The effect of substrate structure on the chemoselectivity of Candida antarctica lipase B-catalyzed acylation of amino-alcohols

Article abstract of DOI:10.1016/j.molcatb.2012.09.006

The selective acylation of multifunctional compounds exhibiting both alcohol and amino groups gives interesting products with many applications in food, cosmetic and pharmaceutical industries, but it is real challenge. The current work describes the different behavior shown by Candida antarctica lipase B (Novozym 435) when catalyzing the O-acylation and N-acylation of bifunctional acyl acceptors. The acylation of three amino-alcohols (alaninol, 4-amino-1-pentanol and 6-amino-1-hexanol) was studied using myristic acid as an acyl donor. To achieve this, a structure-reactivity study was performed in tert-amyl alcohol as a solvent, comparing the three amino-alcohols as acyl acceptors and a series of structurally related amines, namely (R)-sec-butylamine, 1-methoxy-2-propylamine and 1,2-diaminopropane. These substrates were designed to investigate the effect of the group located in β-position of the amino group on the acyl acceptor: the more nucleophilic the group, the more the apparent maximal velocity (V_max,app) of N-acylation increases. Moreover, the crucial role of the carbon chain length between the alcohol and amino groups on the chemoselectivity was also demonstrated. The chemoselectivity for the N-acylation was improved when the carbon chain included two carbons (alaninol) whereas the chemoselectivity for the O-acylation was improved when the carbon chain included four carbons or more (4-amino-1-pentanol and 6-amino-1-hexanol). These results provided new insights for the selective synthesis of amides or esters produced from the acylation of bifunctional substrates.

Full text of DOI:10.1016/j.molcatb.2012.09.006

Journal of Molecular Catalysis B: Enzymatic 85–86 (2013) 193–199
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
The effect of substrate structure on the chemoselectivity of Candida antarctica
lipase B-catalyzed acylation of amino-alcohols
Florian Le Joubioux, Yesmine Ben Henda, Nicolas Bridiau, Oussama Achour,
Marianne Graber, Thierry Maugard∗
UMR 7266 CNRS-ULR, LIENSs, Equipe Approches Moléculaires, Environnement-Santé, Université de La Rochelle, Avenue Michel Crépeau, 17042 La Rochelle, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 24 January 2012
Received in revised form 24 August 2012
Accepted 12 September 2012
Available online 19 September 2012
The selective acylation of multifunctional compounds exhibiting both alcohol and amino groups gives
interesting products with many applications in food, cosmetic and pharmaceutical industries, but it is
real challenge. The current work describes the different behavior shown by Candida antarctica lipase
B (Novozym 435) when catalyzing the O-acylation and N-acylation of bifunctional acyl acceptors. The
acylation of three amino-alcohols (alaninol, 4-amino-1-pentanol and 6-amino-1-hexanol) was studied
using myristic acid as an acyl donor. To achieve this, a structure-reactivity study was performed in tert-
amyl alcohol as a solvent, comparing the three amino-alcohols as acyl acceptors and a series of structurally
related amines, namely (R)-sec-butylamine, 1-methoxy-2-propylamine and 1,2-diaminopropane. These
substrates were designed to investigate the effect of the group located in ␤-position of the amino group
Keywords:
N-acylation
O-acylation
Candida antarctica lipase B
Organic solvent
Chemoselectivity
on the acyl acceptor: the more nucleophilic the group, the more the apparent maximal velocity (Vmax,app
)
of N-acylation increases. Moreover, the crucial role of the carbon chain length between the alcohol and
amino groups on the chemoselectivity was also demonstrated. The chemoselectivity for the N-acylation
was improved when the carbon chain included two carbons (alaninol) whereas the chemoselectivity for
the O-acylation was improved when the carbon chain included four carbons or more (4-amino-1-pentanol
and 6-amino-1-hexanol).
These results provided new insights for the selective synthesis of amides or esters produced from the
acylation of bifunctional substrates.
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
use of biocatalysts can be an interesting alternative, which offers
a clean way to perform chemical processes under mild reaction
The selective acylation of amino-alcohols has applications in
many areas, such as pharmaceutics and cosmetics, through the syn-
thesis of ceramides [1,2] as potential anti-viral or anti-tumor drugs
[3,4] and anti-oxidant stabilizers [5], or for the environmental, food,
and agricultural industries, through the synthesis of glucamide or
amino-acid based surfactants [6–8].
Chemical acylation of amino-alcohols is well established but the
methods used are faced with several limitations. They need fas-
tidious steps of alcohol group protection and deprotection for the
control of chemoselectivity and stereoselectivity. The high temper-
atures often required for chemical synthesis also preclude the use
of fragile molecules and may cause coloration of final products. In
addition, the coproduction of salts and the use of toxic solvents
(dimethylformamide, methanol, etc.) that must be eliminated at
the end of the reaction increase the cost of the processes. The
conditions and with a high degree of selectivity. The use of immo-
bilized enzymes in organic media, in particular lipases (E.C. 3.1.1.3)
provides several advantages such as shifting of the thermodynamic
equilibrium in favor of synthesis over hydrolysis reaction, increas-
ing solubility of non-polar substrates, eliminating side reactions,
making easier enzyme recovery and increasing enzyme thermosta-
bility. Lipases are the most used enzymes for organic synthesis.
They have been used to catalyze O-acylation, transesterification
and N-acylation reactions to synthesize various multifunctional
derivates related to pharmaceuticals, cosmetics and foods. How-
ever, despite the large amount of studies on key enzyme properties
in biocatalysis, their chemoselectivity is still not completely under-
stood. It is therefore still necessary to optimize the output of
lipase-catalyzed reactions and make efforts to understand lipase
chemoselectivity.
Despite the attractive properties of lipases in organic solvents,
few studies have been devoted to the lipase-catalyzed acyla-
tion of bifunctional molecules exhibiting both amino and alcohol
groups such as ethanolamine, diethanolamine, 2-amino-1-butanol,
∗
Corresponding author. Fax: +33 546458277.
E-mail address: tmaugard@univ-lr.fr (T. Maugard).
1381-1177/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcatb.2012.09.006

194
F. Le Joubioux et al. / Journal of Molecular Catalysis B: Enzymatic 85–86 (2013) 193–199
Table 1
6-amino-1-hexanol, serine and amino-alcohols with variable car-
bon chain length [9–14]. In such reactions, the lipase was seen to
catalyze O-acylation or N-acylation, with a chemoselectivity which
is largely dependent on amino-alcohol structure.
Elution gradient for HPLC analysis of reaction samples resulting from the acylation
of amino-alcohols 1, 9, 11 and 13.
Time (min)
Solvent A:
acetonitrile/water/acetic
acid (77/23/0.1, v/v/v) (%)
Solvent B:
methanol/acetic acid
(100/0.1, v/v) (%)
Among lipases used in organic synthesis, Candida antarctica
lipase B is well known for its ability to convert alcohols and amines
into esters and amides in various organic solvents [8,15] and seems
to be the ideal enzyme for the acylation of compounds such as
amino-alcohols. In the present work, we thus investigated the
C. antarctica lipase B-catalyzed acylation of various amines and
amino-alcohols as acyl acceptors, using myristic acid as an acyl
donor. The results obtained under a kinetic approach were ana-
lyzed by comparing the apparent kinetic parameters Vmax,app and
Km,app obtained for each acyl acceptor.
0
20
23
80
82
90
100
100
0
0
100
100
0
0
100
100
0
0
2.4. HPLC and structural analysis
Structural and quantitative analysis of reaction products were
conducted using a LC/MS-ES system from Agilent (1100 LC/MSD
Trap mass spectrometer VL) with a C18 Prontosil 120-5-C18-AQ
reversed-phase column (250 mm × 4 mm, 5 ␮m; Bischoff chro-
matography). Products were detected and quantified by differential
refractometry and UV detection at 210 nm. An external calibra-
tion was performed with pure myristic acid. Then calibrations for
individual acylation products were obtained after mass balance in
reaction conditions enabling to obtain only these products with
myristic acid as an acyl donor. Low-resolution mass spectral anal-
yses were obtained by electrospray in the positive detection mode.
Nitrogen was used as the drying gas at 15 l min−1 and 350 ◦C at a
nebulizer pressure of 4 bar. The scan range was 50–1000m/z using
five averages and 13,000m/z per second resolution. The capillary
voltage was 4000 V. Processing was done offline using HP Chem-
station software.
2. Materials and methods
2.1. Enzyme and chemicals
Novozym® 435 (immobilized C. antarctica lipase B), was kindly
provided by Novozymes A/S, Bagsvaerd, Denmark. ( )-Alaninol
1 (98%), (R)-alaninol (R)-1 (98%), (S)-alaninol (S)-1 (98%), (R)-
sec-butylamine 3 (99%), ( )-1-methoxy-2-propylamine 5 (95%),
(
)-1,2-diaminopropane 7 (≥98%), ( )-isopropanolamine 9 (93%),
and 6-amino-1-hexanol 13 (≥97%), as well as tert-amyl alcohol
(99%) were purchased from Sigma–Aldrich (St. Louis, USA) while
(
)-4-amino-1-pentanol 11 was from Santa Cruz Biotechnology
(USA). Myristic acid and acetic acid were from Fluka (St. Quentin-
Fallavier, Switzerland). All chemicals were dried over molecular
sieves. Pure water was obtained via a Milli-Q system (Millipore,
France). Acetonitrile and methanol were purchased from Carlo
ERBA (Val-de-Reuil, France).
Various eluent systems were used depending on the acyl
acceptor used. Reaction samples resulting from the acylation of
(R)-sec-butylamine 3 were eluted with acetonitrile/water/acetic
acid (90/10/0.1, v/v/v) at room temperature and at a flow rate of
1 ml min−1. Reaction samples resulting from the acylation of 1-
methoxy-2-propylamine 5 and 1,2-diaminopropane 7 were eluted
with methanol/water/acetic acid (95/5/0.1 and 93/7/0.1, v/v/v,
2.2. Enzymatic reactions
In all cases, reactions were carried out in tert-amyl alcohol at
55 ◦C in screw-capped tubes. 2 ml reaction mixtures containing var-
ious amounts of acyl acceptor substrates (25–350 mM) and 175 mM
of myristic acid as an acyl donor were incubated for 10 min prior
to addition of 50 g l−1 of C. antarctica lipase B for the acylation of
sec-butylamine 3 or 5 g l−1 of C. antarctica lipase B for the acylation
of another acyl acceptor. 100 ␮l samples were taken at intervals
and centrifuged at 14,000 rpm. The supernatants were then ana-
lyzed by LC–MS, leading to the determination and quantification
of remaining substrates and synthesized products. Initial rate mea-
surements were performed according to a previously established
procedure [16]. The initial rates were calculated from the linear
relationship of the total concentration of products against reaction
time (0–2 h).
respectively) at room temperature and at a flow rate of 1 ml min−1
.
The elution of reaction samples resulting from the acylation of
amino-alcohols 1, 9, 11 and 13, was carried out at room temper-
ature and at a flow rate of 1 ml min−1, using a gradient that was
derived from two eluent mixtures (Table 1).
In order to perform the purification and characterization of
acylated products, 20 ml reaction mixtures containing 50 mM of
the acyl acceptor and 175 mM of myristic acid in tert-amyl alco-
hol were incubated for 24 h in presence of 15 g l−1 of C. antarctica
lipase B. Purified products were then characterized by 1H NMR and
IR after purification via preparative HPLC using a ProntoPrep C18
reversed-phase column (250 mm × 20 mm, 10 ␮m; Bischoff chro-
matography) eluted with the gradient given in Table 1, at room
temperature and at a flow rate of 5 ml min−1 1H NMR was recorded
.
2.3. Evaluation of the chemoselectivity
on a JEOL-JNM LA400 spectrometer (400 MHz), with tetramethylsi-
lane as an internal reference. Samples were studied as solutions in
CDCl3. Infrared (IR) spectra were recorded from 400 to 4000 cm−1
with a resolution of 4 cm−1 using a 100 ATR spectrometer (Perkin-
Elmer, United States).
The chemoselectivity of C. antarctica lipase B during the acyla-
tion of amino-alcohols was studied by comparing the alcohol group
O-acylation and the amino group N-acylation, and then calculated
via the apparent catalytic efficiency ratio (Eq. (1)) [17], which was
transformed into an apparent maximal velocity ratio (Eq. (2)) owing
to the fact that Km,app were identical for a given amino-alcohol.
2.4.1. N-myristyl 2-amino-1-propanol 2a
m/z (LR-ESI+) C17H35NO2 (M+H+), found: 286.4, calculated for:
286.48. IR vmax (cm−1): 3100–3500 (O H, alcohol and N H, amide),
2800–3000 (CH of myristyl chain), 1638 (C O, amide), 1543 (N H,
amide). 1H NMR (400 MHz, CDCl3, ı ppm): ı 0.88 (t, 3H, J = 6.06 Hz,
CH2 CH3), 1.17 (d, 3H, J = 6.06 Hz, CH CH3), 1.25 (m, 20H,
CH2 of myristyl chain), 1.63 (m, 2H, CH2 CH2 CO NH of
myristyl chain), 2.19 (t, 2H, J = 6.06 Hz, CH2 CH2 CO NH of
Vmax,app O−acylation/Km,app
Vmax,app N−acylation/Km,app
C =
C =
(1)
(2)
Vmax,app O−acylation
Vmax,app N−acylation

F. Le Joubioux et al. / Journal of Molecular Catalysis B: Enzymatic 85–86 (2013) 193–199
195
myristyl chain), 3.04 (s, 1H, OH), 3.27 (dd, 1H, J = 5 Hz, J = 10 Hz,
CH CH2 OH), 3.46 (dd, 1H, J = 3.7 Hz, J = 11 Hz, CH CH2 OH),
4.07 (m, 1H, CH ), 5.7 (s, 1H, NH ).
J = 6.55 Hz, 2x CH2 CH3), 1.22 (d, 3H, J = 6.39 Hz, CH CH3), 1.25
(m, 40H, CH2 of myristyl chain), 1.61 (m, 4H, 2x CH2 CH2 CO
of myristyl chain), 2.16 (t, 2H, J = 6.62 Hz, CH2 CH2 CO
O
of myristyl chain), 2.3 (t, 2H, J = 6.62 Hz, CH2 CH2 CO NH
of myristyl chain), 3.37 (m, 1H, CH CH2 NH ), 3.45 (m, 1H,
CH CH2 NH ), 4.99 (m, 1H, CH ), 5.69 (s, 1H, C NH CH2 ).
2.4.2. N,O-dimyristyl 2-amino-1-propanol 2c
m/z (LR-ESI+) C31H62NO3 (M+Na+), found: 518.6, calculated for:
518.85. IR vmax (cm−1): 3301 (N H, amide), 2800–3000 (CH of
myristyl chain), 1737 (C O, ester), 1643 (C O, amide), 1542 (N H,
amide). 1H NMR (400 MHz, CDCl3, ı ppm): ı 0.88 (t, 6H, J = 7.6 Hz, 2x
CH2 CH3), 1.16 (d, 3H, J = 7.6 Hz, CH CH3), 1.25 (m, 40H, CH2
of myristyl chain), 1.6 (m, 4H, 2x CH2 CH2 CO of myristyl
2.4.8. N-myristyl 4-amino-1-pentanol 12a
m/z (LR-ESI+) C19H39NO2 (M+H+), found: 314.2 calculated for:
314.53. IR vmax (cm−1): 3200–3500 (O H, alcohol and N H, amide),
2800–3000 (CH of myristyl chain), 1639 (C O, amide), 1545 (N H,
amide). 1H NMR (400 MHz, CDCl3, ı ppm): ı 0.88 (t, 3H, J = 6.58 Hz,
CH2 CH3), 1.14 (d, 3H, J = 6.23 Hz, CH CH3), 1.25 (m, 20H,
CH2 of myristyl chain), 1.53 (m, 4H, CH CH2 CH2 CH2 OH),
1.63 (m, 2H, CH2 CH2 CO NH of myristyl chain), 2.14 (t, 2H,
J = 7.27 Hz, CH2 CH2 CO NH of myristyl chain), 2.94 (s, 1H,
OH), 3.67 (m, 2H, CH2 CH2 OH), 4.06 (m, 1H, CH ), 5.28 (s,
1H, NH ).
chain), 2.14 (t, 2H, J = 7.2 Hz, CH2 CH2 CO
2.32 (t, 2H, J = 7.2 Hz, CH2 CH2 CO NH of myristyl chain), 4 (dd,
1H, J = 4.4 Hz, J = 10.7 Hz, CH CH2 ), 4.13 (dd, 1H, J = 4.9 Hz,
J = 10 Hz, CH CH2 ), 4.29 (m, 1H, CH ), 5.54 (d, 1H, J = 7.3 Hz,
NH CH2 ).
O of myristyl chain),
O
O
C
2.4.3. N-myristyl sec-butylamine 4a
m/z (LR-ESI+) C18H37NO (M+H+), found: 284.2, calculated for:
284.51. 1H NMR (400 MHz, CDCl3, ı ppm): ı 0.89 (m, 6H, 2x
CH2 CH3), 1.1 (d, 3H, J = 7 Hz, CH CH3), 1.25 (m, 20H, CH2
of myristyl chain), 1.43 (m, 2H, CH CH2 CH3), 1.6 (m, 2H,
CH2 CH2 CO NH of myristyl chain), 2.12 (st, 2H, J = 7 Hz,
CH2 CH2 CO NH of myristyl chain), 3.9 (st, 1H, J = 6.8 Hz,
CH ), 5.17 (s, 1H, NH ).
2.4.9. O-myristyl 4-amino-1-pentanol 12b
m/z (LR-ESI+) C19H39NO2 (M+H+), found: 314.2 calculated
for: 314.53. IR vmax (cm−1): 3291 (N H, amine), 2800–3000
(CH of myristyl chain), 1736 (C O, ester), 1557 (N H, amine).
1H NMR (400 MHz, CDCl3, ı ppm): ı 0.88 (t, 3H, J = 6.99 Hz,
CH2 CH3), 1.14 (d, 3H, J = 8 Hz, CH CH3), 1.25 (m, 20H,
CH2 of myristyl chain), 1.52 (m, 4H, CH CH2 CH2 CH2
O
),
2.4.4. N-myristyl 1-methoxy-2-propylamine 6a
1.62 (m, 2H, CH2 CH2 CO
1H, J = 7.16 Hz, CH2 CH2 CO
1H, J = 7.5 Hz, CH2 CH2 CO
O
O
O
of myristyl chain), 2.22 (t,
of myristyl chain), 2.29 (t,
of myristyl chain), 3.43 (m,
m/z (LR-ESI+) C18H37NO2 (M+H+), found: 300.3, calculated for:
300.51. IR vmax (cm−1): 3304 (N H, amide), 2800–3000 (CH of
myristyl chain), 1634 (C O, amide), 1544 (N H, amide). 1H NMR
(400 MHz, CDCl3, ı ppm): ı 0.88 (t, 3H, J = 6.79 Hz, CH2 CH3),
1.17 (d, 3H, J = 6.17 Hz, CH CH3), 1.26 (m, 20H, CH2 of myristyl
chain), 1.62 (m, 2H, CH2 CH2 CO NH of myristyl chain), 2.15
(t, 2H, J = 6.8 Hz, CH2 CH2 CO NH of myristyl chain), 2.19 (t,
1H, J = 6.8 Hz, CH CH2 OCH3), 3.36 (m, 3H, OCH3), 4.06 (t, 1H,
J = 6.45 Hz, CH CH2 OCH3), 4.16 (m, 1H, CH ), 5.62 (d, 1H,
J = 5.84 Hz, NH ).
1H,–CH2 CH2
CH ).
O ), 3.69 (m, 1H, CH2 CH2 O ), 4.1 (m, 1H,
2.4.10. N,O-dimyristyl 4-amino-1-pentanol 12c
m/z (LR-ESI+) C33H66NO3 (M+Na+), found: 546.2 calculated for:
546.9. IR vmax (cm−1): 3304 (N H, amide), 2800–3000 (CH of
myristyl chain), 1732 (C O, ester), 1640 (C O, amide), 1546 (N H,
amide). 1H NMR (400 MHz, CDCl3, ı ppm): ı 0.88 (t, 6H, J = 7.43 Hz,
2x CH2 CH3), 1.14 (d, 3H, J = 6.83 Hz, CH CH3), 1.25 (m, 40H,
2.4.5. 1-N-myristyl 1,2-diaminopropane 8a
CH2 of myristyl chain), 1.53 (m, 4H, CH CH2 CH2 CH2
1.6 (m, 4H, 2x CH2 CH2 CO of myristyl chain), 2.14 (t,
4H, J = 7.08 Hz, CH2 CH2 CO of myristyl chain), 3.68 (m, 2H,
O C),
m/z (LR-ESI+) C17H36N2O (M+H+), found: 285.4, calculated
for: 285.49. 1H NMR (400 MHz, CDCl3, ı ppm): ı 0.88 (t, 3H,
J = 6.67 Hz, CH2 CH3), 1.22 (d, 3H, J = 6.67 Hz, CH CH3), 1.25
CH2 CH2
NH ).
O C), 4.07 (m, 1H, CH ), 5.27 (d, 1H, J = 6.86 Hz,
(m, 20H, CH2
CH2 CH2 CO NH of myristyl chain), 1.96 (s, 2H, NH2),
2.19 (t, 2H, J = 7.3 Hz, CH2 CH2 CO NH of myristyl
of myristyl chain), 1.61 (q, 2H, J = 6.67 Hz,
2.4.11. N-myristyl aminohexanol 14a
chain), 3.21 (st, 1H, J = 5.9 Hz, CH ), 3.27 (qd, 1H, J = 2.6 Hz,
J = 11.8 Hz, CH CH2 NH ), 3.46 (qd, 1H, J = 3.24 Hz, J = 13.6 Hz,
CH CH2 NH ), 7.22 (t, 1H, J = 4.42 Hz, NH ).
m/z (LR-ESI+) C20H41NO2 (M+H+), found: 329.5 calculated
for: 328.56. IR vmax (cm−1): 3385 (O H, alcohol), 3314 (N H,
amide), 2800–3000 (CH of myristyl chain), 1634 (C O, amide),
1534 (N H, amide). 1H NMR (400 MHz, CDCl3,
ı
ppm):
0.88 (t, 3H, J = 7.5 Hz, CH2 CH3), 1.25 (m, 20H, CH2
myristyl chain), 1.51 (m, 2H, CH2 CH2 CO
ı
of
2.4.6. N-myristyl 1-amino-2-propanol 10a
m/z (LR-ESI+) C17H35NO2 (M+H+), found: 286.4, calculated for:
286.48. IR vmax (cm−1): 3100–3500 (O H, alcohol and N H, amide),
2800–3000 (CH of myristyl chain), 1619 (C O, amide), 1571 (N H,
amide). 1H NMR (400 MHz, CDCl3, ı ppm): ı 0.88 (t, 3H, J = 6.55 Hz,
CH2 CH3), 1.19 (d, 3H, J = 5.61 Hz, CH CH3), 1.25 (m, 20H,
CH2 of myristyl chain), 1.64 (m, 2H, CH2 CH2 CO NH
of myristyl chain), 2.21 (t, 2H, J = 7.49 Hz, CH2 CH2 CO NH
of myristyl chain), 2.72 (d, 1H, J = 3.74 Hz, OH), 3.13 (m, 1H,
CH CH2 NH ), 3.44 (m, 1H, CH CH2 NH ), 3.92 (m, 1H,
CH ), 5.95 (s, 1H, NH ).
O
of myristyl
chain), 1.59 (m, 4H, CH2 CH2 CH2 CH2 OH), 2.26 (t, 2H,
J = 7.65 Hz, CH2 CH2 CO OH of myristyl chain), 2.72 (s, 1H,
OH), 3.25 (t, 2H, J = 7.07 Hz,
CH2 CH2 OH), 3.63 (t, 2H,
J = 7.29 Hz, CH2 CH2 NH CO CH2), 5.41 (s, 1H, NH ).
2.4.12. O-myristyl aminohexanol 14b
m/z (LR-ESI+) C20H41NO2 (M+H+), found: 329.5 calculated
for: 328.56. IR vmax (cm−1): 3400 (N H, amine), 2800–3000
(CH of myristyl chain), 1736 (C O, ester), 1544 (N H, amine).
1H NMR (400 MHz, CDCl3, ı ppm): ı 0.88 (t, 3H, J = 7.28 Hz,
2.4.7. N,O-dimyristyl 1-amino-2-propanol 10c
m/z (LR-ESI+) C31H62NO3 (M+Na+), found: 518.6, calculated
for: 518.85. IR vmax (cm−1): 3291 (N H, amide), 2800–3000 (CH
of myristyl chain), 1725 (C O, ester), 1641 (C O, amide), 1557
(N H, amide). 1H NMR (400 MHz, CDCl3, ı ppm): ı 0.88 (t, 6H,
CH2 CH3), 1.25 (m, 20H,
(m, 2H, CH2 CH2 CO
4H, CH2 CH2 CH2 CH2 NH2), 2.28 (t, 2H, J = 7.65 Hz,
CH2 CH2 CO of myristyl chain), 2.81 (s, 2H, NH2),
CH2
of myristyl chain), 1.55
O
of myristyl chain), 1.62 (m,
O

196
F. Le Joubioux et al. / Journal of Molecular Catalysis B: Enzymatic 85–86 (2013) 193–199
3,0
2,5
2,0
1,5
1,0
0,5
0,0
3.64 (t, 2H, J = 6.47 Hz, CH2 CH2 NH2), 4.04 (t, 2H, J = 6.47 Hz,
CH2 CH2
O CO CH2).
2.4.13. N,O-dimyristyl aminohexanol 14c
m/z (LR-ESI+) C34H67NO3 (M+Na+), found: 560.7, calculated
for: 560.93. IR vmax (cm−1): 3298 (N H, amide), 2800–3000
(CH of myristyl chain), 1726 (C O, ester), 1635 (C O, amide),
1547 (N H, amide). 1H NMR (400 MHz, CDCl3,
ı
ppm):
ı
0.88 (t, 6H, J = 6.48 Hz, 2x
CH2 of myristyl chain), 1.5 (m, 4H,
myristyl chain), 1.6 (m, 4H, CH2 CH2 CH2 CH2
CH2 CH3), 1.25 (m, 40H,
CH2 CH2 CO
of
O
C), 2.15
(t, 2H, J = 7.8 Hz, CH2 CH2 CO NH ), 2.29 (t, 2H, J = 7.8 Hz,
CH2 CH2 CO NH ), 3.24 (q, 2H, J = 6.5 Hz, CH2 CH2 NH ),
4.06 (t, 2H, J = 5.9 Hz, CH2 CH2
O CO CH2), 5.4 (s, 1H, NH ).
2.5. Analysis of the ionization state of substrates
-0,04
-0,02
0,00
0,02
0,04
0,06
1/[( )-alaninol] (mM-1)
The ionization state of myristic acid in tert-amyl alcohol was
investigated using infrared spectroscopy analysis. Infrared (IR)
spectra of samples containing 175 mM myristic acid and from 0
to 250 mM alaninol were recorded from 1500 to 1800 cm−1 with
a resolution of 4 cm−1 using a 100 ATR spectrometer (Perkin-
Elmer, United States). Before the interpretation of data, a treatment
(base line correction, smoothing and normalization min–max) was
applied to spectra.
Fig. 1. Reciprocal initial rates of myristic acid conversion (᭹) and ( )-alaninol N-
acylation (ꢀ) versus reciprocal ( )-alaninol concentrations. Reactions were carried
out at 55 ◦C using a fixed concentration of myristic acid (175 mM) and 5 g l−1 of C.
antarctica lipase B in 2 ml of tert-amyl alcohol. The data represent the averages of
triplicate runs whose standard deviations were always lower than 15%.
This salt complex makes the ion forms of both substrates non reac-
tive (NH3 amine form and COO− fatty acid form) and therefore
+
3. Results and discussion
leads to the overestimation of the substrate concentrations that
are really available for the enzyme in the reaction medium. This
ion-pair complex was already described by Maugard et al. [16] as
a limiting factor of the lipase-catalyzed acylation under conditions
where it was less soluble than free substrates. To ascertain this
hypothesis, the composition of the medium, especially the carbonyl
species, was analyzed by infrared spectroscopy at the start of the
reaction, for four concentrations of ( )-alaninol within the range
50–250 mM (Fig. 2). When only myristic acid was solubilized in tert-
In order to explore the chemoselectivity of the enzyme, kinetic
studies were investigated for the acylation of several acyl accep-
tors including various amines, methoxyamine and amino-alcohols
(Table 2), catalyzed by C. antarctica lipase B using myristic acid
(175 mM) as an acyl donor and tert-amyl alcohol as a solvent. From
the LC–MS analysis, the decrease in myristic acid concentration
was always seen to be concomitant with the synthesis of acylated
products. The purification and the structural elucidation by mass
spectroscopy, IR and NMR analyses led to identify the structure of
the acylated products described in Table 2. In absence of enzyme,
no product was detected within 2 days.
The acylation of ( )-alaninol (amino-alcohol 1, Table 2, entry
1) was chosen as the model reaction to study the kinetic behavior
of C. antarctica lipase B when catalyzing the acylation of amino-
alcohols. To achieve this, we determined the apparent kinetic
parameters of the amide and ester synthesis from myristic acid
and ( )-alaninol. As no mono-O-acylation product was detected
under our experimental conditions, systematic analysis of the rates
of myristic acid conversion and ( )-alaninol N-acylation were
conducted, by varying ( )-alaninol concentration. This analysis
revealed Lineweaver–Burk reciprocal plots presented in Fig. 1. The
intercepts of the y-axis and the x-axis gave the values of Vmax,app
and Km,app. The Vmax,app were found to be 4.9 mmol h−1 g−1 for
myristic acid conversion and 4.3 mmol h−1 g−1 for ( )-alaninol N-
acylation. The Vmax,app of O-acylation of the N-acylated product 2a
was extremely low (0.3 mmol h−1 g−1), which explained the sim-
ilarities observed between the reciprocal rate values of myristic
acid conversion and ( )-alaninol N-acylation. The Km,app value of
amyl alcohol, only one carbonyl band was observed at 1710 cm−1
,
corresponding to the acid form. When the concentration of ( )-
alaninol was increased, the carbonyl acid band disappeared in favor
of a band at 1562 cm−1 corresponding to a carboxylate ion. This
additional band demonstrated the formation of an ion-pair com-
plex between myristic acid and alaninol when using an ( )-alaninol
concentration superior or equal to 100 mM.
The kinetic studies resulting from the acylation of other acyl
acceptors (Table 2) were determined using the same method as the
one used above for the acylation of ( )-alaninol. In all kinetic pro-
files, a decrease in initial rates was observed for high concentrations
of amino substrates, most probably due to the formation of an ion-
pair complex between substrates similar to the complex described
above in case of ( )-alaninol acylation. The apparent kinetic param-
eters Km,app and Vmax,app resulting from the acylation of all acyl
acceptors were determined and are given in Table 3.
Firstly, we compared the acylation resulting from a series of
amines (3, 5 and 7) structurally related to amino-alcohol 1 (alani-
nol) with different groups in ␤ position of the amino group (Fig. 3
and Table 2, entries 1–4).
The Km,app of amine 3 ((R)-sec-butylamine) (619 mM; Table 3,
entry 4) was 3-fold and 5-fold higher than the Km,app of amino-
alcohols 1 and (R)-1 (( )-alaninol and (R)-alaninol) (182 mM and
118 mM; Table 3, entries 1 and 2), respectively. This showed a
better affinity of C. antarctica lipase B toward a bifunctional amino-
alcohol rather than a monofunctional amine. On the other hand,
the Km,app values of bifunctional compounds 1 (( )-alaninol), 5
(( )-1-methoxy-2-propylamine) and 7 (( )-1,2-diaminopropane),
structurally related substrates with two carbons between the
two functional groups, were of the same order (Km,app equal
(
)-alaninol was found to be 182 mM.
For high concentrations of amino-alcohol (superior to 100 mM),
a decrease in initial rates was observed. This phenomenon was most
likely due to an inhibitor effect similar to an excess substrate inhi-
bition probably due to an interaction between myristic acid and
the amino group of alaninol. Indeed, the presence of an amino
substrate and a fatty acid in an organic solvent generally leads
to the formation of an ion-pair complex between both substrates,
depending on the acido-basic conditions of the medium [16,18,19].

F. Le Joubioux et al. / Journal of Molecular Catalysis B: Enzymatic 85–86 (2013) 193–199
197
Table 2
Structure of substrates and products resulting from the acylation of various acyl acceptors catalyzed by C. antarctica lipase B using myristic acid as an acyl donor in tert-amyl
alcohol.
Entry
Acyl acceptor
Products
Amide
Ester
Amido-ester
1
–
N-myristyl
alaninol (1)
2-amino-1-propanol
N,O-dimyristyl
2-amino-1-propanol (2c)
(2a)
2
3
–
–
–
(R)-sec-
butylamine (3)
N-myristyl
sec-butylamine (4a)
( )-1-methoxy-
2-propylamine
(5)
–
N-myristyl
1-méthoxy-2-
propylamine (6a)
4
–
–
( )-1,2-diamino
propane (7)
1-N-myristyl
1,2-diaminopropane
(8a)
5
–
isopropanolamine
N-myristyl
1-amino-2-propanol (10a)
(9)
N,O-dimyristyl
1-amino-2-propanol (10c)
( )-4-amino-1-
pentanol
(11)
6
N-myristyl
4-amino-1-pentanol
(12a)
O-myristyl
4-amino-1-pentanol (12b)
N,O-dimyristyl
4-amino-1-pentanol (12c)
6-amino-1-
hexanol (13)
7
N-myristyl
aminohexanol (14a)
O-myristyl
aminohexanol (14b)
N,O-dimyristyl
aminohexanol (14c)
Table 3
Apparent kinetics parameters for the N-acylation and O-acylation of various acyl acceptors using C. antarctica lipase B in tert-amyl alcohol at 55 ◦C. Myristic acid (175 mM)
was used as an acyl donor.
Entry
Acyl acceptor
)-Alaninol 1
(R)-alaninol (R)-1
(S)-alaninol (S)-1
(R)-sec-butylamine 3
Km,app (mM)
Vmax,app of N-acylation (mmol h−1 g−1
)
Vmax,app of O-acylation (mmol h−1 g−1
)
1
2
3
4
5
6
7
8
9
(
182
118
122
619
185
252
103
75
4.3
5.5
5.4
0.2
1.6
7.6
2.5
1.1
1
–
–
–
–
–
–
–
7.3
10.1
(
(
(
(
)-1-Methoxy-2-propylamine 5
)-1,2-Diaminopropane 7
)-Isopropanolamine 9
)-4-Amino-1-pentanol 11
6-Amino-1-hexanol 13
63

198
F. Le Joubioux et al. / Journal of Molecular Catalysis B: Enzymatic 85–86 (2013) 193–199
Fig. 2. IR analysis of mixtures containing 175 mM of myristic acid and various concentrations of ( )-alaninol in tert-amyl alcohol.
2.5 mmol h−1 g−1) > methoxyamine 5 (1.6 mmol h−1 g−1) ꢁ amine
3 (0.2 mmol h−1 g−1). The first substrate diamine 7, which exhibits
the highest Vmax,app values was mono-N-acylated only in position
1. Indeed, only mono-amide 8a was detected (Table 2, entry 4).
The last substrate amine 3 is a monofunctional amine with no
nucleophilic group in ␤-position of an amino group. From these
results, we could thus conclude that the presence of a nucleophilic
group (alcohol group, methoxy group or a second amino group) in
␤-position of the acyl-acceptor amino group was responsible for
the enhancement of Vmax,app of N-acylation of this amino group
and that the more nucleophilic group ( NH2 > OH > OCH3) in
␤-position, the higher the reactivity for the amino group.
to 182 mM, 185 mM and 252 mM; Table 3, entries 1, 5 and 6).
On the contrary, Vmax,app values of N-acylation for compounds
1, 5 and 7 were variable (from 1.6 to 7.6 mmol h−1 g−1), which
showed that C. antarctica lipase B catalytic efficiency was mainly
depending on its differential catalytic activity for these substrates.
Furthermore, in a previous work we have compared the acy-
lation of mono-amine 3 ((R)-sec-butylamine) and a structurally
similar secondary alcohol: (R)-2-butanol. We noticed that the
Vmax,app of O-acylation of (R)-2-butanol was 23-fold higher than
the Vmax,app of N-acylation of amine 3 [20]. These results were in
contrast with those obtained for the acylation of (R)-alaninol, for
which no mono-O-acylation product was detected and a value of
Vmax,app equal to 5.5 mmol h−1 g−1 was obtained for N-acylation.
Moreover, it was shown that the stereochemistry of compounds
did not influence the chemoselectivity during the acylation of
alaninol because similar results were observed for the racemic
Secondly, we compared the acylation of three amino-alcohols
that exhibited a variable carbon chain length between the amino
and alcohol groups: amino-alcohols 1 (two carbons), 11 (four car-
bons) and 13 (six carbons) (Table 2, entries 1, 6 and 7). In terms
of Km,app values, the Km,app of amino-alcohol 11 (75 mM; Table 3,
entry 8) was in the same order than the Km,app of amino-alcohol
13 (63 mM; Table 3, entry 9), whereas the Km,app of amino-alcohol
1 (182 mM; Table 3, entry 1) was higher. This pointed out a better
affinity of C. antarctica lipase B toward long chain amino-alcohols 11
and 13 than for short amino-alcohols 1. On the other hand, the acy-
lation of long chain amino-alcohols 11 and 13 by C. antarctica lipase
B gave Vmax,app of N-acylation of 1.1 and 1 mmol h−1 g−1 (Table 3,
entries 8 and 9), respectively, as short chain amino-alcohol 1 was
(
)-alaninol and the R and S pure enantiomers of alaninol: no
mono-O-acylation product was detected and values of Vmax,app of
N-acylation close to 5 mmol h−1 g−1 were obtained in all cases.
Furthermore, by comparing the acylation of ( )-alaninol 1 (2-
amino-1-propanol) with the acylation of ( )-isopropanolamine 9
(1-amino-2-propanol), for which the positions of the alcohol and
amine functions are interchanged (Table 2, entries 1 and 5), similar
results were obtained: Vmax,app of N-acylation of the same order
were observed, equal to 4.3 mmol h−1 g−1 and 2.5 mmol h−1 g−1
,
N-acylated 4-fold faster (Vmax,app of N-acylation: 4.3 mmol h−1 g−1
;
respectively (Table 3, entries 1 and 7) and no mono-O-acylation
product was detected. These results seem to indicate that the posi-
tion (1 or 2) of the alcohol and amine functions has no impact on
the chemoselectivity of the lipase. Finally, to better understand
the influence of substrate structure on the chemoselectivity of C.
antarctica lipase B, we compared the Vmax,app of mono-N-acylation
of structurally related amines (Fig. 3, compounds 1, 3, 5, 7 and
9, Table 3, entries 1 and 4–7). The order of Vmax,app values was
found to be: diamine 7 (7.6 mmol h−1 g−1) > amino-alcohol 1 and
9 (( )-alaninol and ( )-isopropanolamine) (4.3 mmol h−1 g−1 and
Table 3, entry 1) than long chain amino-alcohols. In contrast,
Vmax,app of O-acylation of 7.3 mmol h−1 g−1 for amino-alcohol 11
and 10.1 mmol h−1 g−1 for amino-alcohol 13 (Table 3, entries 8 and
9) were obtained, whereas no mono-O-acylated product and only
trace amounts of amido-ester 2c were detected during the acyla-
tion of short chain amino-alcohol 1. This was attributed to the fact
that the reaction could take place at the amino group (N-acylation)
and/or alcohol group (O-acylation) of long chain amino-alcohols
11 and 13, giving either N- or O-acylated products 12a, 12b, 14a
and 14b (Table 2, entries 6 and 7), while the mono-O-acylation
of short chain amino-alcohol 1 did not occur. Starting from these
results, we could calculate the chemoselectivity ratio (Eq. (2)) of
the C. antarctica lipase B-catalyzed acylation of long chain amino-
alcohols 11 and 13, which was close to 6.6 and 10.1, respectively.
From these results, we could conclude that the increase in the car-
bon chain length between the alcohol and amino groups of long
chain amino-alcohols was concomitant with the increase in the
chemoselectivity of C. antarctica lipase B for the O-acylation of these
substrates.
Fig. 3. Alaninol (1) and structurally related amines (3, 5, 7).

F. Le Joubioux et al. / Journal of Molecular Catalysis B: Enzymatic 85–86 (2013) 193–199
199
Up to now in the literature, it has been suggested that the
lipase-catalyzed chemoselective N-acylation of short chain amino-
alcohols proceeds through an initial O-acylation followed by
spontaneous O- to N-acyl migration to give the N-acylated prod-
uct [12,13]. According to this hypothesis, the acyl migration would
be slower when the carbon chain length of the amino alcohol
increases, which could be in accordance with the experimental
results obtained for the acylation of amino-alcohols 1, 9, 11 and
13. However, the increase in the N-acylation rate observed for
the acylation of bifunctional compounds, which are structurally
related to alaninol 1 but exhibit no alcohol group in their struc-
ture (1-methoxy-2-propylamine 5 and ( )-1,2-diaminopropane 7)
and thus cannot possibly be O-acetylated, seems to rule out this
hypothesis. On the other hand, several works have demonstrated
that spontaneous acyl migration in 1,2-amino-alcohol substrates
proceeds with a rate much lower than 1 min−1 [21,22]. Thus,
according to the experimental data of alaninol 1, reaction rates
for spontaneous O- to N-acyl migrations are at least one to
two orders of magnitude slower than reaction rates for enzyme-
catalyzed acylation of this amino-alcohol (N-acylation of alaninol:
Vmax,app = 4.3 mmol h−1 g−1 which corresponds to kcat = 50 min−1).
Under our experimental conditions, if lipase-catalyzed N-acylation
of alaninol only occurred through an initial O-acylation followed
by spontaneous O- to N-acyl migration, the O-acylated intermedi-
ate would accumulate and reach a steady-state concentration up
to values within the range 1–100 mM. But, no O-acylated prod-
uct was found despite the detection limit of the LC/MS-ES system
being 2 ␮M. These data show that the hypothesis of acyl migra-
tion alone does not enable to explain experimental results resulting
from the acylation of structurally related amines (1, 3, 5 and 7) and
amino-alcohols with various carbon chain length (1, 11 and 13).
For this reason, we formulated the following postulate: the pres-
ence of a nucleophilic group in ␤-position of the amino group of
the acyl acceptor amine would result in the enhancement of the
Vmax,app of N-acylation of this amino group. This might be due to
the formation of an intramolecular interaction between the amino
group and the nucleophilic group located in ␤-position, which is
strengthened by the fact that this interaction could not occur for
long chain amino-alcohols 11 and 13, considering the longer dis-
tance between both functional groups, giving as a result a decrease
in the Vmax,app of N-acylation. In order to confirm and develop
this new postulate, molecular modeling studies are currently in
progress to precisely understand the C. antarctica lipase B-catalyzed
N-acylation mechanism using a methoxy-amine, a di-amine, a short
chain amino-alcohol or a long chain amino-alcohol as an acyl accep-
tor.
N-acylation and thus the enzyme activity. Moreover, the crucial
role of the carbon chain length between the alcohol and amino
groups was highlighted in the C. antarctica lipase B-catalyzed acy-
lation of amino-alcohols. The Vmax,app of N-acylation was indeed
improved when the carbon chain included two carbons (alaninol 1
and isopropanolamine 9) whereas the Vmax,app of O-acylation was
improved when the carbon chain included four carbons or more
(4-amino-1-pentanol 11 and 6-amino-1-hexanol 13). The present
investigation demonstrated the great influence of substrate struc-
ture on the chemoselectivity of C. antarctica lipase B, providing new
insights for the selective synthesis of amides or esters produced
from the acylation of bifunctional substrates.
Finally, the ability to understand and control the chemoselec-
tivity of C. antarctica lipase B, apart from its interest in the specific
acylation of bifunctional substrates, constitutes a promising enzy-
matic way to acylate other heterofunctional compounds such as
precursors of ceramide synthesis (e.g. sphingoid bases or other
amino-polyols) or precursors of amino-acid based surfactant syn-
thesis (e.g. amino-acids or peptides).
Acknowledgments
This study was supported by the Centre National de la Recherche
Scientifique and the French ANR (National Research Agency)
through the EXPENANTIO project (CP2P program: Chimie et
Procédés pour le Développement Durable).
References
[1] M. Bakke, M. Takizawa, T. Sugai, H. Ohta, J. Org. Chem. 63 (1998) 6929–6938.
[2] L. Lassalle, F. Yvergnaux. FR Patent 2,855,823 (2004).
[3] H. Garg, N. Francella, K.A. Tony, L.A. Augustine, J.J. Barchi Jr., J. Fantini, A. Puri,
D.R. Mootoo, R. Blumenthal, Antiviral Res. 80 (2008) 54–61.
[4] M. Fillet, M. Bentires-Alj, V. Deregowski, R. Greimers, J. Gielen, J. Piette, V. Bours,
M.P. Merville, Biochem. Pharmacol. 65 (2003) 1633–1642.
[5] J.F. Molina, Household Pers. Care Today 2 (2008) 12–15.
[6] E.L. Soo, A.B. Salleh, M. Basri, R.N.Z.R.A. Rahman, K. Kamaruddin, J. Biosci. Bio-
eng. 95 (2003) 361–367.
[7] A. Singh, J.D. Van Hamme, O.P. Ward, Biotechnol. Adv. 25 (2007) 99–121.
[8] T. Maugard, M. Remaud-Simeon, D. Petre, P. Monsan, Tetrahedron 53 (1997)
5185–5194.
[9] M. Fernandez-Perez, C. Otero, Enzyme Microb. Technol. 33 (5) (2003) 650–660.
[10] M. Fernandez-Perez, C. Otero, Enzyme Microb. Technol. 28 (6) (2001) 527–536.
[11] V. Gotor, R. Brieva, F. Rebolledo, J. Chem. Soc., Chem. Commun. (1988) 957–958.
[12] T. Furutani, M. Furui, H. Ooshima, J. Kato, Enzyme Microb. Technol. 19 (8) (1996)
578–584.
[13] L.T. Kanerva, E. Vänttinen, T.T. Huuhtanen, M. Dahlqvist, Acta Chem. Scand. 46
(11) (1992) 1101–1105.
[14] O. Torre, V. Gotor-Fernandez, V. Gotor, Tetrahedron: Asymmetry 17 (5) (2006)
860–866.
[15] S. Tawaki, A.M. Klibanov, Biocatal. Biotransform. 8 (1993) 3–19.
[16] T. Maugard, M. Remaud-Simeon, D. Petre, P. Monsan, Tetrahedron 53 (1997)
7587–7594.
[17] A. Fersht, Enzyme Structure and Mechanism, 2nd ed., W.H. Freeman and Com-
pany, New York, 1985, pp. 98–120.
4. Conclusion
[18] E. Husson, C. Humeau, F. Blanchard, X. Framboisier, I. Marc, I. Chevalot, J. Mol.
Catal. B: Enzym. 55 (2008) 110–117.
[19] M. Fernandez-Perez, C. Otero, Enzyme Microb. Technol. 28 (2001) 527–536.
[20] F. Le Joubioux, O. Achour, N. Bridiau, M. Graber, T. Maugard, J. Mol. Catal. B:
Enzym. 70 (2011) 108–113.
In this work, we investigated the C. antarctica lipase B-catalyzed
acylation of various amines and amino-alcohols as acyl acceptors,
using myristic acid as an acyl donor, and showed that the presence
of a nucleophilic group ( NH2 or OH or OCH3) in ␤-position
of the amino group of the acyl acceptor enhances the Vmax,app of
[21] R. Greenhalgh, R.M. Heggie, M.A. Weinberger, Can. J. Chem. 41 (1963) 1662.
[22] S.P. Markey, T. Dudding, T.-C.L. Wang, J. Lipid Res. 41 (2000) 657.