Lipase catalysis in the preparation of 3-(1-amino-3-butenyl)pyridine enantiomers

Article abstract of DOI:10.1016/j.tetasy.2012.09.003

The kinetic resolution of 1-(3-pyridyl)buten-3-ylamine with activated and non-activated acyl donors and Burkholderia cepacia lipase (lipase PS-D) under dry conditions has been studied. The N-acylation in isopropyl acetate (E >100) and the acidic hydrolysi

Full text of DOI:10.1016/j.tetasy.2012.09.003

Tetrahedron: Asymmetry 23 (2012) 1629–1632
Contents lists available at SciVerse ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Lipase catalysis in the preparation of 3-(1-amino-3-butenyl)pyridine
enantiomers
Ari Hietanen a, Tiina Saloranta b, Reko Leino b, Liisa T. Kanerva a,
⇑
a Institute of Biomedicine/Department of Pharmacology, Drug Development and Therapeutics/Laboratory of Synthetic Drug Chemistry and Department of Chemistry,
University of Turku, FIN-20014 Turku, Finland
b Laboratory of Organic Chemistry, Åbo Akademi University, FI-20500 Åbo, Finland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 2 August 2012
Accepted 6 September 2012
The kinetic resolution of 1-(3-pyridyl)buten-3-ylamine with activated and non-activated acyl donors and
Burkholderia cepacia lipase (lipase PS-D) under dry conditions has been studied. The N-acylation in isopro-
pyl acetate (E >100) and the acidic hydrolysis of the (R)-amide produced gave the corresponding enanti-
omerically enriched amines.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
to the results obtained with 1-phenylethylamine (E >200) even un-
der the harsh reaction conditions using a 1:1 mixture of the amine
In the literature, pyridine-based primary amines (compared to
the corresponding benzene-based analogues) have often turned
out to be difficult substrates for lipases, especially in terms of
enantioselectivity [generally measured with an E (enantiomer ra-
tio) value for the reaction].1 The basic properties of the lone-pair
of electrons on the sp2 nitrogen can lead to side reactions, such
as precipitation as a pyridinium salt, or possibly to different bind-
ing modes of benzene- and pyridine-based substrates at the active
site of a lipase. Thus, low enantioselectivity has been reported for
the CAL-B-catalyzed N-acylation of 1-(2-pyridyl)ethylamine in
neat ethyl acetate (E = 41 at 60 °C and E = 66 at 30 °C) 2,3 compared
with ethyl acetate.4 The N-acylation of 2-(3-pyridyl)propanamine
with ethyl methoxyacetate and CAL-B in THF also proceeded with
low enantioselectivity (E = 52) while that of the 3-indolylpropan-
amine (the lone-pair electrons of the pyrrole nitrogen are incorpo-
rated into the
p
system) gave E >200 under the same conditions.5
Herein we report our efforts in resolving racemic 1-(3-pyri-
dyl)buten-3-ylamine rac-1 by lipase-catalyzed N-acylation
(Scheme 1). Primarily, lipase PS-D (lipase from Burkholderia cepacia
adsorbed on diatomaceous earth) has been of interest since the li-
pase was previously found to be very effective for the N-acylation
of 1-phenylbuten-3-ylamine (E = 200 in neat isopropyl acetate at
N
N
N
NH2
rac-1
R1OH
+
O
O
HN
R
NH2
+ HO-lipase
(R)-2
(S)-1
OR1
R
O-lipase
R
O
Acyl-enzyme
intermediate
N
H2O
NH2
(R)-1
Scheme 1. Transformation of rac-1 into (R)- and (S)-1.
⇑
Corresponding author. Tel.: +358 2 333 6773; fax: +358 2 333 7955.
E-mail address: lkanerva@utu.fi (L.T. Kanerva).
0957-4166/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetasy.2012.09.003

1630
A. Hietanen et al. / Tetrahedron: Asymmetry 23 (2012) 1629–1632
23 °C).6 Nitrogen heterocycles are frequent moieties found in chiral
pharmaceuticals and natural products. Moreover, the terminal
double bond of the butenyl side-chain of 1 allows further synthetic
applications of the molecule. For example, rac-1 was previously
used as a starting material for the synthesis of rac-anatabine
[(1,2,3,6)-tetrahydro-2,30-bipyridine] and rac-anabasine [3-(2-
piperidyl)pyridine], the enantiomers of which are naturally occur-
ring tobacco alcaloids.7
from Candida antarctica, catalyze the hydrolysis of some amides
in water, although the hydrolyses seem to be slow. It is worth men-
tioning that even the so-called residual water in the dry lipase cat-
alyst may cause problems with ester substrates, thus concerning
both the acyl donors and O-acylated product enantiomers in ki-
netic resolutions in dry organic solvents.
2. Results and discussion
The lipase-promoted N-acylation of the primary amino group in
rac-1 is interesting due to the potential to obtain both enantiomers
[one as an amide product (R)-2 and the other as an unreacted
amine (S)-1] from one enzymatic reaction. Although typically con-
nected to ester (and accordingly alcohol rather than amine) sub-
strates, lipases (E.C. 3.1.1.3) have been extensively applied to the
kinetic resolution of primary8–12 and in some cases even second-
ary13–15 amines by N-acylation. While O-acylation by lipases tend
to be reversible (usually an acyl donor is chosen such that the reac-
tion becomes practically irreversible, allowing successful kinetic
resolution), N-acylation is typically considered irreversible. The
stability of amide bonds toward lipases can be explained by previ-
ously published data indicating that esterases/lipases lack a hydro-
gen bond between the substrate amide nitrogen and the enzyme,
which typically facilitates nitrogen inversion during the catalytic
process of serine proteases/amidases at the transition state for
hydrolysis.16 In fact, some lipases, such as lipases A17 and B4,18–24
Several side-reactions are possible when lipase-catalyzed N-
acylations in general, and those of pyridine-based amines in partic-
ular, are considered. (1) Non-enzymatic base-catalyzed N-acylation
leading to the formation of a racemic product may compete with
the lipase-catalyzed N-acylation especially in the presence of acti-
vated acyl donors. Accordingly, both activated and non-activated
acylating agents have been studied. (2) Should any enzymatic
hydrolysis of an acyl donor by the residual water in the enzyme
take place, the pyridine containing compounds may be salted out
of the solution by the obtained carboxylic acid. This issue concerns
the N-acylation of pyridine containing amines in particular, since
there are only limited options for the N-protection of the pyridine
nitrogen. In order to minimize the acid formation, all enzymatic
reactions have been performed in the presence of molecular sieves
(4 Å). (3) Racemization of one of the resolution products is also
possible as is the dimerization of the amine.
Table 1
Kinetic resolution of rac-1 (0.1 M) with activated acyl donors and lipase PS-D (50 mg mLꢀ1) in toluene at 22 °C in the presence of molecular sieves (50 mg mLꢀ1, 4 Å)
Entry
1
Acyl donor
Acyl donor (equiv)
Temp (°C)
Timea (h)
Conversion (%)
ee(R)-2 (%)
ee(S)-1 (%)
E
O
O
O
2
22
30
13
30
5
1
F3C
F3C
F3C
O
2
3
2
2
22
22
30
30
5
0
<1
—
2
1
O
O
—
—
O
4
5
6
7
2
2
2
1
22
22
22
22
30
30
72
1
12
30
52
51
n.d.b
n.d.b
90
12
19
99
91
<10
O
O
Cl
O
O
O
O
O
<10
>100
n.d.c
O
O
O
O
O
89
Cl
Cl
Cl
Cl
Cl
F3C
F3C
F3C
F3C
F3C
O
8
2
4
4
4
4
1
1
1
1
50
50
46
35a
92
94
97
97
90
94
82
53
72
O
O
O
O
9
1
>100
>100
>100
10
11
0.75
0.5
a
b
c
The reaction stopped at the time point given.
Enantiomers did not fully separate in HPLC.
Not determined due to concomitant chemical acylation (conversion 4% in 1 h without enzyme).

A. Hietanen et al. / Tetrahedron: Asymmetry 23 (2012) 1629–1632
1631
Table 2
Kinetic resolution of rac-1 (0.1 M) with lipase PS-D in neat non-activated acyl donors CH3CO2R1 at 22 °C in the presence of molecular sieves (50 mg mLꢀ1, 4 Å)
Entry
R1
Temp (oC)
Lipase (mg mLꢀ1
)
Time (h)
Conversion (%)
ee(R)-2 (%)
ee(S)-1 (%)
E
1
2
3
4
5
6
7
8
Et
i-Pr
22
22
48
22
22
48
22
22
(50)
(50)
(50)
72
72
72
96
96
96
96
96
12
32
50
48
50
53
39
33
64
97
87
96
95
87
97
92
9
46
86
87
94
97
63
45
4
>100
38
>100
>100
58
>100
37
i-Pra
i-Pra
i-Pra
i-Pra
i-Pra,b
i-Pra,c
(75)
(100)
(100)
(100)
(100)
a
b
c
The reaction practically stopped at the time given.
iPrOAc (50 vol %) in PhCH3.
CAL-B (Novozym 435) in the place of lipase PS-D.
As an initial set-up, lipase PS-D was used to acylate rac-1
be stable against racemization in isopropyl acetate and in its solu-
tions with 5 vol % of pyridine, triethylamine, acetic acid or isopro-
pyl alcohol as well as in the presence of silica-gel at room
temperature at least for 24 h.
(0.1 M) with variously activated esters (0.2 M) in toluene in the
presence of molecular sieves (Table 1). The use of a correctly cho-
sen activated acyl donor (an activated ester herein) may drastically
influence the course of the resolution process in two ways pro-
vided that the acyl donor binds correctly at the active site to form
an acyl-enzyme intermediate with the serine hydroxyl of the lipase
(Scheme 1). First, the alkyl portion of the acylating ester RCO2R1
should be such that it leads to irreversible acylation, that is,
R1OH does not enzymatically react with the enantiomerically en-
riched (R)-2 back to (R)-1 leading to the racemization of both 1
and 2. Second, the high reactivity of the activated esters allows fast
acylation at low ester concentrations in organic solvents. 2,2,2-Tri-
fluoroethyl esters with activated alkyl ends (common acyl donors
in the kinetic resolution of various b-amino esters) 9 were all inef-
fective for the present substrate with negligible enantioselectivi-
ties (entries 1–3). Acyl activation as a methoxy- (entries 4 and 5)
or chloro- (entry 6) acetate improved the reactivity, whereas the
enantioselectivity was good only with ethyl chloroacetate (entry
6). The value E <10 (entry 4) observed for the N-acylation with
ethyl methoxyacetate was low compared to the 200 value previ-
ously observed for the lipase-PS-D-catalyzed kinetic resolution of
the benzene-based analogue,6 thus confirming the claim in the
Introduction that pyridine-based substrates are difficult in terms
of enantioselectivity. 2,2,2-Trifluoroethyl chloroacetate as an acyl
and alkyl activated acyl donor (entry 7) gave chemical acylation
(4% in 1 h) when reagents other than lipase PS-D were mixed. Thus,
E may incorrectly measure the enantioselectivity, and so the E va-
lue is not given (entry 7). However, the high ee(R)-2 value (89%,
R = CH2Cl) at around 50% conversion together with high reactivity
encouraged us to develop the conditions further in order to sup-
press chemical acylation. When the temperature was dropped
from 22 to 4 °C, N-acylation without the enzyme was less than
1% in one hour, while the time needed to reach 50% conversion
in the presence of lipase PS-D was unchanged and the enantiopu-
Acylation of rac-1 with non-activated acyl donors (ethyl and
isopropyl acetates) was next studied in order to yield more stable
amides (R)-2 (R = CH3; Table 2). Compared to the reactions with
the activated acyl donors (Table 1), N-acylations typically pro-
ceeded slowly even when the acyl donor was also the solvent
(reaction in neat acyl donor). The reaction of rac-1 (0.1 M) with li-
pase PS-D (50 mg mLꢀ1) in ethyl acetate (entry 1) resulted in lower
reactivity (12% conversion in 72 h) and enantioselectivity (E 4)
than the reaction (32% conversion with E >100) in isopropyl acetate
(entry 2) at room temperature, with the reactions practically ceas-
ing below 50% conversion. Elevated temperature (entry 3) or high-
er enzyme contents (entries 4 and 5) both improved the conversion
to 50%. However, increasing the temperature considerably lowered
the enantioselectivity in isopropyl acetate (entry 2 vs 3 and 5 vs 6).
Reactions in neat isopropyl acetate were preferred over the reac-
tion in solution with toluene (entry 7 vs 5), and CAL-B (Novozym
435), a commonly employed catalyst for the N-acylation of primary
amines,7–12 was considerably less enantioselective (E = 37, entry 8)
than lipase PS-D (E >100, entry 5).
Finally, the preparative scale kinetic resolution of rac-1 in neat
isopropyl acetate was performed at room temperature. The reac-
tion with lipase PS-D (100 mg mLꢀ1) proceeded to 47% conversion
in 96 h. After work-up and purification (see Section 3), amide (R)-2
(R = CH3) was obtained in 44% (ee = 97%) and amine (S)-1 in 39%
(ee = 88%) isolated yields from rac-1. It was also seen that the
acidic hydrolysis of (R)-2 (R = CH3) under microwave irradiation
and the following basic work-up (pH <10) could be used to afford
(R)-1 with a negligible drop in ee. Thus, the hydrolysis of a sample
of (R)-2 (R = CH3; ee = 88%) afforded (R)-1 in 78% isolated yield and
85% ee (see Section 3).
rities were improved (ee = 94%, entry
9
vs ee(R)-2 = 89% and
3. Experimental
ee(S)-1 = 91%, entry 7). Next, studies were continued at 4 °C. When
the ester content now was decreased from 2 to 0.5 equiv, the
ee(R)-2 increased from 92% to 97% (entries 8–11) justifying the
use of E >100 with decreasing chemical N-acylation at lower con-
versions. However, with only 0.5 equiv of 2,2,2-trifluoroethyl chlo-
roacetate (i.e., 1 equiv with respect to one of the enantiomers), the
reaction ceased at 35% conversion (entry 11). When the kinetic res-
olution with 2,2,2-trifluoroethyl chloroacetate (0.75 equiv) was
3.1. Materials and methods
Lipase PS-D was acquired from AmanoEurope and CAL-B (Nov-
ozym 435) from Novozymes. Ethyl methoxyacetate, ethyl chloro-
acetate and isopropyl acetate were purchased from Aldrich.
Isopropyl methoxyacetate and 2,2,2-trifluoroethyl esters were syn-
thesized from the corresponding acid chloride and alcohol while
rac-1 was prepared from 3-pyridinecarboxaldehyde based on a
Barbier-type Zn-mediated allylation7 as previously described. The
purities of the products were checked by NMR before use. Toluene
was from J. T. Baker and dried before use.
performed on
a
preparative scale at 4 °C, ee(S)-1 = 90% and
ee(R)-2 = 97% at 48% conversion were detected in the resolved reac-
tion mixture. However, the work-up and column chromatography
(CH2Cl2/MeOH/Et3N 40:1:1 as an eluent) yielded negligible
amounts of the products [the amide product (R)-2 being impure]
with decreased ee values [ee(S)-1 = 82% and ee(R)-2 only 50%], and
these studies were not continued. Compound (S)-1 was shown to
NMR spectra were recorded with a Bruker Avance 600 MHz
spectrometer. HRMS were measured in ESI+ mode with a Bruker
micrOTOF-Q quadrupole-TOF spectrometer. Optical rotations were

1632
A. Hietanen et al. / Tetrahedron: Asymmetry 23 (2012) 1629–1632
determined with a PerkinElmer 241 or 341 polarimeter, and [
a
]
ing to room temperature, the reaction was treated with aq NH4OH
(24.5%, 6 mL) and the basic solution was extracted with EtOAc
(3 ꢂ 10 mL). The organic phases were combined, dried over NaSO4,
and concentrated to yield (R)-1 (16 mg, 0.109 mmol, 78%,
D
values are given in units of 10ꢀ1 deg cm2 gꢀ1. Flash chromatogra-
phy was performed using silica gel 60 Å (Merck, 230–400 mesh,
enriched with 0.1% Ca).
Small-scale kinetic resolutions were performed in vials with
approximately 1 mL of reaction volume. Agitation, both on the
small and preparative scale, was carried out by shaking. The deter-
mination of E is based on equation E = ln[(1-c)(1-eeS)]/ln[(1-
c)(1+eeS)] with c = eeS/(eeS+eeP) using linear regression where E
is the slope of the line ln[(1-c)(1-eeS)] versus ln[(1-c)(1+eeS)].1
ee = 85%). (R)-1: ½a D25
¼ þ38 (c 0.5, CHCl3). NMR spectra were in
ꢁ
accordance with those reported in the literature.7
Acknowledgements
The authors thank the Academy of Finland for financial support
(research grants #121983 for L.K. and #121334 for RL/Fusing Bio-
catalytic and Chemocatalyzed Reaction Technologies).
3.2. Kinetic resolution of rac-1
One of the acyl donors (0.2 M or 1 mL) was added into a reac-
tion vial containing molecular sieves (4 Å, 50 mg) and the enzyme
(50–100 mg mLꢀ1), toluene (1 mL) when used and rac-1 (0.1 M)
were added. The reaction mixture was shaken (170 rpm) at a given
temperature. Reactions were followed by HPLC equipped with a
Daicel CHIRALCEL OD-H analytical column (40 °C) which was
eluted with 10% iPrOH in hexane (0.8 mL minꢀ1) and detected with
a UV detector at 235 nm. HPLC samples were derivatized either
with acetic or butanoic anhydride.
References
1. Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am. Chem. Soc. 1982, 104,
7294–7299.
2. Skupinska, K. A.; McEachern, E. J.; Baird, I. R.; Skerlj, R. T.; Bridger, G. J. J. Org.
Chem. 2003, 68, 3546–3551.
3. Iglesia, L. E.; Sánchez, V. M.; Rebolledo, F.; Gotor, V. Tetrahedron: Asymmetry
1997, 8, 2675–2677.
4. Päiviö, M.; Perkiö, P.; Kanerva, L. T. Tetrahedron: Asymmetry 2012, 23, 230–236.
5. Rodríguez-Mata, M.; Gotor-Fernández, V.; González-Sabín, J.; Rebolledo, F.;
Gotor, V. Org. Biomol. Chem. 2011, 9, 2274–2278.
6. Hietanen, A.; Saloranta, T.; Rosenberg, S.; Laitinen, E.; Leino, R.; Kanerva, L. T.
Eur. J. Org. Chem. 2010, 909–919.
3.3. Preparative-scale kinetic resolution of rac-1
7. Saloranta, T.; Leino, R. Tetrahedron Lett. 2011, 52, 4619–4621.
8. Busto, E.; Gotor-Fernández, V.; Gotor, V. Chem. Rev. 2011, 111, 3998–4035.
9. Liljeblad, A.; Kanerva, L. T. Tetrahedron 2006, 62, 5831–5854.
10. Höhne, M.; Bornscheuer, U. T. Chem. Cat. Chem. 2009, 1, 42–51.
11. Turner, N. J. Curr. Opin. Chem. Biol. 2010, 14, 115–121.
12. Lee, J. H.; Han, K.; Kim, M.-J.; Park, J. Eur. J. Org. Chem. 2010, 999–1015.
13. Liljeblad, A.; Kiviniemi, A.; Kanerva, L. T. Tetrahedron 2004, 60, 671–677.
14. Hietanen, A.; Lundell, K.; Kanerva, L. T.; Liljeblad, L. Arkivoc 2012, V, 60–74.
15. Gotor-Fernández, V.; Fernández-Torres, P.; Gotor, V. Tetrahedron: Asymmetry
2006, 17, 2558–2564.
16. Syrén, P.-O.; Hult, K. Chem. Cat. Chem. 2011, 3, 853–860.
17. Liljeblad, A.; Kallio, P.; Vainio, M.; Niemi, J.; Kanerva, L. T. Org. Biomol. Chem.
2010, 8, 886–895.
18. Smidt, H.; Fischer, A.; Fischer, P.; Schmid, R. D. Biotechnol. Tech. 1996, 10, 335–
338.
Compound rac-1 (113 mg, 0.764 mmol) was dissolved in i-PrO-
Ac (7.65 mL) and the solution was added to lipase PS-D (765 mg,
100 mg mLꢀ1) and 4 Å molecular sieves (770 mg). After 96 h of
shaking at +22 °C (47% conversion), the reaction was filtered and
the enzyme and molecular sieves were washed with i-PrOAc. The
organic phases were combined, concentrated and purified by col-
umn chromatography (5% Et3N in EtOAc) to give (R)-2 [64 mg,
0.337 mmol, 44%, ee = 97%, purity 94%, ½a D25
¼ þ82 (c 1.0, CHCl3)]
ꢁ
and (S)-1 (44 mg, 0.297 mmol, 39%, ee = 88%, ½a D25
ꢁ
¼ ꢀ39 (c 0.5,
CHCl3). (R)-2: 1H NMR (600.13 MHz, CDCl3, 25 °C): d = 8.76 (s, 1H,
19. Wagegg, T.; Enzelberger, M. M.; Bornscheuer, U. T.; Schimd, R. D. J. Biotechnol.
1998, 61, 75–78.
HAr2); 8.52 (d, JAr6–Ar5 = 3.0 Hz, 1H, HAr6); 7.75 (d, JAr4–Ar5 = 8.40 Hz,
1H, H ); 7.38 (dd, 1H, HAr5); 6.27 (d, JNH–CH = 6.60 Hz, 1H, NH);
5.68 (Amr4 , 1H, CH@); 5.16 (m, 1H, CH); 5.13 (m, 2H, @CH2); 2.61
(m, 2H, –CH2–); 2.03 (s, 3H, COCH3) ppm; HRMS: calcd for
20. Duarte, D. R.; Castillo, E. L.; Bárzana, E.; López-Munguía, A. Biotechnol. Lett.
2000, 22, 1811–1814.
21. Sergeeva, M. V.; Mozhaev, V. V.; Rich, J. O.; Khmelnitsky, Y. L. Biotechnol. Lett.
2000, 22, 1419–1422.
22. Torres-Gavilán, A.; Escalante, J.; Regla, I.; López-Munguía, A.; Castillo, E.
Tetrahedron: Asymmetry 2007, 18, 2621–2624.
23. Ismail, H.; Lau, R. M.; van Rantwijk, F.; Sheldon, R. A. Adv. Synth. Catal. 2008,
350, 1511–1516.
C
11H14N2ONa+ [M+Na]+ 213.0998, found 213.1071. (S)-1: NMR
spectra for (S)-1 were in accordance with those reported in the
literature.7
24. Brem, J.; Bencze, L.-C.; Liljeblad, A.; Turcu, M. C.; Paizs, C.; Irimie, F.-D.;
Kanerva, L. T. Eur. J. Org. Chem. 2012, 3288–3294.
3.4. Hydrolysis of (R)-2 to give (R)-1
(R)-2 (26 mg, 0.138 mmol, ee 88%) was stirred in 2 M aq HCl
(2.6 mL) under microwave heating at 165 °C for 15 min. After cool-