Pseudomonas stutzeri lipase: A useful biocatalyst for aminolysis reactions

Article abstract of DOI:10.1039/c1gc15160f

The use of Pseudomonas stutzeri lipase (PSL) as a biocatalyst for aminolysis reactions with bulky substrates has been investigated. PSL compared favorably to Novozym 435 (immobilized Candida antarctica lipase B, NOV435) in the aminolysis of various bulky methyl esters and amines. While NOV435 demonstrated a higher rate of aminolysis with methyl 2-phenylpropionic acid as the acyl donor, PSL outperformed NOV435 with secondary amines as the nucleophile. Methanol inhibition and a low affinity for bulky acyl donors were found to be the two main reasons for relatively low rates in the PSL-catalyzed aminolysis reactions. It was demonstrated that the use of molsieve 4A had a significant effect on the aminolysis rate and amide yield, since it enabled the effective removal of the inhibiting methanol from the reaction mixture.

Full text of DOI:10.1039/c1gc15160f

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2011, 13, 1791
www.rsc.org/greenchem
PAPER
Pseudomonas stutzeri lipase: a useful biocatalyst for aminolysis reactions
a
a
a
a
b
b
S. van Pelt,* R. L. M. Teeuwen, M. H. A. Janssen, R. A. Sheldon, P. J. Dunn, R. M. Howard,
c
c
c
R. Kumar, I. Mart ´ı nez and J. W. Wong
Received 11th February 2011, Accepted 18th March 2011
DOI: 10.1039/c1gc15160f
The use of Pseudomonas stutzeri lipase (PSL) as a biocatalyst for aminolysis reactions with bulky
ꢀ
R
substrates has been investigated. PSL compared favorably to Novozym 435 (immobilized
Candida antarctica lipase B, NOV435) in the aminolysis of various bulky methyl esters and
amines. While NOV435 demonstrated a higher rate of aminolysis with methyl 2-phenylpropionic
acid as the acyl donor, PSL outperformed NOV435 with secondary amines as the nucleophile.
Methanol inhibition and a low affinity for bulky acyl donors were found to be the two main
reasons for relatively low rates in the PSL-catalyzed aminolysis reactions. It was demonstrated
that the use of molsieve 4A had a significant effect on the aminolysis rate and amide yield, since it
enabled the effective removal of the inhibiting methanol from the reaction mixture.
3
Introduction
shock-sensitive reagents. None of the N-acylation reactions
reviewed in the survey were catalytic and therefore it is not
surprising that there is still a pressing need for the development
of catalytic, environmentally friendly N-acylation processes.
Recently, some relatively efficient catalysts have been devel-
In 2005, the American Chemical Society Green Chemistry In-
stitute Pharmaceutical Roundtable (ACS GCIPR, Roundtable)
was founded with the mission to promote the implementation of
green chemistry and green engineering in the global pharmaceu-
tical industry. During a brainstorming exercise and a subsequent
voting process of the Roundtable, amide formation was identi-
fied as one of the most utilized and problematic syntheses in the
4–6
oped for N-acylation reactions, such as boron-based catalysts,
a heterogeneous silica catalyst, and a heterogeneous sulfated
tungstate catalyst. Amide bond formation can also be catalyzed
by enzymes. Primary amides can be formed by lipase-catalyzed
ammoniolysis reactions of carboxylic acids and esters with
ammonia, or by a hydration reaction of nitriles catalyzed
by nitrile hydratases, or (as a side reaction) by nitrilases.
Aminolysis reactions to form secondary and tertiary amides
can be catalyzed by enzymes, such as proteases, amidases,
acylases and lipases. Lipases have the advantage of having
a high stability, relatively broad substrate specificity and, in
contrast to proteases, acylases and amidases, lipases do not
readily mediate amide hydrolysis. Lipase-catalyzed aminolysis
of carboxylic acids and esters with an amine is therefore a
7
8
1
pharmaceutical industry. More detailed information is available
9
from a recent survey of the reactions used for drug candidates
prepared by three leading pharmaceutical companies. It was
found that acylations comprised 12% of all used reactions and
that N-acylation reactions to the amide were used in 84 (65%)
10
11
2
of the 128 drug candidates reviewed.
12
The majority of the N-acylation methods currently used in
the pharmaceutical industry are based on an acid chloride
intermediate (44%), employing either thionyl or oxalyl chloride,
or are carried out by the use of coupling reagents (36%), such as
13
1
-ethyl-3-(3-dimethylaminopropyl)-carbodiimide hydrochloride
14,15
well-known reaction
but has so far only been employed in
(
EDCI·HCl), 1-propylphosphonic acid cyclic anhydride, and
16
industry for the kinetic resolution of chiral amines, where bulky
amines are reacted with relatively simple esters. The industrial
application of lipase-mediated amine acylation potentially offers
a general method for amide synthesis under mild conditions, but
is still a relatively unexplored field. The limited scope, as well as
the low activity of commercially available enzymes for bulky
substrates, is the main problem.
During screening efforts in our lab, Pseudomonas stutzeri
lipase (PSL), together with Novozym 435, were found to be by
far the most active biocatalysts for the aminolysis of methyl 3-
phenylpropionate and cyclohexylamine. In contrast to Novozym
435, PSL-catalyzed aminolysis reactions have only sporadically
1
N,N¢-carbonyldiimidazole (CDI). In general, these methods
have several disadvantages, such as toxic and/or corrosive by-
products, poor atom economy and, in some cases, they employ
a
CLEA Technologies, Delftechpark 34, 2628, XH, Delft, The
Netherlands. E-mail: s.vanpelt@clea.nl; Fax: +31157600301;
Tel: +31157600300
b
Pfizer Worldwide Research and Development, Department of Chemical
Research and Development, Ramsgate Road, Kent, United Kingdom,
CT13 9NJ
c
Pfizer Worldwide Research and Development, Chemical Research and
Development, Eastern Point Road, Groton, CT, 06340, USA
This journal is © The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1791–1798 | 1791

View Article Online
1
7
been reported in the literature. Therefore, the performance of
PSL in the aminolysis of a group of different bulky methyl esters
and amines is investigated in this manuscript.
Certain other hydrolases were also considered for this study
but were not included, based on their limited substrate scope.
Although the cutinases from Humicola insolens and Fusarium
solani pisi are known to have a wide binding cleft for both
19
the alcohol and the acyl donor, the immobilized cutinase
Results and Discussion
ꢀ
R
from Novozymes (Novozym 51032) was an inferior aminolysis
biocatalyst for these substrates by far, in comparison to NOV435
and PSL (data not shown). Similarly, Candida antarctica lipase
A might be a more suitable enzyme for aminolysis reactions
with 6 and possibly 7. However, this C. antarctica lipase isoform
did not show any activity in the aminolysis reactions of esters
I–IV.
Substrate scope
In order to investigate the substrate scope of PSL-catalyzed
aminolysis, reactions were carried out with various bulky methyl
esters and amines (Fig. 1). The acyl donors were activated
in the form of an ester to prevent complications due to salt
formation between the amine and the free carboxylic acid. In
order to keep the E factor of the transformation as low as
possible, only methyl esters were used. As a comparison, the
exact same reactions were carried out in parallel using Novozym
The differences in substrate specificities between NOV435 and
PSL are obvious (Table 1). NOV435 clearly outperforms PSL in
the aminolysis reactions with methyl 2-phenylpropionate (III)
as the acyl donor, while PSL is a more effective biocatalyst for
aminolysis reactions with secondary amines and with piperidine
in particular. Both PSL and NOV435 were unable to accept
tert-butylamine (6), dipropylamine (7), methyl 2-methyl-2-
phenylpropionate (V), and methyl 3-methyl-2-phenylbutanoate
18
4
35 (NOV435).
(
VI) as substrates. Furthermore, both of the lipases clearly had
difficulties in converting esters that are bulky or branched at the
a-position to the carbonyl moiety.
Several reports in the literature describe the successful directed
evolution of lipases for an increased tolerance of substrate
bulkiness and branching at the a-position to the carbonyl
moiety. This increased tolerance is made possible by replacing
bulky amino acids in the acyl binding pocket close to the catalytic
triad by smaller ones, causing an enlargement of this binding
20,21
pocket to accommodate a larger group.
This technique could
enable the production of a lipase with a substrate scope more
suitable for bulky-to-bulky aminolysis reactions than the lipases
that are commercially available today.
Enantioselectivity
Since 1-phenylethylamine (5) is chiral, the enantiomeric ratio
(
E) of the PSL-catalyzed aminolysis reaction of ester II with
amine 5 was calculated by determining the Vmax and K values
m
for aminolysis reactions with (S)-, and (R)-5 (Fig. 2). Although
PSL followed Kazlauskas’ rule and was clearly (R)-selective for
5, the selectivity was rather modest. No beneficial effect on
Fig. 1 The different methyl esters and amines subjected to aminolysis
catalyzed by PSL and NOV435.
the enantioselectivity of PSL was found when the reaction was
carried out at 21 C rather than the standard 50 C.
◦
◦
-1
-1
Table 1 Rate of aminolysis (mmol min g enzyme ), and conversion (% in brackets) to the amide after 20 h in the aminolysis reaction of 0.1 M
◦
ester with 0.15 M amine using 60 mg PSL or NOV435 at 50 C with 100 mg molsieve 4A in 1 mL MTBE (methyl tert-butyl ether). Unless stated
otherwise, the rates of aminolysis were determined after 30 min
b
I
II
III
IV
a
Amine
PSL
NOV435
PSL
NOV435
PSL
NOV435
PSL
NOV435
a
a
a
1
2
3
4
~0 (<1)
0 (0)
1.7 (73)
0 (0)
2.5 (59)
7.4 (98)
18.4 (99)
11.2 (99)
8 (78)
0.5 (20)
9.4 (99)
1 (34)
8.1 (99)
8.6 (84)
1.1 (55)
0 (0)
0 (0)
0.2 (13)
0.1 (7)
1.1 (43)
0.4 (15)
2.7 (60)
1.7 (53)
0.08 (4)
0.03 (2)
0.07 (5)
0.04 (3)
4 (52)
0.5 (33)
1 (51)
0.9 (43)
0.4 (28)
1.6 (67)
a
a
0.05 (1)
0.2 (16)
3.2 (73)
1.2 (52)
3.9 (57)
1.4 (58)
0.6 (45)
n.d.
b
a
5
3 (52)
c
c
c
c
c
c
8
n.d.
n.d.
0.9 (25)
n.d.
n.d.
n.d.
a
b
c
Rates of aminolysis determined after 20 h of reaction. Racemic mixture. Not determined.
1
792 | Green Chem., 2011, 13, 1791–1798
This journal is © The Royal Society of Chemistry 2011

View Article Online
Fig. 2 Initial aminolysis rates as a function of the concentration of (R)-, and (S)-5. The initial rates were determined by measuring the amide II5
concentration after 10 and 20 min. in the aminolysis reaction with ester II (1 M) in 1 mL MTBE containing 250 mg molsieve 4A at a temperature of
◦
5
0 C. The table shows the E-value of PSL for the aminolysis of ester II with amine 4 and the parameters that were used to calculate it.
Although PSL did not show a high enantioselectivity in
the aminolysis reaction of ester II with amine 5, several
reports on the high selectivity of PSL in esterification and
22–25
hydrolysis reactions can be found in the literature.
However,
in these enantioselective esterifications bulky secondary alcohols
were used, such as 1,2-diarylethanols and benzoins. A highly
enantioselective aminolysis with structurally different amines is,
therefore, a possibility.
Reaction kinetics
The PSL-catalyzed aminolysis of methyl 3-phenylpropionate (II)
with cyclohexylamine (4) in MTBE was chosen as the model
reaction for studying the kinetics of the aminolysis reaction
Fig. 3 The aminolysis of ester II (0.1–2 M) with amine 4 (0.1–3 M),
catalyzed by 10 mg PSL at 50 C in 1 mL MTBE containing 25 mg
◦
(
Scheme 1).
molsieve 4A. Aminolysis rates were determined after 60 min with (᭜) a
varying ester II concentration and an amine 4 concentration that is 50%
higher than the ester concentration or (ꢀ) a fixed ester II concentration
of 1 M and a varying amine 4 concentration.
Effect of substrate concentration on the aminolysis rate
In order to investigate the effect of the substrate concentration
on the PSL-catalyzed aminolysis reaction, aminolysis rates were
determined with different concentrations of ester II and amine
high amine and ester concentrations indicate that PSL did not
suffer from substrate inhibition.
4
(Fig. 3).
It was observed that PSL had a very low affinity for ester
II (K
amine 4 (K
m
> 400 mM) and a substantially higher affinity for
< 100 mM). These results indicate that the
Effect of molsieve 4A
m
formation of the acyl-enzyme intermediate is rate-limiting. This
is not surprising since the biological function of lipases is to
catalyze the hydrolysis of triglycerides to the corresponding
fatty acids and glycerol. Therefore lipases have a rather narrow,
hydrophobic active site, which is reflected in their preference
for linear carboxylic acids. However, far more variation of
alcohol structure is tolerated by lipases, or in this case, of amine
Although all reactants and the solvent (MTBE) were dried
extensively over activated molsieve 3A or 4A before use, the
addition of molsieve 4A to the reaction mixture was still
necessary to prevent substantial ester hydrolysis from taking
place (Fig. 4). Therefore, it is likely that most of the water
necessary for this observed enzymatic ester hydrolysis originated
from the PSL enzyme powder. Drying the PSL enzyme powder
in vacuo over phosphorous pentoxide did not have an effect on
14
structure. Furthermore, the relatively high rates measured at
Scheme 1 The aminolysis of methyl 3-phenylpropionate (II) with cyclohexylamine (4), catalyzed by PSL.
This journal is © The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1791–1798 | 1793

View Article Online
◦
Fig. 4 The aminolysis of ester II (0.1 M) with amine 4 (0.15 M), catalyzed by 60 mg PSL at 50 C in 1 mL MTBE containing (a) 0 mg and (b) 25 mg
molsieve 4A.
◦
Fig. 5 (a) The aminolysis of ester II (0.1 M) with amine 4 (0.15 M), catalyzed by 60 mg PSL at 50 C in 1 mL MTBE containing different amounts
of molsieves; (b) PSL rates of aminolysis were measured after 30 min in reactions using different amounts of molsieves.
the enzyme activity but appeared to result in only the partial
removal of water from this biocatalyst preparation.
concentrations, although in the case of ethanol and n-butanol,
the conversion of ester II continued at an increased rate as the
result of a transesterification reaction of ester II with these
alcohols. The inhibiting effect of methanol is therefore most
likely caused by the competition of methanol with amine 4 as a
nucleophile. Since the transesterification of methanol and ester
II leads to ester II and methanol, this competing reaction is not
evident from following the reaction by HPLC analysis. When
Besides preventing the hydrolysis of ester II, the amount of
molsieve 4A that was added to the reaction had a considerable
effect on the aminolysis rate (Fig. 5). The addition of 250 mg
mL- molsieve 4A compared to 25 mg mL-1 molsieve 4A led to
an increase in the aminolysis rate after 30 min of approximately
1
1
50%. More interestingly, the reaction time necessary to reach
complete conversion of the ester was drastically reduced when
using 250 mg molsieve 4A instead of 25 mg. Besides adsorbing
4
butyl 3-phenylpropionate (IIC ) was used in the aminolysis
with amine 4 under the exact same conditions as with ester
II, molsieve 4A lost its positive effect on the overall rate of
aminolysis, since the produced butanol was unable to enter the
pores of the molsieves and therefore accumulated in the reaction
mixture (Fig. 6b).
˚
water (2.63 A) from the reaction mixture, molsieve 4A will most
26
˚
likely also adsorb methanol (3.8 A), which is a by-product
of the enzymatic aminolysis reaction. This fact prompted us to
investigate the influence of methanol on the rate of aminolysis.
In order to determine the extent of the competitive inhibition
of methanol on the PSL-catalyzed aminolysis reaction, initial
rate experiments were carried out without the presence of
molsieve 4A and with different concentrations of methanol
(Fig. 7a). These initial rate experiments demonstrated that PSL
lost approximately 50% of its initial rate of aminolysis at a
methanol concentration of approximately 25 mM and an amine
concentration of 1100 mM, due to competitive inhibition.
The addition of 250 mg of molsieve 4A to the reaction mixture
led to a more stable initial rate, because the molsieves effectively
removed the competing methanol (Fig. 7b). However, the use of
such high amounts of molsieve 4A resulted in a lower initial rate
The inhibiting effect of methanol
The inhibiting effect of methanol became evident when this
alcohol was added in different concentrations to an aminolysis
reaction in progress (Fig. 6a). By adsorbing the methanol, the
molsieves effectively removed this inhibitor from the reaction
mixture and thus increased the overall rate of aminolysis.
When different amounts of ethanol or n-butanol were added
to the reaction mixture instead of methanol, a similar effect
was observed (data not shown). Amide formation decreased
or even stopped after addition of these alcohols in different
1
794 | Green Chem., 2011, 13, 1791–1798
This journal is © The Royal Society of Chemistry 2011

View Article Online
◦
Fig. 6 (a) The aminolysis of ester II (0.1 M) with amine 4 (0.15 M), catalyzed by 60 mg PSL at 50 C in 1 mL MTBE containing 25 mg molsieve 4A.
After 1 h of reaction, MeOH is added to the reaction mixture to a 0 M, 0.1 M, and 1 M total concentration. (b) The aminolysis of ester II (0.1 M)
◦
and ester IIC
4
(0.1 M) with amine 4 (0.15 M), catalyzed by 60 mg PSL at 50 C in 1 mL MTBE containing 100 mg molsieve 4A.
◦
Fig. 7 The initial aminolysis rate of ester II (1 M) with amine 4 (1.1 M), catalyzed by 7 mg PSL at 50 C in 1 mL MTBE in the presence of different
concentrations of methanol, without molsieves or with 250 mg molsieve 4A in the reaction mixture. Initial rates were determined by measuring the
concentration of amide II4 after (a) 10 min of reaction or after (b) 10, 20, 30, and 60 min of reaction. The numbers above the bars indicate the
theoretical amount of methanol (mM) in the reaction mixture at the time of sampling.
-
1
-1
after 10 min (71 mmol min g PSL ) compared to the initial rate
Reaction engineering
-
1
-1
without the presence of molsieves (92 mmol min mg PSL ). It
is likely that the extremely dry environment that was created
by the high amount of molsieves had a negative effect on PSL
activity, because structurally-important water was stripped at an
From the studies on the reaction kinetics of the PSL-catalyzed
aminolysis of ester II and amine 4 it was concluded that a high
concentration of ester II, as well as a high concentration of
molsieve 4A, had a positive effect on the overall rate of the
aminolysis reaction. To demonstrate this, a reaction was carried
out on a slightly larger scale with a large amount of molsieves
and a high concentration of ester II (Fig. 8).
27
increased rate from the enzyme mantle. The effect of activated
molsieve 5A on the PSL-catalyzed aminolysis reaction of ester
II with amine 4 was also studied. However, molsieve 5A was
found to be less effective than molsieve 4A (data not shown).
Although methanol has a substantial inhibiting effect on the
aminolysis reaction, no product inhibition was observed for
the amide (results not shown). This indicated that the PSL-
catalyzed aminolysis reaction of ester II and amine 4 is a
relatively fast reaction, as long as methanol is effectively removed
from the reaction mixture. The use of vinyl or 2-propenyl esters
could circumvent the problem of competitive inhibition, since
the alcohols released during aminolysis would tautomerize to
28
acetaldehyde and acetone, respectively. However, a potential
drawback of using esters with highly activated alcohol groups,
such as vinyl esters, as well as halogen-ethyl or methyl esters,
oxime esters, and anhydrides, is that these esters may also react
with the amine in the absence of the lipase. Using a more
sterically hindered alcohol moiety could result in less enzyme
inhibition compared to methanol, but would likely also result in
a decreased rate of aminolysis.
Fig. 8 The aminolysis of ester II (1 M) with amine 4 (1.1 M), catalyzed
by 400 mg PSL at 50 C in 8 mL MTBE containing 2000 mg molsieve
4A.
◦
This journal is © The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1791–1798 | 1795

View Article Online
After studying this reaction, the two main limiting factors
were identified. The first limiting factor is product crystalliza-
tion. The reaction in Fig. 8 was not terminated at a final amide
PSL proved to be active in all three of the alternative
solvents, although to a lesser extent than in MTBE. Amide II4
is substantially more soluble in MeTHF and PDME than in
MTBE and TBEE. Both MeTHF and PDME could be suitable
solvents for the removal of methanol by distillation or by using
-
1
II4 concentration of 675 mM (156 g L ) because the PSL had
lost its activity, but rather because of product crystallization.
The crystallization of amide II4 commenced at a concentration
of approximately 250 mM when the reaction mixture was
allowed to cool down to room temperature. At an amide II4
an N purge.
2
Another possible alternative was to carry out the PSL-
catalyzed aminolysis reaction without solvent. By carrying out
the reaction in this way, 1600 mM of amide II4 was produced
before crystallization made further conversion impossible. In
this regard, a mixture of ester II and amine 4 seemed to be the
best solvent for amide II4. The high concentration of the ester
also resulted in a high rate of aminolysis. Although not practical
for these two substrates, the solvent-free aminolysis reaction
could be of use for the production of liquid amide products,
such as II1 and II3. Where the boiling points of the selected
amine and ester are high enough, methanol could be removed
◦
concentration of >600 mM, crystallization at 50 C made
normal stirring impossible and the reaction mixture solidified.
Furthermore, although the use of molsieve 4A can be an effective
method to remove methanol from the reaction mixture during
the reaction, it is not practical to use extremely high relative
loadings of molsieves in reactions that are carried out at high
concentrations of substrate on a large scale. The large amount
of molsieves necessary in such reactions is therefore the second
limitation.
31
from the reaction mixture by applying a vacuum.
Effect of solvent and temperature
Carrying out the PSL-catalyzed aminolysis reactions at
◦
temperatures above 50 C may not only result in a higher
The PSL-catalyzed aminolysis of ester II and amine 4 was
studied in solvents other than MTBE for two reasons.
aminolysis rate and greater solubility of the product amides,
but could also enable the removal of MeOH by distillation or
1
. Higher amide concentrations could be reached if the PSL-
an N purge. Encouragingly, PSL was found to be reasonably
2
catalyzed aminolysis reaction were to be carried out in a solvent
in which amides are more soluble.
stable at elevated temperatures. The residual activity of PSL after
◦
incubation over 43 h at 70 C in MeTHF was 67% (Fig. 10). This
2
. In industry, alternative methods for the in situ removal
of methanol to the use of molsieves would be the removal
of methanol by distillation, by the use of an N purge, or by
result indicates that methanol removal by in situ distillation or
an N
2
purge may be feasible. Furthermore, work is currently
2
29
being carried out to immobilize PSL, which could improve the
pervaporation using modified silica membranes. These meth-
ods are more suitable for scale-up than the use of molsieves. Since
MTBE has a boiling point below MeOH and the MTBE–MeOH
system exhibits a minimum-boiling azeotrope at atmospheric
32
enzyme stability considerably.
The results of the experiments on the effect of elevated
temperatures on the initial rate of the PSL-catalyzed aminolysis
-
1
30
reaction with 60 mg mL enzyme in MeTHF were significantly
influenced by methanol inhibition (Fig. 11a). The increase in the
pressure comprising only 31.5 mol% (14.3 wt %) MeOH,
it might be more convenient to consider alternative reaction
◦
initial rate of aminolysis caused by a 10 C reaction temperature
solvents that have a higher boiling point than methanol (bp:
◦
increase was much less than the factor of 2 that can generally
be expected from the Arrhenius equation. Therefore, it was not
surprising that the increase of the initial rate of aminolysis at
elevated temperatures was higher when greater levels of molsieve
6
4.7 C).
◦
2
-Methyltetrahydrofuran (MeTHF, bp: 78 C), tert-butyl
◦
ethyl ether (TBEE, bp: 72–73 C), and polyglycol DME 500
PDME, bp: >250 C) were tested as alternative solvents to
◦
(
4
A were used in the reaction. Since the lipase is used as a
MTBE. The rate of aminolysis of PSL for ester II with amine 4,
as well as the solubility of amide II4 at room temperature, was
determined for the various solvents (Fig. 9).
suspended solid in these reactions, mass transfer limitations may
influence the rate of aminolysis as well.
◦
Fig. 9 (a) Aminolysis of ester II (0.1 M) with amine 4 (0.15 M), catalyzed by 60 mg PSL at 50 C in 1 mL MTBE, MeTHF, TBEE, and PDME
containing 100 mg molsieve 4A. (b) Maximum solubility of amide II4 at room temperature in MTBE, MeTHF, TBEE, and PDME.
1
796 | Green Chem., 2011, 13, 1791–1798
This journal is © The Royal Society of Chemistry 2011

View Article Online
Experimental
Materials
ꢀ
R
Lipase TL from Pseudomonas stutzeri was purchased from
Meito Sangyo (Nagoya, Japan) and Novozym 435 (immobilized
Candida antarctica lipase B) was kindly supplied by Novozymes
(
(
Bagsvaerd, Denmark). Activated molecular sieve 4A powder
2.5 mm) and activated molecular sieve 3A pellets (1.6 mm) were
obtained from Sigma–Aldrich (Zwijndrecht, the Netherlands).
Molecular sieve 5A powder (<50 mm) was received from Acros
Organics (Geel, Belgium).
Fig. 10 The residual activity of PSL after an incubation of 20 and 43 h
Chemicals
◦
in MeTHF at 50 and 70 C. The residual activities were determined in
◦
the aminolysis of ester II (1.0 M) with amine 4 (1.1 M) at 50 C in 1 mL
Methanol (99.9%, Roth), methyl tert-butyl ether (≥99.5%, Roth),
MTBE with 250 mg molsieve 4A.
2
-methyltetrahydrofuran (99+%, Acros Organics), polyethylene
glycol dimethyl ether (M : 500, Clariant), ethyl tert-butyl ether
99%, Aldrich), 1,3-dimethoxybenzene (≥98%, Aldrich), 1,3,5-
w
(
The amide II4 concentrations reached after 21 h and 44 h of
reaction in MeTHF were similarly dependent on the amount
of molsieve 4A that was added to the reaction (Fig. 11b).
Furthermore, it appears that after 44 h the product concen-
trimethoxybenzene (≥99%, Fluka), N-methylbenzylamine (97%,
Sigma-Aldrich, 1), isopropylamine (99%, Sigma-Aldrich, 2),
piperidine (99%, Sigma-Aldrich, 3), cyclohexylamine (99%,
Sigma-Aldrich, 4), DL(±)-a-methylbenzylamine (≥98%, Fluka,
◦
◦
tration at 70 C was lower than at 50 C, indicating that
increased initial activity at higher temperatures may be less influ-
ential on final conversion than temperature-dependent enzyme-
deactivation.
5
), (S)-(-)-a-methylbenzylamine (>99% (sum of enantiomers),
Sigma-Aldrich, (S)-5), (D)-(+)-a-methylbenzylamine (99% ee,
9+%, Acros Organics, (R)-5), tert-butylamine (99%, Acros
9
Organics, 6), dipropylamine (≥99%, Sigma-Aldrich, 7), ani-
line (≥99.5%, Sigma-Aldrich, 8), methylcyclohexanecarboxylate
(
98%, Sigma-Aldrich, I), methyl 3-phenylpropionate (98%, Alfa
Conclusion
Aesar, II), methyl benzoate (99%, Acros, IV), cyclohexanecar-
boxylic acid (98%, Aldrich), (±)-2-phenylpropionic acid (98%,
Alfa Aesar), benzoic acid (sublimed 99+%, Aldrich), 2-methyl-
2-phenylpropionate (Alfa Aesar), 3-methyl-2-phenylbutanoate
(97%, Alfa Aesar), cyclohexanecarbonyl chloride (98%, Sigma-
Aldrich), 3-phenylpropionyl chloride (98%, Sigma-Aldrich) and
2-phenylpropionylchloride (≥95%, Sigma-Aldrich) were used in
the experiments as received, without additional purification.
We have shown that lipase-catalyzed ester aminolysis has definite
potential as a green and viable alternative to the currently used
non-catalytic methods for amide formation in the pharmaceu-
tical and fine chemical industries. In particular, we have shown
that Pseudomonas stutzeri lipase is a useful aminolysis catalyst
with a relatively broad substrate scope, with respect to both
the amine and the ester. This brings the possibility of lipase-
mediated amine acylation as a green and general method for
amide synthesis under mild conditions a step closer to industrial
practice. Moreover, we expect that the application of modern
molecular biology tools, such as directed evolution, will broaden
the scope further to include even bulky ester substrates that are
not accommodated by wild-type lipases.
4
Butyl 3-phenylpropionate (IIC ), methyl 2-phenylpropionate
(III), methyl 2-methyl-2-phenylpropionate (V), and methyl 3-
methyl-2-phenylbutanoate (VI) were synthesized from their cor-
responding acids by following a procedure from the literature.
3-Phenylpropionic acid (AII) was synthesized by a reaction of
the corresponding acid chloride with water. All amide reference
33
Fig. 11 (a) The initial rates of aminolysis of PSL in the aminolysis of ester II (1 M) with amine 4 (1.1 M), catalyzed by 60 mg PSL at 50, 60, and
◦
7
(
0
C in 1 mL MeTHF containing 100 and 250 mg molsieve 4A. (b) Production of amide II4 after 21 and 44 h in aminolysis reactions of ester II
◦
1 M) with amine 4 (1.1 M), catalyzed by 60 mg PSL at 50, 60, and 70 C in 1 mL MeTHF containing 100 and 250 mg molsieve 4A.
This journal is © The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1791–1798 | 1797

View Article Online
compounds were synthesized from the corresponding amines
and acyl chlorides under Schotten–Baumann conditions.
A. Wells, A. Zaks and T. Y. Zhang, Green Chem., 2007, 9, 411–
420.
J. S. Carey, D. Laffan, C. Thomson and M. T. Williams, Org. Biomol.
Chem., 2006, 4, 2337–2347.
P. J. Dunn, W. Hoffmann, Y. Kang, J. C. Mitchell and M. J. Snowden,
Org. Process Res. Dev., 2005, 9, 956–961.
K. Ishihara, S. Ohara and H. Yamamoto, J. Org. Chem., 1996, 61,
4196–4198.
K. Arnold, B. Davies, R. L. Giles, C. Grosjean, G. E. Smith and A.
Whiting, Adv. Synth. Catal., 2006, 348, 813–820.
R. M. Al-Zoubi, O. Marion and D. G. Hall, Angew. Chem., Int. Ed.,
2008, 47, 2876–2879.
7 J. W. Comerford, J. H. Clark, D. J. Macquarrie and S. W. Breeden,
Chem. Commun., 2009, (18), 2562–2564.
P. S. Chaudhari, S. D. Salim, R. V. Sawant and K. G. Akamanchi,
34
2
3
4
5
6
Enzymatic reactions
All enzyme preparations were dried in vacuo for at least 24 h
over phosphorous pentoxide before use. Reaction solvents were
dried with molsieve 4A and the different amines and esters were
dried over molsieve 3A before use. Stock solutions of various
concentrations of the amines, esters, and internal standards in
the reaction solvent were prepared and stored over molsieve
4
A. The reactions were started by adding the stock solution to a
8
certain amount of molsieve 4A and enzyme in a 1.5 or 10 mL vial
with a PP screw cap containing a natural rubber/TEF septum
and a small magnetic stirrer bar. Reactions were carried out in
an oil bath heated to different temperatures on an IKA RET
basic heater–stirrer plate (875 rpm) with an IKA ETS-D4 Fuzzy
thermocouple. Samples were taken at different time intervals
and added to 1 mL of MeOH in order to quench the reaction
and to dissolve any salts of the amines and carboxylic acids
that might have formed during the reaction. Subsequently, the
samples were centrifuged using a Thermo Scientific Heraeus
Pico 17 centrifuge (13 000 rpm, 10 min) to remove the molsieves
and enzyme. The supernatant was diluted as required in HPLC
eluent and injected directly into the HPLC.
Green Chem., 2010, 12, 1707–1710.
9 M. C. de Zoete, A. C. Kock-van Dalen, F. Van Rantwijk and
R. A. Sheldon, J. Chem. Soc., Chem. Commun., 1993, (24), 1831–
1
832.
1
1
0 S. Prasad and T. C. Bhalla, Biotechnol. Adv., 2010, 28, 725–741.
1 L. Mart ´ı nkov a´ and V. Kren, Curr. Opin. Chem. Biol., 2010, 14, 130–
137.
2 U. T. Bornscheuer, R. J. Kazlauskas, Hydrolases in Organic Syn-
thesis: Regio- and Stereoselective Biotransformations, Wiley-VCH,
Weinheim, 2nd edn, 2006.
1
13 A. Torres-Gavil a´ n, E. Castillo and A. L o´ pez-Mungu ´ı a, J. Mol. Catal.
B: Enzym., 2006, 41, 136–140.
1
1
4 F. van Rantwijk, M. A. P. J. Hacking and R. A. Sheldon, Monatsh.
Chem., 2000, 131, 549–569.
5 V. Gotor-Fern a´ ndez, E. Busto and V. Gotor, Adv. Synth. Catal., 2006,
348, 797–812.
6 K. Ditrich, Synthesis, 2008, 14, 2283–2287.
7 D. R. Allen, V. V. Mozhaev, R. H. Valivety, (Bakers and Daniels),
US Pat., 0 121 435, 2004.
1
1
HPLC analysis
1
1
8 R. A. Sheldon, Green Chem., 2007, 9, 1273–1283.
9 S. Naik, A. Basu, R. Saikia, B. Madan, P. Paul, R. Chaterjee, J.
Brask and A. Svendsen, J. Mol. Catal. B: Enzym., 2010, 65, 18–
23.
The aminolysis reactions were followed by HPLC on a Shimadzu
LC-20AT Prominence liquid chromatograph using a SIL-20AC
Prominence auto sampler. All analyses were carried out using
a 4.6 ¥ 50 Merck Chromolith SpeedROD RP-18e column with
different eluent compositions containing 0.1 v% acetic acid as
2
0 M. T. Reetz, J. D. Carballeira, J. Peyralans, H. H o¨ benreich,
A. Maichele and A. Vogel, Chem.–Eur. J., 2006, 12, 6031–
6
038.
-
1
organic modifier at 1 mL min and a column temperature of
21 K. Engstr o¨ m, J. Nyhl e´ n, A. G. Sandstr o¨ m and J.-E. B a¨ ckvall, J. Am.
◦
2
1 C. Compounds were detected using a Shimadzu SPD-20A
Chem. Soc., 2010, 132, 7038–7042.
2 P. Hoyos, M. Fern a´ ndez, J. V. Sinisterra and A. R. Alc a´ ntara, J. Org.
Chem., 2006, 71, 7632–7637.
3 A. S. Demir, H. Findik and E. K o¨ se, Tetrahedron: Asymmetry, 2004,
15, 777–781.
2
2
2
Prominence UV/VIS detector at 215 nm and a Shimadzu RID-
0A refractive index detector. The aminolysis reactions of ester
1
II with the amines 1–7, as well as the aminolysis reactions of
ester III with the amines 1–5 and the reactions of esters V
4 T. Yamamoto, N. Shibata, M. Takashima, S. Nakamura, T. Toru,
N. Matsunaga and H. Hara, Org. Biomol. Chem., 2008, 6, 1540–
and VI with amine 4 were analyzed using an eluent of H O–
2
1
543.
MeOH (60 : 40, v/v) and 1,3-dimethoxybenzene as the internal
standard. The aminolysis reactions of ester I with the amines
2
2
2
5 M.-J. Kim, Y. K. Choi, S. Kim, D. Kim, K. Han, S.-B. Ko and J.
Park, Org. Lett., 2008, 6, 1295–1298.
6 T. C. Bowen, H. Kalipcilar, J. L. Falconer and R. D. Noble, J. Membr.
Sci., 2003, 215, 235–247.
1
–5 were analyzed using an eluent of H O–MeOH (75 : 25,
2
v/v) using 1,3-dimethoxybenzene as the internal standard and
the aminolysis reactions of ester IV with the amines 1–5 were
7 L. Kvittingen, Tetrahedron, 1994, 50, 8253–8274.
28 Y.-F. Wang, J. J. Lalonde, M. Momongan, D. E. Bergbreiter and
analyzed using an eluent of H
2
O–MeOH (80 : 20, v/v) using
C.-H. Wong, J. Am. Chem. Soc., 1988, 110, 7200–7205.
2
3
3
9 S. Ray and S. K. Ray, J. Membr. Sci., 2006, 278, 279–289.
0 K. Alm and M. Ciprian, J. Chem. Eng. Data, 1980, 25, 100.
1 I. Mart ´ı nez, A. Markovits, R. Chamy and A. Markovits, Appl.
Biochem. Biotechnol., 2004, 112, 55–62.
1
,3,5-trimethoxybenzene as the internal standard. The methods
mentioned above allowed for baseline separation of the ester,
carboxylic acid, amide and internal standard.
3
2 P. Hoyos, A. Buthe, M. B. Ansorge-Schumacher, J. V. Sinisterra
and A. R. Alc a´ ntara, J. Mol. Catal. B: Enzym., 2008, 52–53, 133–
1
39.
References
3
3 B. D. Hosangadi and R. H. Dave, Tetrahedron Lett., 1996, 37, 6375–
6378.
1
D. J. C. Constable, P. J. Dunn, J. D. Hayler, G. R. Humphrey, J.
L. Leazer, R. J. Linderman, K. Lorenz, J. Manley, B. A. Pearlman,
34 C. Schotten, Ber. Dtsch. Chem. Ges., 1884, 17, 2544.
1
798 | Green Chem., 2011, 13, 1791–1798
This journal is © The Royal Society of Chemistry 2011