography and its absolute stereochemistry was con-
firmed by NMR analysis of its Mosher amide. 6a was converted
to 8 (Scheme 1) which was reacted with ethyl acrylate in presence
of Hoveyda-Grubbs catalyst to furnish intermediate 9. The alkene 9
was a suitable precursor for aza-Michael reaction. A recent litera-
ture described the acid catalyzed aza-Michael cyclization using
Although a number of synthetic methods are available to con-
struct a wide variety of piperidine analogs, intramolecular Michael
reaction is still the most popular route to generate 2,6-disubstitued
analogs in a chiral fashion [4]. A plethora of organic and inorganic
catalysts are known to promote this reaction, but none of these are
applicable in large-scale synthesis. A recent example [4e] of a fully
functionalized
a,b-unsaturated carbonyl compound caught our
⇑
Corresponding author.
E-mail address: bashoke@hotmail.com (A. Bhattacharjee).
https://doi.org/10.1016/j.tetlet.2018.12.061
040-4039/Ó 2018 Elsevier Ltd. All rights reserved.
0
Please cite this article as: S. Chowdhury, J. Mao, J. Martynow et al., Diastereoselective scalable synthesis of 2,6-trans-Piperidines using an aza-Michael reac-
tion, Tetrahedron Letters, https://doi.org/10.1016/j.tetlet.2018.12.061

2
S. Chowdhury et al. / Tetrahedron Letters xxx (xxxx) xxx
Table 1
Attempted aza-Michael reaction of Cbz protected amine 9.
Entry Conditiona,b
Resultc
1
2
3
4
Triflic acid (0.2 equiv.), CH
up to 5 h.
2
Cl
2
2
, À20 °C, No Reaction
Triflic acid (0.2 equiv.), CH
up to 5 h.
2
Cl
, 25 °C,
No Reaction
No Reaction
Triflic acid (1 equiv.), CH
up to 5 h.
2
Cl
2
2
, À20 °C,
Triflic acid (1 equiv.), CH
to 5 h.
2
Cl
, 25 °C, up Cbz group was partially
cleaved. No cyclization
2 3 2 2
Scheme 1. Initial synthesis plan: (a) (S)-tert-Butanesulfinamide, Cs CO , CH Cl ,
9
7% (b) Pent-4-enylmagnesium bromide, THF, À78 °C, 70% (c) (1) HCl, MeOH. (2)
Na CO , CbzCl, THF/Water, 87%, (d) Ethyl acrylate, Hoveyda-Grubbs-II, CH Cl
0 °C, 95%. Note: Then acid catalyzed cyclisation of 9 (Table 1) followed by
reduction of ester using LAH.
occurred.
2
3
2
2
,
5
Triflic acid (4 equiv.), CH
2
Cl
2
, 25 °C 1 h Cbz group was cleaved at
4
then at 45 °C for 3 h. Cooled down to
room temperature and 3.9 equiv. of
45 °C.
Upon addition of Et N,
3
Et
3
N was added dropwise and stirred
cyclization occurred.
overnight (ꢀ14 h).
Yield: 81%; trans/cis
=
44/56.
6
7
2
1 M HCl/Et O (0.6 mol%)
No reaction
t-BuOK (0.8 equiv.), THF, 25 °C, 30 min. Yield: 46%; trans/cis
60/40
=
a
b
c
Reactions were run in 1 mmol scale.
Volume of the solvent was 5 ml.
Yields were defined as percentage peak area at 214 nm.
Table 2
Base-catalyzed cyclization of 12.
Scheme 2. Pent-4-enyl magnesium bromide addition.
triflic acid (TfOH) as catalyst to prepare trans-2,6-disubstituted
piperidine [4e]. We attempted to use similar conditions to obtain
trans-2,6-disubstituted piperidine but our substrates failed to
yield any product (entry 1–4, Table 1).
Using slightly modified conditions (entry 5, 4 equiv. TfOH and
3
.9 equiv. triethyl amine) resulted in loss of Cbz followed by cycli-
Entry Conditiona,b
Resultc
sation in favor of the cis-isomer. Recently [5] HCl was also used as
catalyst for aza-Michael reactions. However, addition of 1 M HCl in
1
2
3
t-BuOK (1.0 equiv.), THF, 25 °C,
0 min.
Traces of product. Complicated
mixture
3
2
Et O (0.6 mol%) (entry 6, Table 1) did not show any reaction for the
t-BuOK (1.05 equiv.), THF, À75 °C to Traces of product. Complicated
starting material. At this point, we attempted base-catalyzed
cyclization reactions. Use of potassium tert-butoxide (entry 7,
Table 1) improved the selectivity (trans/cis = 60/40) but reduced
the yield to 46%. Since a chiral tert-butyl sulfinyl group [6] is
known to influence the stereochemical outcome of reactions, we
envisioned that use of intermediate 12 (Table 2) instead of 9
0 °C, 1 h.
mixture
No product
Hunig’s base (2.0 equiv.), ACN,
80 °C, 16 h
4
5
6
7
8
TEA (2.0 equiv.), ACN, 80 °C, 16 h
DBU (2.0 equiv.), ACN, 80 °C, 16 h
DBU (4.0 equiv.), DMA, 80 °C, 16 h
No product
Yield: 56% trans/cis = 60/40
Yield: 56% trans/cis = 60/40
Yield: 70%, trans/cis = 60/40
Yield = 85%, trans/cis = 90/10
Cs
Cs
2
CO
3
(5 equiv.), ACN, 80 °C, 16 h
(5 equiv.), DMA, 40 °C, 4 h
2
CO
3
(Scheme 1) as a possible starting material could provide better
a
b
c
trans/cis selectivity. It would also eliminate protection-deprotec-
tion steps.
Reactions were run in 1 mmol scale.
Volume of the solvent was 5 ml.
Yields were defined as percentage peak area at 214 nm.
Pure diastereomer 6a was treated with ethyl acrylate in the
presence of Hoveyda-Grubbs catalyst to afford compound 12 in
excellent yield. Since the tert-butyl sulfinyl group is acid sensitive,
we planned to explore base catalyzed cyclization reactions. The
results are summarized in Table 2.
2 3
Cs CO (5 equiv.), DMA, 40 °C, 4 h (entry 8) provided more than
8
5% yield with good selectivity (trans/cis = 90/10).
At this stage we have solved the problem of making trans-piper-
idine with excellent yield and selectivity. However, the poor selec-
tivity of the first step (Scheme 1, from 5a to 6a) made this route
We tried several conditions (entries 1–6), but failed to obtained
promising results. However, to our surprise, use of Cs
0 °C, 16 h (entry 7) provided a good yield (>70%) with moderate
selectivity, 13/14 (trans/cis = 60:40). After further optimization,
2 3 3
CO , CH CN,
8
Please cite this article as: S. Chowdhury, J. Mao, J. Martynow et al., Diastereoselective scalable synthesis of 2,6-trans-Piperidines using an aza-Michael reac-
tion, Tetrahedron Letters, https://doi.org/10.1016/j.tetlet.2018.12.061

S. Chowdhury et al. / Tetrahedron Letters xxx (xxxx) xxx
3
Scheme 3. Allyl zinc reagent route: (a) allyl bromide, Zn, LiCl, DMF, H
-BBN, Pd(PPh , K PO , ethyl-3-iodoacrylate, 65% (c) Cs CO , DMA, 85% (d) LAH,
complicated mixture.
2
O, 100% (b)
9
)
3 4
3
4
2
3
Fig. 1. Plausible mechanism for base mediated selectivity.
and coupled with 3-iodo acrylate to form 12. The stereochemical
and spectroscopic data of 12 were identical as obtained from 6a.
During the work-up and isolation process of 12, we observed
that the ethyl ester was partially hydrolyzed to the carboxylic acid.
To avoid this hydrolysis, we planned to utilize a tert-butyl ester as
a sterically hindered counterpart, which theoretically should
retard the hydrolysis process as shown in Scheme 4. We prepared
the tert-butyl acrylate 15 following the same procedure described
for 12 (Scheme 3). Surprisingly, use of intermediate 15 not only
eliminated the hydrolysis of the ester but also improved trans/cis
selectivity (no cis isomer of 16 has been detected). Although high
yield and selectivity have been achieved, we faced operational
challenges (efficient stirring, work-up) during hundred-gram scale
Scheme 4. Utilization of tert-butyl ester – the optimized synthesis: (a) 9-BBN, Pd
PPh , K PO , tert-butyl-3-iodoacrylate, THF/Water, 82% (b) TBAF, THF, 82% (c)
DIBAL, THF, 88%
(
)
3 4
3
4
2 3
synthesis due to poor solubility of Cs CO . In our further optimiza-
tion process, we discovered that the use of TBAF as a base [9] made
the reaction mixture completely homogeneous and provided an
excellent yield of 16 as a single diastereomer. After achieving high
yield and selectivity, our next goal was to directly reduce ester 16
to alcohol 1.
unattractive for consideration in multigram-scale syntheses. To
make this process more efficient we decided to use an allyl zinc
reagent, which are well known [7] to afford better selectivity when
added to sulfinimine. Accordingly, we revised the synthetic scheme
as shown below (Scheme 3). Conversion of 5a to 2 went smoothly
to a single diastereomer. Compound 2 was reacted with 9-BBN [8]
Attempted reduction of 16 with LAH gave a complicated mix-
ture including de-brominated product. Use of both LiBH
ture of LiBH /LiEt BH cleanly reduces the ester albeit at a very slow
pace. Finally DIBAL [10] was found to be the reagent of choice for
4
or a mix-
4
3
Scheme 5. Oxidation of sulfinamide to sulfonamide.
Please cite this article as: S. Chowdhury, J. Mao, J. Martynow et al., Diastereoselective scalable synthesis of 2,6-trans-Piperidines using an aza-Michael reac-
tion, Tetrahedron Letters, https://doi.org/10.1016/j.tetlet.2018.12.061

4
S. Chowdhury et al. / Tetrahedron Letters xxx (xxxx) xxx
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.tetlet.2018.12.061. These data include
MOL files and InChiKeys of the most important compounds
described in this article.
Scheme 6. Trans-cis isomerization does not occur.
References
[
1] (a) G.M. Struntz, J.A. Findlay, in: A. Brossi (Ed.), The Alkaloids, Academic Press,
New York, 1985, pp. 89–193;
(
b) M. Schneider, in: S.W. Pelletier (Ed.), Alkaloids: Chemical and Biological
Perspectives, Pergamon, Oxford, 1996, pp. 155–299;
c) H. Takahata, H. Ouchi, M. Ichinose, H. Nemoto, Org. Lett. 4 (2002) 3459–
462;
(
3
(
(
d) Y. Ying, J. Hong, Org. Lett. 13 (2011) 796–799;
e) M. Guerola, M. Sánchez-Roselló, C. Mulet, C. del Pozo, S. Fustero, Org. Lett.
1
7 (2015) 960–963;
f) H. Li, J. Wu, Org. Lett. 17 (2015) 5424–5427.
2] (a) J.S. Weintraub, P.M. Sabol, J.M. Kane, D.R. Borcherding, Tetrahedron 59
(
[
(
(
(
(
2003) 2953–2989;
b) M.G.P. Buffat, Tetrahedron 60 (2004) 1701–1729;
c) J. Cossy, Chem. Rec. 5 (2005) 70–80;
Fig. 2. ‘‘Matched (SS)” and ‘‘mismatched (SR)” systems.
d) C. Gnamm, C.M. Krauter, K. Broedner, G. Helmchen, Chem. Eur. J. 15 (2009)
2
(
1
050–2054;
e) C. Gnamm, C.M. Krauter, K. Broedner, G. Helmchen, Chem. Eur. J. 15 (2009)
0514–10532;
the reduction in large scale. After reduction, the desired trans-
piperidine 1 was isolated by crystallization. NMR spectra of both
the ester and the alcohol were not conclusive for structural assign-
ment. So, 1 was oxidized to 17 (Scheme 5). The X-ray structure of
(f) R.S.C. Kumar, E. Sreedhar, V.G. Reddy, K.S. Babu, J.M. Rao, Tetrahedron
Asymmetry 20 (2009) 1160–1163;
(
5
(
g) R.S.C. Kumar, V.G. Reddy, K.S. Babu, J.M. Rao, Chem. Lett. 38 (2009) 564–
65;
g) A. Guérinot, A. Serra-Muns, C. Gnamm, C. Bensoussan, S. Reymond, J. Cossy,
1
7 confirmed the desired stereochemistry.
After completion of the synthesis of 1, we also successfully
extended this methodology to other analogs (see SI for 16b-16e).
This unique feature of the intra-molecular aza-Michael reaction
with high stereoselectivity can plausibly be explained as shown
in Fig. 1. Both the size of the acrylate and stereochemistry of the
sulfinyl group help to bind the counter ion effectively (18) to gen-
erate the trans-ring system (16). Further mechanistic detail will be
published elsewhere.
Org. Lett. 12 (2010) 1808–1811.
3] D. Xiao, C. Wang, H.-C. Tsui, A. Palani, R. Aslanian, V.A. Buevich, Tetrahedron
Lett. 54 (2013) 6199–6203.
4] (a) K. Takasu, S. Maiti, M. Ihara, Heterocycles 59 (2003) 51;
(b) M. Bandini, A. Eichholzer, M. Tragni, A. Umani-Ronchi, Angew. Chem., Int.
Ed. 47 (2008) 3238;
[
[
(
(
c) A. Rolfe, K. Young, P.R. Hanson, Eur. J. Org. Chem. (2008) 5254;
d) Y. Ying, J. Hong, Org. Lett. 13 (2011) 796–799;
(e) C. Zhong, Y. Wang, A.W. Hung, S.L. Schreiber, D.W. Young, Org. Lett. 13
2011) 5556–5559.
(
Treatment of 16 with base (Cs
2 3
CO , 1–10 equiv., DMA, 40 °C,
[
[
5] N. Salih, H. Adams, F.W.R. Jackson, J. Org. Chem. 81 (2016) 8386–8393.
6] M.T. Robak, M.A. Herbage, J.A. Ellman, Chem. Rev. 110 (2010) 3600–3740.
4
h) did not show any isomerization to the thermodynamically
more stable cis-form 19 (Scheme 6).
[7] X.-W. Sun, M.-H. Xu, G.-Q. Lin, Org. Lett. 8 (2006) 4979–4982.
[
[
8] M. Gärtner, R. Weihofen, G. Helmchen, Chem. Eur. J. 17 (2011) 7605–7622.
9] G.V.M. Sharma, V.G. Reddy, A.S. Chander, R.K. Reddy, Tetrahedron Asymmetry
We also observed (Fig. 2) that only compound 15 [11] (matched
stereochemistry) and its enantiomer delivered desired products
with high yield and selectivity, whereas compound 20 (mis-
matched stereochemistry) and its enantiomer showed poor ten-
dency towards cyclization.
This is the first example of hitherto unknown large-scale syn-
thesis of 2,6-trans-piperidine analogs from an extremely poor
Michael acceptor. Further application of this methodology to other
systems is in progress. The important product 1 was utilized to
prepare a novel class of anti-infective agents.
13 (2002) 21–24.
[10] R. White, O. Epstein, J.B. Human, X.M. Zheng, K. Sham, Q. Liu, N. Chen, Y. Cheng,
PCT Int. Appl. (2014), WO 2014078314A1.
11] Compound 15 is an interesting intermediate. It can be easily converted to
trans-alcohol 21 or cis-alcohol 23 using simple chemical manipulation as
shown below.
[
Acknowledgments
We thank Professor William L. Jorgensen, Sterling Professor of
Chemistry, Yale University, New Haven, for valuable suggestions.
We acknowledge Dr. Mark Haddam, Albany Molecular Research
Inc., Albany, NY for crystallizing compound 17. Dr. Brandon
Mercado in the Department of Chemistry in Yale University is
appreciated for contributions involving X-ray crystallography.
Reagent: (a) TBAF, THF. (b) (i) LiBH
d) LiBH
4 3
, (ii) HCl, MeOH (c) (i) HCl, MeOH (ii) Et N
(
4
.
Please cite this article as: S. Chowdhury, J. Mao, J. Martynow et al., Diastereoselective scalable synthesis of 2,6-trans-Piperidines using an aza-Michael reac-
tion, Tetrahedron Letters, https://doi.org/10.1016/j.tetlet.2018.12.061