Enzymatic synthesis of activated esters and their subsequent use in enzyme-based peptide synthesis

Article abstract of DOI:10.1016/j.molcatb.2011.03.012

Chemoenzymatic peptide synthesis is potentially the most cost-efficient technology for the synthesis of short and medium-sized peptides. However, there are still some limitations when challenging peptides, e.g. containing sterically demanding acyl donors, non-proteinogenic amino acids or proline residues, are to be synthesized. To remedy these limitations, special ester moieties have been used that are specifically recognized by the enzyme, e.g. guanidinophenyl, carboxamidomethyl (Cam) or trifluoroethyl (Tfe) esters, which, unfortunately, are notoriously difficult to synthesize chemically. Herein, we demonstrate that Cam and Tfe esters are very useful for Alcalase-CLEA mediated peptide synthesis using sterically demanding and non-proteinogenic acyl donors as well as poor nucleophiles, and combinations thereof. Furthermore, these esters can be efficiently synthesized by using the lipase Cal-B or Alcalase-CLEA. Finally, it is shown that the ester synthesis by Cal-B and subsequent peptide synthesis by Alcalase-CLEA can be performed simultaneously using a two-enzyme-one-pot approach with glycolamide or 2,2,2-trifluoroethanol as additive.

Full text of DOI:10.1016/j.molcatb.2011.03.012

Journal of Molecular Catalysis B: Enzymatic 71 (2011) 79–84
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Enzymatic synthesis of activated esters and their subsequent use in
enzyme-based peptide synthesis
Timo Nuijensa,b, Claudia Cusana, Annette C.H.M. Schepersa, John A.W. Kruijtzerb,
Dirk T.S. Rijkersb, Rob M.J. Liskampb, Peter J.L.M. Quaedfliega,∗
a DSM Innovative Synthesis B.V., P.O. Box 18, NL-6160 MD Geleen, The Netherlands
b Medicinal Chemistry and Chemical Biology, Utrecht Institute for Pharmaceutical Sciences, Utrecht University, P.O. Box 80082, NL-3508 TB Utrecht, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Chemoenzymatic peptide synthesis is potentially the most cost-efficient technology for the synthesis of
short and medium-sized peptides. However, there are still some limitations when challenging peptides,
e.g. containing sterically demanding acyl donors, non-proteinogenic amino acids or proline residues, are
to be synthesized. To remedy these limitations, special ester moieties have been used that are specifically
recognized by the enzyme, e.g. guanidinophenyl, carboxamidomethyl (Cam) or trifluoroethyl (Tfe) esters,
which, unfortunately, are notoriously difficult to synthesize chemically. Herein, we demonstrate that Cam
and Tfe esters are very useful for Alcalase-CLEA mediated peptide synthesis using sterically demanding
and non-proteinogenic acyl donors as well as poor nucleophiles, and combinations thereof. Furthermore,
these esters can be efficiently synthesized by using the lipase Cal-B or Alcalase-CLEA. Finally, it is shown
that the ester synthesis by Cal-B and subsequent peptide synthesis by Alcalase-CLEA can be performed
simultaneously using a two-enzyme-one-pot approach with glycolamide or 2,2,2-trifluoroethanol as
additive.
Received 14 December 2010
Received in revised form 29 March 2011
Accepted 29 March 2011
Available online 5 April 2011
Keywords:
Esterification
Kinetic coupling
Peptide synthesis
Alcalase-CLEA
Cal-B
© 2011 Elsevier B.V. All rights reserved.
1. Introduction
monly used for the synthesis of small peptides containing two to
ten amino acid residues. Also this approach requires expensive cou-
In the past few decades, a large number of peptides have been
introduced onto the market since they can be used either as a
therapeutic or as prodrug [1], and an even increasing number is
in clinical trails. Additionally, peptides have found applications
as nutritional additive or as a cosmetic ingredient [2]. Despite
this demand for peptides, their production on large scale remains
expensive and time consuming [3]. Common peptide synthesis
approaches are fermentation, solid-phase or solution-phase chem-
ical peptide synthesis, and chemo-enzymatic peptide synthesis [4].
Currently, the fermentative approach is well feasible for large pep-
tides (>50 amino acid residues) and proteins containing natural
fragments and requires a large development effort for each indi-
vidual peptide/protein. Solid phase peptide synthesis (SPPS) is the
most commonly used method for medium-sized and long peptides
(10–50 amino acid residues). However, SPPS requires besides full
protection of the functionalized side chains, also expensive and
environmentally unfriendly coupling reagents in at least stoichio-
metric amounts. Furthermore, the requirement of functionalized
resins and the use of reagent excess makes SPPS an expensive
method. Solution phase chemical peptide synthesis is most com-
pling reagents and of major concern is the uncontrolled C-terminal
racemization during fragment assembly. Finally, chemo-enzymatic
peptide synthesis, wherein peptide fragments are elongated enzy-
matically, has been studied in academia during the past decades
and proved to be suitable for certain short peptide sequences up
to five amino acid residues [4]. The application of enzymes as
coupling reagent is a promising alternative since functionalized
amino acid side chain do not require full protection and most
importantly, C-terminal racemization is completely absent during
fragment assembly, which is beneficial for the characterization and
purification of the final peptide.
There are two approaches toward enzymatic peptide synthe-
sis, either the thermodynamically, or the kinetically controlled
approach [5]. In thermodynamically controlled peptide synthesis,
an N-terminally protected acyl donor reacts with a C-terminally
protected amino acid acceptor as the nucleophile, resulting in
the formation of the peptide bond, and one water molecule is
expelled. Thermodynamically controlled peptide synthesis is, how-
ever, rather slow and the thermodynamic equilibrium between
product and starting materials needs to be shifted into the syn-
thetic direction, for example by product precipitation, by water
withdrawal, or by using organic solvents, to obtain a high product
yield. This is in contrast to the kinetically controlled peptide synthe-
sis, in which an N-terminally protected and C-terminally activated
∗
Corresponding author. Tel.: +31 46 4761592; fax: +31 46 4767604.
E-mail address: peter.quaedflieg@dsm.com (P.J.L.M. Quaedflieg).
1381-1177/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2011.03.012

80
T. Nuijens et al. / Journal of Molecular Catalysis B: Enzymatic 71 (2011) 79–84
Fig. 1. Representative examples of acyl donors that are specifically recognized by certain proteases.
amino acid ester reacts preferentially with a C-terminally protected
acyl acceptor, to give a high product yield in a generally shorter
reaction time.
relative to DMSO-d6 (39.50 ppm) or CDCl3 (77.00). Column chro-
matography was carried out using silica gel, Merck grade 9385
˚
60 A. Analytical HPLC diagrams were recorded on an HP1090 Liquid
The high selectivity of enzymes restricts the number of amino
acids that will be recognized. Therefore, coupling of sterically
demanding amino acid as acyl donors (valine, isoleucine, threo-
nine), notoriously weak nucleophiles (proline), or d- and other
non-proteinogenic amino acid residues, remains rather challeng-
ing. To remedy the restricted access toward the primary specificity
pocket, designed acyl donors with an ester moiety that is specif-
ically recognized by the enzyme, among others, guanidinophenyl
(Gp) [6], carboxamidomethyl (Cam) [7] or trifluoroethyl (Tfe) [8]
esters have been used (Fig. 1). Thus, successful couplings of steri-
cally demanding acyl donors decorated with these activated ester
moieties have been reported. Also, in the presence of these acti-
vated esters, enzyme mediated couplings of weak nucleophiles
and non-proteinogenic amino acid residues, became feasible [9].
Although these active esters broaden the scope of enzymatic
peptide synthesis, their chemical synthesis is however not straight-
forward [10] since highly reactive coupling reagents are required
to couple the rather poor nucleophilic alcohol derivatives which
increases the risk of racemization [11].
Herein, we report a highly promising enzymatic approach
toward the synthesis of Cam and Tfe active esters by means of
cross-linked enzyme aggregates of Alcalase (Alcalase-CLEA) [12]
and immobilized Candida antartica lipase B (Cal-B). Additionally,
their subsequent application is demonstrated in an enzymatic pep-
tide synthesis approach using a number of challenging acyl donors
and nucleophiles to obtain not easily accessible dipeptides. Further-
more, we explored a ‘two-enzyme-one-pot’ approach, in which the
activated ester is synthesized by Cal-B and simultaneously used as
a substrate by Alcalase-CLEA to elongate the peptide sequence.
Chromatograph, using a reversed-phase column (Inertsil ODS-3,
C18, 5 ␮m particle size, 150 mm × 4.6 mm internal diameter) at
40 ◦C. UV detection was performed at 220 nm using a UV–VIS 204
Linear spectrometer (Varian). The gradient program was: 0–25 min
linear gradient ramp from 5% to 98% eluent B and from 25.1 to
30 min at 5% eluent B (eluent A: 0.5 mL/L methane sulfonic acid
(MSA) in H2O; eluent B: 0.5 mL/L MSA in acetonitrile). Prepara-
tive HPLC was performed on a Varian PrepStar system using a
stationary-phase column (Pursuit XRs, C18, 10 ␮m particle size,
500 mm × 41.4 mm) at room temperature. UV detection was per-
formed at 220 nm and 254 nm using a UV–VIS Varian ProStar
spectrometer. The gradient program was 20% eluent B and 80%
eluent A to 95% eluent B and 5% eluent A in 30 min (eluent A:
0.1 mL/L formic acid in H2O; eluent B: 0.1 mL/L formic acid in ace-
tonitrile) with a flow rate of 80 mL/min and an injection volume
of 10 mL. Pure fractions were pooled and lyophilized. The flow-
injection analysis (FIA) experiments to determine the exact mass
were performed on an Agilent 1100 LC-MS system (Agilent, Wald-
bronn, Germany). The ESI-MS was run in positive mode, with the
following conditions: m/z 50–3200, 175 V fragmentor, 0.94 cycl/s,
350 ◦C drying gas temperature, 10 L N2/min drying gas, 45 psi g
nebuliser pressure and 4 kV capillary voltage. The exact mass was
determined using an internal referent to recalibrate the m/z axis
for each measurement. Reference dipeptides [13] as well as N-Cbz-
protected amino acid methyl [14], Cam [15] and Tfe [16] esters were
synthesized according to the literature procedures. Analytical data
of all known compounds was compared to those reported in the
literature which is cited in the supplementary information.
2.1.1. Cbz-d-Phe-OTfe
This compound was synthesized from Cbz-d-Phe-OH accord-
ing to the literature procedure [16] and obtained as a white solid;
Rt(HPLC) 22.37 min; 1H NMR (CDCl3, 300 MHz): ı = 2.99–3.10 (m,
2H), 4.37–4.46 (m, 2H), 4.65–4.71 (m, 1H), 5.02–5.09 (m, 3H),
7.02–7.30 (m, 10H); 13C NMR (CDCl3, 75 MHz): ı = 37.8, 54.5, 60.6,
61.1, 67.1, 127.3, 128.0, 128.2, 128.4, 128.7, 129.0, 134.9, 170.1;
FIA-ESI(+)-TOF-MS: m/z [M + H]+ calcd for C19H19F3NO4: 382.1261;
found: 382.1249.
2. Materials and methods
2.1. Chemicals and enzyme preparation
Unless stated otherwise, chemicals were obtained from com-
mercial sources and used without further purification. Prior to
use, Alcalase-CLEA (3 g, Type OM, CLEA-technologies, 580 U/g) was
suspended in tBuOH (100 mL) and crushed with a spatula. After
being collected by filtration, the enzyme was resuspended in
MTBE or THF (50 mL) and isolated by filtration. A stock solution
of 2-hydroxyacetamide (CAS 598-42-5 from Sigma–Aldrich) was
prepared by dissolving 2-hydroxyacetamide (1.25 g) in HPLC grade
acetonitrile (200 mL) and MgHPO4 (5.0 g) was added. The obtained
suspension was stirred for 30 min at room temperature followed
2.2. Enzymatic peptide synthesis
Alcalase-CLEA (100 mg) was added to a mixture of MTBE or
˚
THF (3 mL) containing activated 3 A molsieves (200 mg). Sub-
sequently, the N-Cbz-protected amino acid or dipeptide ester
(100 mg) followed by a C-terminally protected amino acid or dipep-
tide (1.5 equiv) were added to the enzyme suspension. The reaction
mixture was shaken at 150 rpm at 50 ◦C for 16 h. After filtra-
tion, the remaining solids were subsequently washed with EtOAc
(3 × 10 mL), CH2CL2 (3 × 10 mL) and MeOH (3 × 10 mL). The com-
bined filtrate was concentrated in vacuo and the resulting residue
was purified by one of the following three different methods.
by filtration. Cal-B was purchased from Novozymes (immobilized
®
˚
Novozym -435, LC 200204). Molsieves (3 A, 8–12 Mesh, Acros)
were activated under reduced pressure at 200 ◦C. tBuOH was dried
on activated molsieves prior to use. 1H (300 MHz) and 13C NMR
(75 MHz) spectra were recorded on a Bruker Avance NMR spec-
trometer. 1H chemical shifts are given in ppm (ı) relative to TMS
(0.00 ppm) or DMSO-d6 (2.50 ppm) and 13C NMR chemical shifts

T. Nuijens et al. / Journal of Molecular Catalysis B: Enzymatic 71 (2011) 79–84
81
In first method, depending on the solubility of the product, the
peptides were purified by column chromatography using EtOAc/n-
heptane or CH2Cl2/MeOH as eluent. In the second method, the
residue was redissolved in EtOAc or CH2Cl2 (50 mL) and the solution
was washed with sat. aq. NaHCO3 (25 mL), 1 mM aq. HCl (25 mL),
and brine (25 mL). The organic layer was dried (Na2SO4), concen-
trated in vacuo, and the volatiles were co-evaporated with toluene
(2 × 20 mL) and CHCl3 (2 × 20 mL). The third method was purifica-
tion of the peptides by preparative HPLC.
(50 mg). The reaction mixture was shaken at 150 rpm at 50 ◦C for
72 h. Purification of the ester was performed as described in Section
2.2.
2.4. Cam or Tfe ester synthesis using Cal-B
Cal-B (100 mg) was added to a mixture of MTBE or ace-
˚
tonitrile (3 mL), 3 A molsieves (100 mg), carbamoylmethanol or
2,2,2-trifluoroethanol (200 mg), and N-Cbz-protected amino acid or
dipeptide (50 mg). The reaction mixture was shaken at 150 rpm at
50 ◦C for 16 h. Purification of the ester was performed as described
in Section 2.2.
2.2.1. Cbz-l-Phe-l-Pro-NH2
White solid; Rt(HPLC) 16.53 min; 1H NMR (CDCl3, 300 MHz):
ı = 1.65–1.69 (m, 3H), 2.00–2.25 (m, 1H), 2.85–3.01 (m, 2H),
3.37–3.57 (m, 1H), 4.33–4.48 (m, 1H), 4.68 (q, 1H, J = 15.0 and
7.5 Hz), 5.00 (q, 2H, J = 18.9 and 11.4 Hz) 5.45 (s, 1H), 5.72 (d,
1H, J = 8.7 Hz), 6.32 (s, 1H), 7.15–7.26 (m, 11H); 13C NMR (CDCl3,
75 MHz): ı = 24.4, 28.2, 38.4, 47.8, 54.6, 60.0, 67.1, 127.3, 127.9,
128.2, 128.5, 128.6, 129.3, 135.7, 136.1, 156.4, 171.4, 173.5; FIA-
ESI(+)-TOF-MS: m/z [M + H]+ calcd for C22H26N3O4: 396.1918;
found: 396.1925.
2.4.1. Cbz-l-Ala-l-Ala-OTfe
White solid; Rt(HPLC) 18.06 min; 1H NMR (CDCl3, 300 MHz):
ı = 1.30-1.36 (m, 6H), 4.21–4.59 (m, 4H), 5.03 (s, 2H), 5.37 (d,
1H, J = 6.6 Hz), 6.67 (s, 1H), 7.19–7.26 (m, 5H); 13C NMR (CDCl3,
75 MHz): ı = 17.6, 18.4, 47.9, 50.3, 60.1, 60.6, 61.1, 61.6, 67.1, 67.2,
120.8, 124.5, 128.0, 128.2, 128.5, 136.1, 156.0, 171.2, 172.1; FIA-
ESI(+)-TOF-MS: m/z [M + H]+ calcd for C16H20F3N2O5: 377.1319;
found: 377.1341.
2.2.2. Cbz-d-Phe-l-Pro-NH2
White solid; Rt(HPLC) 16.54 min; 1H NMR (CDCl3, 300 MHz):
ı = 1.49–2.12 (m, 5H), 2.50 (q, 1H, J = 7.5 Hz), 2.93 (d, 2H, J = 6.6 Hz),
3.49–3.54 (m, 1H), 4.33–4.50 (m, 2H), 4.98 (d, 2H, J = 3.3 Hz), 5.41
(s, 1H), 5.85 (d, 1H, J = 6.6 Hz), 6.71 (s, 1H), 7.14–7.27 (m, 11H);
13C NMR (CDCl3, 75 MHz): ı = 24.4, 28.2, 38.4, 47.8, 54.6, 60.0, 67.1,
127.3, 127.9, 128.2, 128.5, 128.6, 129.3, 135.7, 136.1, 156.4, 171.4,
173.5; FIA-ESI(+)-TOF-MS: m/z [M + H]+ calcd for C22H26N3O4:
396.1918; found: 396.1920.
2.5. Dipeptide synthesis with simultaneous use of Alcalase-CLEA
and Cal-B
To a mixture of Alcalase-CLEA (25 mg) and Cal-B (100 mg) in
˚
acetonitrile (3 mL) was added 3 A molsieves (200 mg) and the
appropriate alcohol (200 mg). Then, the N-Cbz-protected amino
acid (50 mg) and subsequently the amino acid amide (1.0 equiv)
were added. The obtained reaction mixture was shaken at 150 rpm
at 50 ◦C for 16 h.
2.2.3. Cbz-d-Phe-l-Leu-NH2
2.6. tBu-ester hydrolysis using Alcalase-CLEA
White solid; Rt(HPLC) 18.37 min; 1H NMR (DMSO-d6, 300 MHz):
ı = 0.77 (dd, 6H, J = 10.2 and 6.3 Hz), 1.24–1.43 (m, 3H), 2.74–2.95
(m, 2H), 4.13–4.32 (m, 2H), 4.95 (s, 1H), 7.01 (s, 1H), 7.25–7.32
(m, 11H), 7.57 (d, 1H, J = 7.8 Hz), 8.14 (d, 1H, J = 8.1 Hz); 13C NMR
(DMSO-d6, 75 MHz): ı = 21.2, 23.0, 23.9, 37.4, 50.6, 56.2, 65.2, 126.1,
127.4, 127.6, 127.9, 128.2, 129.1, 136.8, 137.5, 155.8, 171.0, 173.9;
FIA-ESI(+)-TOF-MS: m/z [M + H]+ calcd for C23H30N3O4: 412.2231;
found: 412.2215.
Alcalase-CLEA (20 mg) was suspended in dioxane/water (2 mL,
9/1, v/v) or DMF/water (2 mL, 1/1, v/v) and then the dipeptide
tBu-ester (50 mg) was added. The reaction mixture was shaken at
150 rpm at 37 ◦C for 16 h. Purification of the dipeptide carboxylic
acid was performed as described in Section 2.2.
3. Results and discussion
2.2.4. Cbz–l-Val–l-Pro-NH2
White solid; Rt(HPLC) 14.52 min; 1H NMR (CDCl3, 300 MHz):
ı = 0.90 (dd, 6H, J = 11.1 and 6.9 Hz), 1.87–2.31 (m, 6H), 3.50–3.58
(m, 1H), 3.64–3.70 (m, 1H), 4.24–4.29 (m, 1H), 4.50–4.54 (m, 1H),
5.02 (d, 2H, J = 5.1 Hz), 5.45–5.51 (m, 2H), 6.74 (s, 1H), 7.19–7.29 (m,
6H); 13C NMR (CDCl3, 75 MHz): ı = 18.3, 18.9, 24.4, 29.1, 29.7, 46.9,
57.8, 59.1, 65.3, 79.1, 127.5, 127.6, 128.2, 137.0, 156.1, 170.0, 173.3;
FIA-ESI(+)-TOF-MS: m/z [M + H]+ calcd for C18H26N3O4: 348.1918;
found: 348.1908.
3.1. Alcalase-CLEA catalyzed peptide synthesis using Cam and Tfe
esters
As reported previously by us, Alcalase-CLEA mediated peptide
synthesis was highly efficient in anhydrous organic solvents for the
coupling of N-terminally protected amino acid C-terminal methyl
esters with C-terminally protected amino acid nucleophiles [17].
Encouraged by these results, we decided to explore the scope of
coupling reactions in which the poor nucleophile proline was used.
Although the subtilisin catalyzed coupling of proline was described
in the literature [18], others observed that proline as nucleophile
gave no conversion at all [19]. In our hands, coupling of Cbz-l-
Phe-OMe with H-l-Pro-OtBu in the presence of Alcalase-CLEA in
anhydrous THF gave an almost quantitative conversion (98%) after
24 h to the dipeptide) as judged by HPLC analysis, and the dipeptide
Cbz-l-Phe-l-Pro-OtBu was isolated in 92% yield. Noteworthy, only
50 mol% excess of H-l-Pro-OtBu was used, which is a rather small
amount for enzymatic coupling reactions. Despite these promising
data, in case of more challenging acyl donors like Cbz-d-Phe-OMe
or Cbz-l-Val-OMe, the conversion toward dipeptides Cbz-d-Phe-
Pro-OtBu and Cbz-l-Val-Pro-OtBu dropped significantly to 24% and
32%, respectively. In order to improve the coupling yield, we inves-
tigated the versatility of Cam and Tfe active esters in Alcalase-CLEA
mediated peptide synthesis using a number of challenging acyl
2.2.5. Cbz-l-Ala-l-Ala-Gly-l-Phe-NH2
White solid; Rt(HPLC) 14.28 min; 1H NMR (DMSO-d6, 300 MHz):
ı = 1.16–1.29 (m, 6H), 2.74–3.06 (m, 2H), 3.56–3.77 (m, 2H),
4.05–4.12 (m, 1H), 4.22–4.46 (m, 1H), 5.02 (d, 2H, J = 6.9 Hz),
7.10–7.48 (m, 14H), 7.93–8.21 (m, 3H); 13C NMR (DMSO-d6,
75 MHz): ı = 17.9, 48.2, 53.7, 65.3, 65.4, 126.1, 127.6, 127.9, 128.2,
129.0, 136.9, 137.9, 138.0, 155.5, 168.2, 168.3, 172.3, 172.6, 172.7;
FIA-ESI(+)-TOF-MS: m/z [M + H]+ calcd for C25H32N5O6: 498.2347;
found: 498.2346.
2.3. Cam or Tfe ester synthesis using Alcalase-CLEA
Alcalase-CLEA (100 mg) was added to a mixture of MTBE
˚
or THF (3 mL), 3 A molsieves (200 mg), glycolamide or 2,2,2-
trifluoroethanol (200 mg), and N-Cbz-protected amino acid

82
T. Nuijens et al. / Journal of Molecular Catalysis B: Enzymatic 71 (2011) 79–84
Scheme 1. Alcalase mediated synthesis of dipeptides using Tfe and Cam esters.
Table 1
Table 2
Alcalase-CLEA mediated dipeptide synthesis using Cam and Tfe estersa.
Alcalase-CLEA mediated dipeptide synthesis with H-l-Pro-NH2 as nucleophilea.
Entry Acyl donor
Nucleophile
Dipeptide
Yield (%)b
Entry
Acyl donor
Initial rateb
Conversione (%)
1
2
3
4
5
6
7
8
Cbz-l-Phe-OCam
H-l-Leu-OtBu Cbz-l-Phe-l-Leu-OtBu 87
1
2
3
4
5
6
7
8
Cbz-l-Ala-OMe
Cbz-l-Ala-OTfe
Cbz-l-Ala-OCam
Cbz-l-Phe-OMe
Cbz-l-Phe-OTfe
Cbz-l-Phe-OCam
Cbz-d-Ala-OMe
Cbz-d-Ala-OTfe
Cbz-d-Ala-OCam
Cbz-d-Phe-OMe
Cbz-d-Phe-OTfe
Cbz-d-Phe-OCam
Cbz-l-Val-OMe
Cbz-l-Val-OTfe
Cbz-l-Val-OCam
10.6
11.9
12.8
11.0
12.8
13.1
1.7
2.8
5.1
0.8
1.9
77c
90c
95c
81c
93c
98c
35d
52d
97d
12d
29d
65d
11d
21d
50d
Cbz-l-Phe-OCam
Cbz-l-Phe-OCam
Cbz-l-Phe-OCam
Cbz-l-Phe-OTfe
Cbz-l-Phe-OTfe
Cbz-l-Ala-OTfe
Cbz-l-Ala-OTfe
Cbz-d-Ala-OTfe
Cbz-d-Ala-OCam
Cbz-d-Ala-OCam
Cbz-d-Phe-OTfe
H-l-Pro-OtBu
H-l-Leu-NH2
H-l-Pro-NH2
H-l-Pro-NH2
H-l-Pro-NH2
H-l-Leu-NH2
H-l-Pro-NH2
H-l-Leu-NH2
H-l-Leu-NH2
H-l-Pro-NH2
H-l-Leu-NH2
Cbz-l-Phe-l-Pro-OtBu
Cbz-l-Phe-l-Leu-NH2
Cbz-l-Phe-l-Pro-NH2
Cbz-l-Phe-l-Pro-NH2
Cbz-l-Phe-l-Pro-NH2
Cbz-l-Ala-l-Leu-NH2
Cbz-l-Ala-l-Pro-NH2
Cbz-d-Ala-l-Leu-NH2
Cbz-d-Ala-l-Leu-NH2
Cbz-d-Ala-l-Pro-NH2
Cbz-d-Phe-l-Leu-NH2
Cbz-d-Phe-l-Leu-NH2
Cbz-d-Phe-l-Pro-NH2
Cbz-l-Val-l-Leu-NH2
Cbz-l-Val-l-Pro-NH2
92
91
91
90
93
94
90
45
92
69
36
93
50
93
76
9
9
10
11
12
13
14
15
16
10
11
12
13
14
15
3.5
0.9
1.4
2.9
Cbz-d-Phe-OCam H-l-Leu-NH2
Cbz-d-Phe-OCam H-l-Pro-NH2
Cbz-l-Val-OCam
Cbz-l-Val-OCam
H-l-Leu-NH2
H-l-Pro-NH2
a
Reaction conditions: see Section 2.2, only 25 mg of Alcalase-CLEA was used and
the products were not isolated
a
Reaction conditions: see Section 2.2.
Isolated yield.
b
b
Initial rates were determined by HPLC and are expressed as %conversion per
hour per 25 mg of Alcalase-CLEA.
c
HPLC conversion after: 10 h.
HPLC conversion after: 24 h.
Conversions to dipeptide product compared to acyl donor starting material.
d
donors and nucleophiles, as is shown in Table 1 and Scheme 1. Gp
esters were not included in this study because they have poor solu-
bility in anhydrous organic solvents and are recognized by arginine
specific proteases, such as chymotrypsin and not by Alcalase.
Evidently, the Cam as well as Tfe active esters were found to be
very good acyl donors in Alcalase-CLEA mediated dipeptide syn-
theses. Good yields were obtained in entries 1–8 and in entries
9–16 using more challenging acyl donors. Not only d-amino acids
but also valine could be used in combination with the poor nucle-
ophilic proline. There was almost no difference in efficiency in case
of H-l-Leu-NH2 or H-l-Pro-NH2 as the nucleophilic species when
the acyl donors Cbz-l-Phe and Cbz-l-Ala were used irrespective of
the active ester. However, in case of more challenging acyl donors
as Cbz-d-Phe, Cbz-d-Ala and Cbz-d-Val, the Cam esters turned out
to be superior, while the coupling efficiency of proline was lower
compared to leucine. These results triggered us to investigate the
type of active ester in more detail in order to optimize such difficult
coupling reactions.
e
Therefore, we tried to synthesize Cbz-l-Phe-OCam as well as
Cbz-l-Phe-OTfe in the presence of the respective alcohols and
Alcalase-CLEA (reaction conditions see Section 2.3). Both esters
could be obtained, however, their formation was rather slow and
as a result of this the yield was relatively low (24% and 57%,
respectively). Since the lipase Cal-B is known to catalyze ester
synthesis in anhydrous organic solvents, we used this enzyme
for the synthesis of the desired Cam and Tfe active esters. We
found that Cal-B was highly efficient, and several amino acids
as well as dipeptides were accepted as substrate, as shown in
Table 3.
3.3. Fully enzymatic peptide synthesis
Since the active esters were synthesized enzymatically, and
on their turn can be used as versatile acyl donors in the next
3.2. Differences between the amino acid methyl, Cam and Tfe
ester species and their respective enzymatic synthesis
Table 3
Cal-B mediated synthesis of Tfe and Cam estersa.
In order to optimize the efficiency of the enzyme mediated
dipeptide synthesis, the coupling of several amino acid methyl,
Cam, and Tfe esters to H-l-Pro-NH2 was investigated in the pres-
ence of Alcalase-CLEA, as shown in Table 2.
Gratifyingly, these results showed that the Cam esters, and to
a lesser extent the Tfe esters, were in all cases more efficient acyl
donors than the commonly used methyl ester.
During the course of our research we found that much more
amino acid esters like methyl, ethyl, benzyl, as well as tert-butyl
were accessible by Alcalase-CLEA mediated ester synthesis [20].
Entry
Ester product
Yield (%)b
1
2
3
4
5
6
7
Cbz-Gly-OTfe
93
91
96
80
93
65
77
Cbz-l-Pro-OTfe
Cbz-l-Ala-OTfe
Cbz-l-Ala-OCam
Cbz-d-Ala-OTfe
Cbz-d-Ala-OCam
Cbz-l-Ala-l-Ala-OTfe
a
Reaction conditions: see Section 2.4.
Isolated yield.
b

T. Nuijens et al. / Journal of Molecular Catalysis B: Enzymatic 71 (2011) 79–84
83
Table 4
Fully enzymatic peptide synthesis using Cal-B and Alcalase-CLEA.
Entry
Ester producta
Peptide productb
Yield (%)c
Deprotected productd
Yield (%)e
1
2
3
4
5
6
7
8
Cbz-Gly-OTfe
Cbz-Gly-OTfe
Cbz-Gly-OTfe
Cbz-l-Ala-OTfe
Cbz-l-Ala-OTfe
Cbz-l-Phe-OTfe
Cbz-l-Ala-l-Ala-OTfe
Cbz-d-Ala-OCam
Cbz-Gly-l-Phe-NH2 [22]
Cbz-Gly-l-Leu-NH2 [22]
Cbz-Gly-l-Phe-OtBu
89
88
90
91
92
87
85
75
Cbz-Gly-l-Phe-OH [22]
Cbz-l-Ala-l-Leu-OH [22]
Cbz-l-Ala-l-Phe-OH [22]
Cbz-l-Phe-l-Leu-OH [22]
88
86
90
84
Cbz-l-Ala-l-Leu-OtBu
Cbz-l-Ala-l-Phe-OtBu
Cbz-l-Phe-l-Leu-OtBu
Cbz-l-Ala-l-Ala-Gly-l-Phe-NH2
Cbz-d-Ala-d-Ala-OtBu
a
Reaction conditions: Sections 2.4 and 2.3 for entry 6.
Reaction conditions Section 2.2.
Purified yield compared to acyl donor.
Reaction conditions: Section 2.6.
b
c
d
e
Purified yields compared to acyl donor.
coupling step, we explored the fully enzymatic synthesis of sev-
eral biologically interesting peptides, as shown in Table 4 and
Scheme 2.
Cal-B and Alcalase-CLEA in a one pot approach. Indeed, starting
from an N-terminally protected amino acid as the acyl donor and
an amino acid amide as the nucleophile, in the presence of both
enzymes with either 2,2,2-trifluoroethanol or glycolamide as addi-
tive, the corresponding dipeptide was obtained, as shown in Table 5
and Scheme 3.
Although Alcalase-CLEA and Cal-B were separately capable of
catalyzing dipeptide formation, the obtained yields were signifi-
cantly higher if both enzymes were combined, probably because
the esterification equilibrium as mediated by Cal-B is shifted to
the ester side since the ester is consumed in the Alcalase-CLEA
mediated coupling reaction. Without addition of an alcohol the
dipeptide yields remained low. No deactivation of the enzymes was
observed, probably due to the fact both enzymes were immobilized
and water was absent. For optimal results, only one equivalent of
nucleophile was necessary. Thus, by applying this two-enzyme-
one-pot approach, dipeptides were obtained in high yields starting
from readily accessible amino acids with a free carboxylic acid func-
tionality acyl donor and an amino acid with a free N␣-amino moiety
nucleophile.
Dipeptide amides were easily accessible (entries 1–2), as well as
dipeptide tert-butyl esters (entries 3–6). With respect to the latter,
enzymatic hydrolysis [21] by Alcalase-CLEA gave the dipeptide acid
that could be used as substrate for Cal-B to give the corresponding
Tfe active ester as the dipeptide acyl donor for the next coupling
step. Interestingly, also a dipeptide Tfe ester was recognized by the
enzyme and reacted with a dipeptide amide to give a tetrapep-
tide in high yield (entry 7). Furthermore, not only proteinogenic
amino acids, also H-d-alanine tert-butyl ester was recognized by
the enzyme as a nucleophile to give the all d-dipeptide, Cbz-d-Ala-
d-Ala-OtBu, (entry 8).
3.4. Simultaneous esterification and peptide coupling:
two-enzymes-one-pot approach
Finally, we investigated the possibility of performing the ester-
ification and subsequent dipeptide formation in the presence of
Scheme 2. Cal-B or Alcalase-CLEA mediated Tfe or Cam ester synthesis followed by Alcalase-CLEA catalyzed peptide synthesis and subsequent tert-butyl ester hydrolysis
according to literature reference [21] (for entry 3–6 Table 4).

84
T. Nuijens et al. / Journal of Molecular Catalysis B: Enzymatic 71 (2011) 79–84
Table 5
Dipeptide synthesis using a Cal-B and Alcalase-CLEA two-enzyme-one-pot approach with 2,2,2-trifluoroethanol or glycolamide as additivea.
Product
Enzyme(s)
Alcohol
Ester (%)
Dipeptide (%)
Cbz-l-Ala-l-Leu-NH2
Cbz-l-Ala-l-Leu-NH2
Cbz-l-Ala-l-Leu-NH2
Cbz-l-Ala-l-Pro-NH2
Cbz-l-Ala-l-Pro-NH2
Cbz-l-Ala-l-Pro-NH2
Cbz-l-Ala-l-Leu-NH2
Cbz-l-Ala-l-Leu-NH2
Cbz-l-Ala-l-Pro-NH2
Cbz-l-Ala-l-Pro-NH2
Alcalase-CLEA
Cal-B
Cal-B and Alcalase-CLEA
Alcalase-CLEA
Cal-B
Cal-B and Alcalase-CLEA
Alcalase-CLEA
Cal-B and Alcalase-CLEA
Alcalase-CLEA
Tfe-OH
Tfe-OH
Tfe-OH
Tfe-OH
Tfe-OH
Tfe-OH
Cam-OH
Cam-OH
Cam-OH
Cam-OH
1
30
2
0
32
1
0
1
0
3
13
31
87
9
14
27
15
88
11
31
Cal-B and Alcalase-CLEA
a
Reaction conditions: Section 2.5, conversions were calculated by HPLC assuming that absorption coefficients of starting material and products were identical.
Scheme 3. Simultaneous esterification and peptide synthesis using a two-enzyme-one-pot approach.
4. Conclusions
[3] D. Hans, P.R. Young, D.D.P. Fairlie, Med. Chem. 2 (2006) 627–646.
[4] F. Guzmán, S. Barberis, A. Illanes, J. Biotechnol. 10 (2007) 279–314.
[5] F. Bordusa, Chem. Rev. 102 (2002) 4817–4867.
[6] V. Cerˇovsky, F. Bordusa, J. Pept. Res. 55 (2000) 325–329.
[7] T. Miyazawa, E. Ensatsu, N. Yabuuchi, R. Yanagihara, T. Yamada, J. Chem. Soc.
Perkin Trans. 1 (2002) 390–395.
Herein, we have shown that Cam and Tfe active esters are very
useful to achieve highly efficient Alcalase-CLEA mediated peptide
synthesis. These active esters allow the use of sterically demanding
and non-proteinogenic acyl donors as well as poor nucleophiles,
and combinations thereof. Furthermore, the Cam and Tfe active
esters based on amino acids can be enzymatically synthesized
by the lipase Cal-B. Finally, a fully enzymatic peptide synthesis
approach was developed by combination of two enzymes in which
the esterification is performed by Cal-B, while Alcalase-CLEA is
responsible for peptide synthesis, also in a two-enzyme-one-pot
approach. We are currently exploring whether this technology can
be used for the synthesis of longer peptides using also the more
hydrophilic amino acid residues.
ˇ
[8] T. Miyazawa, S. Nakajo, M. Nishikawa, K. Hamahara, K. Imagawa, E. Ensatsu, R.
Yanagihara, T. Yamada, J. Chem. Soc. Perkin Trans. 1 (2001) 82–86.
[9] (a) S.M.A. Salam, K.-I. Kagawa, K. Kawashiro, Biotechnol. Lett. 27 (2005)
1199–1203;
(b) F. Bordusa, Braz. J. Med. Biol. Res. 33 (2000) 469–485.
[10] M. Capellas, M.D. Benaiges, G. Caminal, G. Gonzalez, J. Lopez-Santín, P. Clapés,
Biotechnol. Bioeng. 50 (2000) 700–708.
[11] M.K. Dhaon, R.K. Olsen, K. Ramasamy, J. Org. Chem. 47 (1982) 1962–1965.
[12] R.A. Sheldon, Biochem. Soc. Trans. 35 (2007) 1583.
[13] W. König, R. Geiger, Chem. Ber. 103 (1970) 788–798.
[14] W. H. J. Boesten, P.J. L. M. Quaedflieg, WO Patent No. 6222013, 1999.
[15] T. Miyazawa, E. Ensatsu, M. Hiramatsu, R. Yanagihara, T. Yamada, J. Chem. Soc.
Perkin Trans. 1 (2002) 396–401.
[16] O.J. Park, H.G. Park, J.W. Yang, Biotechnol. Lett. (1996) 473–478.
[17] T. Nuijens, C. Cusan, T.J.G.M. van Dooren, H.M. Moody, R. Merkx, J.A.W. Kruijtzer,
D.T.S. Rijkers, R.M.J. Liskamp, P.J.L.M. Quaedflieg, Adv. Synth. Catal. 352 (2010)
2399–2404.
Appendix A. Supplementary data
[18] S.T. Chen, S.Y. Chen, C.L. Kao, K.T. Wang, Bioorg. Med. Chem. Lett. 4 (1994)
443–448.
[19] J.U. Klein, A. Prykhodzka, V. Cerˇovsky, J. Pept. Sci 6 (2000) 541–549.
[20] (a) T. Nuijens, C. Cusan, J.A.W. Kruijtzer, D.T.S. Rijkers, R.M.J. Liskamp, P.J.L.M.
Quaedflieg, Synthesis 5 (2009) 809–814;
(b) T. Nuijens, J.A.W. Kruijtzer, C. Cusan, D.T.S. Rijkers, R.M.J. Liskamp, P.J.L.M.
Quaedflieg, Tetrahedron Lett. 50 (2009) 2719–2721.
[21] F. Gini, I.F. Eggen, D.J. van Zoelen, C. Boeriu, Chimica Oggi 27 (2009)
24–26.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molcatb.2011.03.012.
ˇ
References
[1] (a) B. Steffansen, C.U. Nielsen, S. Frokjaer, Eur. J. Pharm. Biopharm. 60 (2005)
241–245;
(b) K. Adermann, Curr. Prot. Pept. Sci. 6 (2005) 205–206.
[2] (a) H. Korhonen, A. Pihlanto, Curr. Pharm. Des. 9 (2003) 1297–1308;
(b) J. Torres-Llanez, B. Vallejo-Cordoba, A.F. González-Córdova, Arch. Latinoam.
Nutr. 55 (2005) 111–117;
[22] (a) D.I. Mundy, W.J. Strittmatter, Cell 40 (1985) 645–656;
(b) A. Koide, M. Yoshizawa, K. Kurachi, Eur. J. Biochem. 117 (1981)
383–388;
(c) K. Olesen, M.C. Kielland-Brandt, Protein Eng. 6 (1993) 409–415;
(d) Y. Bai, R. Hayashi, J. Biol. Chem. 254 (1979) 8473–8479.
(c) H. Korhonen, A. Pihlanto, Curr. Pharm. Des. 13 (2007) 829–843;
(d) M.B. Roberfroid, J. Nutr. 137 (2007) 2493–2502.