Lipase-catalyzed transesterification of ethyl formate to octyl formate

Article abstract of DOI:10.1016/j.molcatb.2014.03.016

The preparation of octyl formate via lipase-catalyzed transesterification of ethyl formate with 1-octanol is demonstrated. To shift the equilibrium of the reaction, ethyl formate was added in surplus but could be partially recovered for subsequent reactions. The same was true for the biocatalyst (Novo435), which could be reused at least 27 times. This method gives simple access to a hydrophobic formic acid ester, which can be used as a reactive organic phase in biocatalytic redox reactions. The enzymatically prepared octyl formate can be utilized by formate dehydrogenase to regenerate NADH from NAD⁺.

Full text of DOI:10.1016/j.molcatb.2014.03.016

Journal of Molecular Catalysis B: Enzymatic 105 (2014) 7–10
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Lipase-catalyzed transesterification of ethyl formate to octyl formate
1
1
Leonie M.G. Janssen , Remco van Oosten , Caroline E. Paul, Isabel W.C.E. Arends,
Frank Hollmann
∗
Department of Biotechnology, Delft University of Technology, Julianalaan 136, 2628BL Delft, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
The preparation of octyl formate via lipase-catalyzed transesterification of ethyl formate with 1-octanol
is demonstrated. To shift the equilibrium of the reaction, ethyl formate was added in surplus but could
be partially recovered for subsequent reactions. The same was true for the biocatalyst (Novo435), which
could be reused at least 27 times.
Received 12 November 2013
Received in revised form 16 March 2014
Accepted 25 March 2014
Available online 1 April 2014
This method gives simple access to a hydrophobic formic acid ester, which can be used as a reactive
organic phase in biocatalytic redox reactions. The enzymatically prepared octyl formate can be utilized
Keywords:
Lipase
Transesterification
NADH regeneration
Formate dehydrogenase
+
by formate dehydrogenase to regenerate NADH from NAD .
©
2014 Elsevier B.V. All rights reserved.
1
. Introduction
To attain a commercially and ecologically feasible synthesis of
formyl esters, a range of challenges has to be addressed. First, the
cost contribution of the biocatalyst to the overall process has to
be reduced, e.g. by recycling [8]. Second, the enormous surplus of
substrate, needed to shift the equilibrium of the transesterification
reaction, should be reduced [9].
The use of hydrolases (E.C. 3.1) for the interconversion of car-
boxylic acids, ester, and amides is nowadays a well-established
procedure also utilized on industrial scale [1–3]. Interestingly, the
formation of formic acid esters, despite several promising early
contributions [4], has so far failed to be the focus of research. This is
rather astonishing as some formyl esters are valuable compounds,
e.g. ingredients in flavors and fragrances [5].
2. Materials and methods
We became interested in the enzymatic synthesis of formic acid
esters in the course of an ongoing project evaluating formic acid
esters, such as octyl formate, as hydrophobic formate equivalents
to serve as an organic phase in two-liquid phase reaction setups for
oxidoreductases [6]. This project was inspired by the work of Bertau
and co-workers who first demonstrated that formate dehydroge-
nase also accepts formic acid esters [7]. Therefore, we envisioned
a simple and scalable method to produce formic acid esters. Aim-
ing at an environmentally acceptable process, lipases are amongst
the first catalysts to be considered. Overall, we designed an enzy-
matic production system of formic acid esters from commercially
available ethyl formate (Scheme 1). As biocatalyst we chose the
immobilized lipase B from Candida antarctica commercialized as
Novo435.
2
.1. Chemicals and enzymes
All chemicals were purchased from Sigma–Aldrich (Zwijn-
drecht, The Netherlands) in the highest purity available and used
without further purification.
The immobilized lipase B from Candida antarctica (trade name
Novo435) was kindly donated by Novozymes (Bagsværd, Denmark)
and used as received.
Formate dehydrogenase (FDH) was obtained by expressing fdh
in E. coli JM101 carrying the plasmid pBTac2 (C23S) following a
published procedure [10]. For the experiments reported here, crude
cell extracts of the recombinant E. coli cells were used.
2.2. Methods
Throughout the manuscript enzyme activities are expressed
as international units (U) with 1 U being the enzyme activity
producing 1 ␮mol of product (octyl formate in case of the lipase-
catalyzed transesterification and NADH in case of the formate
∗ Corresponding author. Tel.: +31 152781957.
E-mail address: f.hollmann@tudelft.nl (F. Hollmann).
1
+
Both authors contributed equally to this work.
dehydrogenase-catalyzed reduction of NAD ) per minute.
http://dx.doi.org/10.1016/j.molcatb.2014.03.016
1
381-1177/© 2014 Elsevier B.V. All rights reserved.

8
L.M.G. Janssen et al. / Journal of Molecular Catalysis B: Enzymatic 105 (2014) 7–10
00
1
7
5
2
5
0
5
0
Scheme 1. Lipase-catalyzed transesterification of ethyl formate with various alco-
hols.
All transesterification reactions were carried out using a three-
necked flask equipped with magnetic top-stirring, temperature
control and reflux cooling. The reaction mixture was exposed to
the ambient atmosphere and no extra drying or degassing measures
were taken. Unless indicated otherwise, the reagents were added
to the flask (amounts are specified in the captions of the figures
and tables) and incubated for at least 15 min at the envisaged reac-
tion temperature. Prior to starting the reaction by addition of the
biocatalysts, a zero-sample was taken. Samples were taken at dif-
ferent time intervals to follow the reaction progress. The stirring
was interrupted during the sampling to allow the biocatalyst to
settle; then a sample (0.1 ml) was taken from the reaction mix-
ture, diluted with 0.9 ml heptane (containing 100 mM dodecane as
internal standard) mixture and analyzed by GC.
0
4
8
12
16
20
24
reacꢀon ꢀme [h]
Fig. 1. Influence of the ratio of 1-octanol to ethyl formate on the time-course of the
transesterification reaction. General conditions: m(Novo435) = 0.1 g; m(ethyl for-
mate) = 48 g (0.65 mol); m(1-Octanol) = 2 g (♦), 5 g (ꢀ), 10 g (ꢀ), 20 g (ꢁ), 40 g (ꢂ),
−1 ◦
8
0 g (᭹) (0.025, 0.062, 0.12, 0.25, 0.5 and 1 mol mol ); T = 40 C.
3. Results and discussion
We began our study by investigating the effect of varying ratios
of the starting materials (ethyl formate and 1-octanol) over the time
course of the transesterification reaction (Fig. 1). It is worth men-
tioning that all experiments reported here were performed under
neat conditions.
2.3. Recycling experiments
After every cycle, the biocatalyst was filtrated and washed two
When using approximately equimolar concentrations of the
starting material, only 35% conversion was observed after 24 h.
Upon prolonged reaction time (3 days), this value increased to 48%
indicating an equilibrium constant of the transesterification reac-
tion of around 1. In this set of experiments, because both initial
concentrations of 1-octanol and ethyl formate were varied simul-
taneously (neat reaction conditions) kinetic parameters could not
be determined at this stage. It is, however, interesting to note that
the biocatalyst’s specific activity (as determined in the first hour)
times with 20 ml portions of 1-octanol. When stored over night or
over the weekend, the biocatalyst was kept under 1-octanol at room
temperature (this was done in order to prevent possibly remaining
ethyl formate to hydrolyze; the resulting formic acid has been pre-
viously demonstrated to inactivate the biocatalyst). Ethyl formate
was (partially recovered) from the reaction mixture by distillation
under reduced pressure. Unless otherwise indicated, the distillate
was weighted, supplemented with fresh ethyl formate (to a total
of 48 g, typically 15–20 g of fresh ethyl formate were added), 5 g of
−
1
−1
gradually increased from 1.77 U mg
to 4.99 U mg Novo435.
Novo435
1
-octanol and used for the next reaction cycle. The remaining crude
For subsequent experiments, we used a molar ratio of 94:6
between ethyl formate and 1-octanol. The resulting product
composition of approximately 9:1 of octyl formate to 1-octanol
appeared acceptable in view of the envisioned application as reac-
tive organic phase for NADH-dependent reactions (vide infra).
When using inactivated catalyst under otherwise identical reac-
tion conditions, less than 2% conversion was observed within 24 h.
Also, formic acid was not efficiently converted by Novo435, which
we attribute to the inactivity of CALB in the presence of strong acids
octyl formate product was collected and used for the FDH-activity
assay.
2.4. GC analytics
Product analysis was performed on a Shimadzu 2014 gas chro-
◦
matograph (direct injection, 320 C) using a CP SIL 5 CB column
(
(
50 m, 0.53 mm, 1.0 ␮m) and the following temperature program
◦
◦
◦
50 C hold for 4 min, raise to 110 C (20 C per min) hold for 3 min,
[
11]. The temperature-dependence of the transesterification rate is
◦
◦
raise to 350 C (20 C per min) hold for 1 min).
shown in Fig. 2. For further experiments we chose a reaction tem-
Sample preparation entailed dilution of the sample (0.1 ml) with
◦
perature of 40 C as it appeared to be a good compromise between
0.9 ml heptane (containing 100 mM dodecane as internal standard).
biocatalyst activity (increasing with temperature) and undesired
evaporation of ethyl formate.
Octyl formate and 1-octanol formations were based on calibra-
tion curves using authentic standards.
Catalyst regeneration is of the utmost importance to attain
economic feasibility of an enzymatic process, especially if low-
value-added compounds are envisaged. Therefore, we evaluated
the recycling of the catalyst Novo435 (Fig. 3). It should be men-
tioned here that we also aimed at recycling the non-reacted starting
material (ethyl formate). Thus, after each reaction, not only the cat-
alysts were recovered (by filtration) but also the non-reacted ethyl
formate was distilled off the crude reaction mixture. For the next
cycle, the distillate was supplemented with fresh ethyl formate and
2.5. Formate dehydrogenase assay
To test the enzymatic activity of FDH with the crude octyl
formate product, a spectroscopic assay following the character-
istic absorbance of reduced nicotinamide cofactors at 340 nm
−
1
−1
(
(
2
ε = 6220 M cm ) was used. For this assay, 10 ml TRIS buffer
+
50 mM, pH 7.5) containing 1 mM NAD were supplemented with
5 ␮l FDH crude extract (approximately 5 U ml FDH activity in
−
1
1
-octanol. In these experiments, a molar ratio of 94:6 between ethyl
the stock). To this solution, 1 ml of the crude octyl formate prod-
uct (pooled products from recycling experiments 1–20) was added.
formate and 1-octanol was used, which would allow for a maximal
conversion of 1-octanol of approximately 90% (Fig. 1). We assumed
that this purity would be sufficient for the envisaged application
as ‘hydrophobized formic acid’ since 1-octanol will represent the
major side-product of this concept (Scheme 2).
◦
The resulting emulsion was gently shaken at 30 C. Samples were
taken at different time intervals, then centrifuged and analyzed on
a Shimadzu UV-2401 PC spectrophotometer.

L.M.G. Janssen et al. / Journal of Molecular Catalysis B: Enzymatic 105 (2014) 7–10
9
100
7
5
2
5
0
5
0
0
5
15
60
300
Reacꢀon ꢀme [min]
Fig. 2. Temperature-dependence of the Novo435-catalyzed transesterification
of ethyl formate with 1-octanol. From left to right: reaction temperatures of
◦
◦
◦
◦
◦
◦ ◦
C and 60 C, respectively. General condi-
0
C, 10 C, 20 C, 30 C, 40 C, 50
tions: m(Novo435) = 0.1 g; m(ethyl formate) = 48 g (0.65 mol), m(1-octanol) = 5 g
(
0.04 mol).
6
5
4
3
2
1
100
Scheme 2. The ‘hydrophobized formates concept’ utilizing esters of formic acid with
hydrophobic alcohols. Instead of the conventional formic acid salts, the correspond-
ing esters are utilized as stoichiometric reductant to promote NADH regeneration.
At the same time, the esters as well as the corresponding alcohols serve as organic
phase.
7
5
50
separation. Further experiments clarifying this [noun] are currently
underway.
Nevertheless, the recycling experiments demonstrate that
Novo435 can be reused at least 27 times for the envisaged transes-
terification reaction. The activity loss under these conditions was
maximally 45%.
2
0
5
0
0
5
10
15
20
25
30
recycle No
Overall, from cycles 1 to 20, 96 g of product, containing 88.5%
w/w) of octyl formate and 11.5% (w/w) of 1-octanol, were obtained.
This crude product mixture was evaluated as organic phase to
Fig. 3. Recycling experiments for the Novo435-catalyzed transesterification of ethyl
formate with 1-octanol: (ꢀ) conversion after 24 h and (ꢁ) initial rate. General con-
(
ditions: m(Novo435) = 0.1 g; m(ethyl formate) = 48 g (0.65 mol), m(1-octanol) = 5 g
◦
(
0.04 mol), T = 40 C. After each cycle, the enzyme was filtrated off, washed with 1-
promote formate dehydrogenase (FDH)-catalyzed regeneration of
octanol and reused or stored at ambient temperature. Ethyl formate was recovered
under vacuum from the filtrate, supplemented with fresh ethyl formate to a total of
NADH from NAD+ (vide infra). Summing up all the reagents used
in these cycles enabled us to calculate an overall E-factor [12] of
.95 with respect to octyl formate. Compared to the E-factor of a
4
8 g. After addition of 5 g of 1-octanol, a new reaction cycle was initiated by addition
5
of the recovered enzyme.
single use (8.7), this represents only a marginal improvement. The
major reason for this lies in the loss of ethyl formate during the
recycling procedure (operating under open atmosphere without
taking precautions to prevent the escape of volatile solvents into
the environment). As a result, for every new cycle, more ethyl for-
mate than theoretically necessary had to be supplemented for the
next cycle. Nevertheless, we are confident that optimized reaction
and recycling conditions will lead to improved recycling efficiencies
and decreased environmental impact due to waste formation.
Finally, we evaluated the applicability of the crude product mix-
tures (as obtained from the recycling experiments) as reactive
organic phase in formate-dehydrogenase-catalyzed reduction of
Fig. 3 shows the initial product formation rates and final con-
versions (of 1-octanol after 24 h reaction time) determined in these
recycling experiments. While in every experiment, maximal con-
version was observed after 24 h, there was a significant drop in
initial rate. Within the first 5 cycles, the initial rate dropped by
more than 50% and then mostly remained at this low level. In cycle
2
0, however, we did not recycle the remaining ethyl formate but
utilized fresh one. Pleasingly, the initial transesterification activity
−
1
of the recycled Novo435 increased significantly (from 1.1 U mg to
4
−
1
.0 U mg in cycles 19 and 20, respectively). Using recycled ethyl
+
NAD . As shown in Fig. 4, the crude octyl formate preparation was
formate again led to a significant drop of initial rate (cycles 21–23),
which could be overcome by using fresh ethyl formate (cycles 24
and 27). We attribute this peculiar behavior to the accumulation
of ethanol in the recovered ethyl formate thereby competing with
readily accepted by formate dehydrogenase to regenerate NADH
+
from NAD . The apparent NADH-formation activity observed in
this experiment (approx. 0.014 U) falls back by one order of mag-
nitude below the theoretical FDH activity applied (0.126 U), which
we attribute to the diffusion limitation of octyl formate into the
aqueous medium due to the reaction setup. This finding is in line
with our previous observation with commercial octyl formate [6].
Nevertheless, this experiment demonstrates that even the crude
product from the recycling experiments (containing significant
amounts of 1-octanol) can be applied to promote FDH-catalyzed
NADH-regeneration reactions.
1
-octanol as nucleophile for the enzyme-formyl-ester and reduc-
ing the apparent initial rate. Indeed, NMR measurements revealed
the presence of ethanol in the recovered ethyl formate. Princi-
pally, the boiling points of ethyl formate and ethanol should be
sufficiently separated to enable clean distillative separation. Ethyl
formate and ethanol potentially form an azeotrope, complicat-
ing their separation. Alternatively, the simple rotary evaporator
distillation conditions were not sufficiently precise for a clean

1
0
L.M.G. Janssen et al. / Journal of Molecular Catalysis B: Enzymatic 105 (2014) 7–10
.1
0
The major limitation of the proposed procedure lies in the
0
0
0
.6
.4
.2
0
reversibility of the transesterification reaction. In order to shift
the equilibrium, we have used large molar excesses of ethyl for-
mate. The resulting waste formation could be partially alleviated
by recycling the acyl donor. Its rather low boiling point prevented
simple distillative in situ product removal. Further investigations
in our group will therefore focus on the use of less volatile for-
mates (e.g. glycerol triesters, etc.) as acyldonors for the synthesis
of hydrophobic formic acid esters.
0
0
0
0
.08
.06
.04
.02
0
0
20 40 60 80
t [min]
Acknowledgements
3
00
350
400
λ [nm]
450
500
This project was supported by the BIOTRAINS Marie Curie Initial
Training Network, financed by the European Union through the 7th
Framework People Programme (grant agreement number 238531).
Fig. 4. Time course of the FDH-catalyzed reduction of NAD+ using octyl formate
produced in this study. General conditions: aqueous phase 9 ml TRIS (50 mM, pH
.5), [NAD ] = 1 mM and [FDH] = 0.014 U ml (crude enzyme preparation); organic
phase 1 ml crude octyl formate (pooled fractions 1–20 of Fig. 3), UV spectra were
measured after 0, 2, 15, 23, 28, 43 and 73 min.
+
−1
7
References
[
[
[
1] U. Bornscheuer, R. Kazlauskas, Hydrolases in Organic Synthesis, 2nd ed., Wiley-
VCH, Weinheim, 2006.
2] K. Faber, Biotransformations in Organic Chemistry, 6th ed., Springer, Berlin,
2011.
3] K. Drauz, H. Groeger, O. May, Enzyme Catalysis in Organic Synthesis, Wiley-
VCH, Weinheim, 2012.
4
. Conclusions
[
[
[
4] H.S. Bevinakatti, R.V. Newadkar, Biotechnol. Lett. 11 (1989) 785–788.
5] A. Herrmann, Angew. Chem. Int. Ed. 46 (2007) 5836–5863.
6] E. Churakova, B. Tomaszewski, K. Buehler, A. Schmid, I.W.C.E. Arends, F. Holl-
mann, Top. Catal. 57 (2014) 385–391.
Very recently, we have demonstrated that hydrophobic esters
of formic acid can be utilized as reactive organic phases for NADH-
dependent oxidoreductases reactions (Scheme 2) [6].
In the present study, we have demonstrated that the simple
transesterification of commercially available ethyl formate using
commercially available) Novo435 represents an efficient access to
the desired product. The parameters determining the efficiency of
the system have been investigated. Furthermore the suitability of
even the crude product to promote FDH-catalyzed regeneration of
NADH has been demonstrated.
[7] P. Frohlich, K. Albert, M. Bertau, Org. Biomol. Chem. 9 (2011) 7941–7950.
[
8] P. Tufvesson, J. Lima-Ramos, M. Nordblad, J.M. Woodley, Org. Process Res. Dev.
5 (2010) 266–274.
1
[
9] Y. Ni, D. Holtmann, F. Hollmann, ChemCatChem (2014), http://dx.doi.org/
10.1002/cctc.201300976 (in press).
(
[
[
[
10] H. Slusarczyk, S. Felber, M.R. Kula, M. Pohl, Eur. J. Biochem. 267 (2000)
280–1289.
11] F. Hollmann, P. Grzebyk, V. Heinrichs, K. Doderer, O. Thum, J. Mol. Catal. B:
Enzym. 57 (2009) 257–261.
1
12] R.A. Sheldon, Chem. Commun. (2008) 3352–3365.