Papain-Specific Activating Esters in Aqueous Dipeptide Synthesis

Article abstract of DOI:10.1002/cbic.201200017

Enzymatic peptide synthesis has the potential to be a viable alternative for chemical peptide synthesis. Because of the increasing commercial interest in peptides, new and improved enzymatic synthesis methods are desirable. In recently developed enzymatic strategies such as substrate mimetic approaches and enzyme-specific activation, use of the guanidinophenyl ester (OGp) group has been shown to suffer from some drawbacks. OGp esters are sensitive to spontaneous chemical hydrolysis and the group is expensive to synthesize and therefore not suitable for large-scale applications. On the basis of earlier computational studies, we hypothesized that OGp might be replaceable by simpler ester groups to make the enzyme-specific activation approach to peptide bond formation more accessible. To this end, a set of potential activating esters (Z-Gly-Act) was designed, synthesized, and evaluated. Both the benzyl (OBn) and the dimethylaminophenyl (ODmap) esters gave promising results. For these esters, the scope of a model dipeptide synthesis reaction under aqueous conditions was investigated by varying the amino acid donor. The results were compared with those obtained from a previous study of Z-X_AA-OGp esters. Computational docking analysis of the set of esters was performed in order to provide insight into the differences in the reactivities of all the potential activating esters. Finally, selected ODmap- and OBn-activated amino acids were applied in the synthesis of two biologically active dipeptides on preparative scales.

Full text of DOI:10.1002/cbic.201200017

DOI: 10.1002/cbic.201200017
Papain-Specific Activating Esters in Aqueous Dipeptide
Synthesis
Roseri J. A. C. de Beer,[a] Barbara Zarzycka,[b] Michiel Mariman,[a] Helene I. V. Amatdjais-
Groenen,[a] Marc J. Mulders,[a] Peter J. L. M. Quaedflieg,[c] Floris L. van Delft,[a]
Sander B. Nabuurs,[b] and Floris P. J. T. Rutjes*[a]
Enzymatic peptide synthesis has the potential to be a viable al-
ternative for chemical peptide synthesis. Because of the in-
creasing commercial interest in peptides, new and improved
enzymatic synthesis methods are desirable. In recently devel-
oped enzymatic strategies such as substrate mimetic ap-
proaches and enzyme-specific activation, use of the guanidino-
phenyl ester (OGp) group has been shown to suffer from some
drawbacks. OGp esters are sensitive to spontaneous chemical
hydrolysis and the group is expensive to synthesize and there-
fore not suitable for large-scale applications. On the basis of
earlier computational studies, we hypothesized that OGp
might be replaceable by simpler ester groups to make the
enzyme-specific activation approach to peptide bond forma-
tion more accessible. To this end, a set of potential activating
esters (Z-Gly-Act) was designed, synthesized, and evaluated.
Both the benzyl (OBn) and the dimethylaminophenyl (ODmap)
esters gave promising results. For these esters, the scope of a
model dipeptide synthesis reaction under aqueous conditions
was investigated by varying the amino acid donor. The results
were compared with those obtained from a previous study of
Z-XAA-OGp esters. Computational docking analysis of the set of
esters was performed in order to provide insight into the dif-
ferences in the reactivities of all the potential activating esters.
Finally, selected ODmap- and OBn-activated amino acids were
applied in the synthesis of two biologically active dipeptides
on preparative scales.
Introduction
The market share of peptide-based drugs is steadily increasing
and will likely continue to do so, because a large number of
therapeutic peptides are in clinical trials.[1] Other relevant appli-
cations of peptides include their use as nutrients[2] and cosmet-
ic ingredients.[3] Large-scale peptide synthesis, however, is an
expensive and time-consuming procedure, and this conflicts
with the considerable commercial interest.[4]
es of such a chemoenzymatic approach include the high selec-
tivities of the enzymes, minimal need for side chain protection,
and absence of racemization. On the other hand, the generally
narrow substrate specificities of enzymes, in particular the ex-
clusion of unnatural amino acids, severely restricts their univer-
sal application. In addition, undesired product hydrolysis when
using proteases can be problematic, even though strategies
have been developed to minimize this side reaction.[9] By em-
ploying the principle of substrate mimetics,[10] the challenge of
inadequate substrate recognition can be overcome. In the sub-
strate mimetics approach, the amino acid side chain—crucial
for enzyme recognition—is transferred to the ester leaving
group so that the enzymatic process becomes independent of
the amino acid to be coupled. The guanidinophenyl (OGp)
group (Scheme 1, below), a mimetic of the arginine side chain,
has been extensively applied in academic research with much
Solid-phase peptide synthesis (SPPS) is the most commonly
used and most versatile strategy for the synthesis of peptides
in general.[5] The requirement for side-chain protection, the
need to use excesses of reagents, and the risk of racemization
are downsides of this method, but are to a large extent com-
pensated by its capability to incorporate virtually any amino
acid. A strategy commonly used in cases of larger peptides is
to produce shorter fragments by SPPS and to couple them in
solution to make the synthesis more convergent.[6] Alternative-
ly, fragments with unprotected side chains can be coupled in
aqueous solution by the chemoselective native chemical liga-
tion method, which involves a C-terminal thioester and an N-
terminal cysteine.[7] For large-scale production of long peptides
(>50 amino acid residues) and protein sequences containing
only natural amino acids, fermentation is currently most feasi-
ble, but this requires significant development efforts for each
individual peptide or protein.[8]
[a] R. J. A. C. de Beer, M. Mariman, H. I. V. Amatdjais-Groenen, M. J. Mulders,
Dr. F. L. van Delft, Prof. F. P. J. T. Rutjes
Institute for Molecules and Materials, Radboud University Nijmegen
Heyendaalseweg 135, 6525 AJ Nijmegen (The Netherlands)
E-mail: f.rutjes@science.ru.nl
[b] B. Zarzycka, Dr. S. B. Nabuurs
Computational Drug Discovery
Center for Molecular and Biomolecular Informatics
Radboud University Nijmegen Medical Centre
P.O. Box 9101, 6500 HB Nijmegen (The Netherlands)
A relatively new and rather unexplored area is the use of en-
zymatic hydrolysis of readily available existing proteins to re-
lease specific peptides.[4a] An alternative use of these proteases,
though, is their potential to create peptide bonds.[9] Advantag-
[c] Dr. P. J. L. M. Quaedflieg
DSM Innovative Synthesis BV
P.O. Box 18, 6160 MD Geleen (The Netherlands)
ChemBioChem 2012, 13, 1319 – 1326
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1319

F. P. J. T. Rutjes et al.
success.[11] However, the OGp group is unsuitable for industrial
application because it is expensive and difficult to work with
due to the positively charged (under relevant conditions) gua-
nidino group.
alities (e.g., hydrogen bond donors or acceptors) were incorpo-
rated to allow for interactions with the binding groove of
papain. This is also important for gaining insight into the spa-
tial, electronic, and chemical requirements for papain-specific
activating esters. Secondly, the stability to spontaneous ester
hydrolysis needs to be improved. Chemical background hydrol-
ysis for OGp esters is substantial (ꢀ5% within 30 min)[12] and
this is undesirable because it will increase when longer reac-
tion times are required in more difficult couplings. Aliphatic or
benzylic esters are considerably more stable than phenolic
esters. Thirdly, the new activating esters must preferably be
cheap and easy to synthesize. A chemoenzymatic route—using
Alcalase-CLEA,[13] for example—would be particularly attractive
in this respect. N-Protected amino acids can be converted into
their alkyl or benzyl esters by this enzyme in near-anhydrous
organic solvents. Furthermore, arylamides can be obtained by
ester interconversion under the same conditions. The more
labile phenyl esters are beyond the scope of this method.
These considerations led to a set of potentially activating
moieties, shown in Scheme 1. These were chemically synthe-
sized as the corresponding Z-Gly-OH esters. In addition, four
activated esters that have been described in the literature in
combination with papain—OMe,[14] OCam,[14a,15] OTfe,[15] and
OBn,[14a,16]—were also included for comparison. The synthesis
details of all ester substrates can be found in the Supporting
Information.
We recently reported that the OGp ester group can also be
used in chemoenzymatic dipeptide synthesis under aqueous
conditions with the protease papain.[12] In this case, the OGp
moiety is likely not mimicking the natural substrate, because
the guanidine group is predicted to bind to papain in
a manner different from that of the arginine side chain. We
named this phenomenon enzyme-specific activation, because
no reaction takes place in the absence of the enzyme. From
these findings we hypothesized that it might not be essential
to have an arginine-like activating ester for papain, as long as
sufficient specific interactions with the enzyme can be pro-
duced. A set of potentially enzyme-specific activating esters
was therefore designed, synthesized, and evaluated for effec-
tiveness in papain-catalyzed dipeptide formation.
Results and Discussion
Design of potentially enzyme-specific activating esters
In designing enzyme-specific activating esters for papain, sev-
eral requirements had to be fulfilled. Firstly, sufficient function-
Scheme 1. Structures of potential activating esters. OGp: p-guanidinophenyl ester. OGb: p-guanidinobenzyl ester. NGp: p-guanidinobenzyl amide. OAb: p-ami-
dinobenzyl ester. OBn: benzyl ester. OCam: carbamoylmethyl ester. O3Cam: carbamoylpropyl ester. O4Cam: carbamoylbutyl ester. O3G: 3-guanidinopropyl
ester. O4G: 4-guanidinobutyl ester. O5G: 5-guanidinopentyl ester. ODmap: p-(dimethylamino)phenyl ester. OTmap: p-(trimethylammonium)phenyl ester.
ODmape: p-(dimethylamino)phenethyl ester. OTmape: p-(trimethylammonium)phenethyl ester. O4A: 4-aminobutyl ester. O5A: 5-aminopentyl ester. OMe:
methyl ester. OTfe: 2,2,2-trifluoroethyl ester.
1320
www.chembiochem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2012, 13, 1319 – 1326

Papain-Specific Activating Esters
Evaluation of potentially enzyme-specific activating esters
enzymatic synthesis and hydrolysis remained constant over
time as measured after 24 h.
Next, the esters were evaluated for their ability to form the di-
peptide Z-Gly-Phe-NH2 with papain catalysis. The levels of con-
version were determined by a previously described assay
(Scheme 2).[12]
Ideally, the enzyme-specific activating group should give a
fast reaction, a high percentage of synthesis, and no hydrolysis.
In none of the entries are all these criteria met at the same
time, except for the phenolic ODmap ester (entry 6). However,
the liberated dimethylaminophenol suffers from decomposi-
tion, as deduced from the large amount of additional peaks in
the HPLC chromatogram. The phenolic ester OTmap (entry 5)
reacted comparably rapidly as OGp, but had a higher sponta-
neous hydrolysis rate. The introduction of two extra carbons
(cf. ODmape and OTmape, entries 7 and 8, respectively) result-
ed in significant increases in reaction time. The benzyl ester
(OBn, entry 17) was superior to OGp in many respects, except
for the time needed to reach full conversion of the starting
material. The other benzylic esters, OGb and OAb (entries 2
and 4, respectively), reacted even more slowly and showed dif-
ferent percentages of enzymatic hydrolysis, which is remark-
able in view of their similarity. The introduction of an amide
bond (entry 3) resulted in complete inactivity. Of the set of ali-
phatic potentially activating esters, only the activated OCam
and OTfe esters (entries 14 and 18, respectively) reacted rea-
sonably rapidly. As anticipated from their chemical properties,
background hydrolysis was low for all aliphatic esters. The
enzymatic hydrolysis levels, on the other hand, were much
higher in all cases. These results are in line with earlier studies
showing that the Z-Gly esters of OMe,[14b] OBn,[16] OCam,[14a]
and OTfe[15] can be used in combination with papain. Because
of the variation in documented reaction conditions—in organic
solvent, in frozen aqueous solutions, varying equivalents, im-
mobilized papain—it was, however, difficult to compare them
properly. The fact that these re-
Scheme 2. Enzymatic dipeptide synthesis with papain. a) 3.5 mm papain,
2 mm pTSA, 10% (v/v) DMF, 0.2m HEPES buffer (pH 8), 0.2m NaCl, 20 mm
CaCl2.
Table 1 shows either the times it takes to reach 100% con-
version or the levels of conversion after 3 h. Background hy-
drolysis of the esters was determined from blank reactions in
which no papain was present. The indicated percentages of
actions could not just be per-
Table 1. Various experimentally tested Z-Gly-Act esters.[a]
formed in buffer is consistent
with the relatively high enzymat-
ic hydrolysis values we find for
these esters.
Although the enzymatic reac-
tions proceed through the same
acyl-enzyme intermediate (Z-Gly-
papain), differences in S/H ratio
are observed, This might be be-
cause the first step (acylation of
the enzyme) is not the rate-de-
termining step for each ester.
Another explanation might be
that the leaving groups have
varying affinities for the enzyme
and therefore influence the
attack of the nucleophile (water
or H-Phe-NH2) to larger or small-
er extents.
Experimental
t [min] Conv [%]
Background
Z-Gly-OH [%]
Enzymatic
Z-Gly-Phe-NH2 [%]
Act
Z-Gly-OH [%]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
OGp
OGb
NGp
OAb
OTmap
ODmap
OTmape
ODmape
O3G
O4G
O5G
O4A
O5A
OCam
O3Cam
O4Cam
OBn
OTfe
OMe
20
180
180
180
15
100
97
–
5.6
3.3
–
92.0
92.5
–
2.4
1.2
–
10.3
0
0.2
99
2.2
8.4
1.2
0.4
0.1
1.8
1.6
1.2
2.5
0.6
1.8
1.1
0.2
0
86.5
91.6
98.6
58.1
18.9
68.6
66.8
54.7
53.9
56.7
89.9
67.8
81.9
97.5
91.5
58.8
100
100
61
21
86
83
66
69
70
100
80
97
100
100
70
10
180
180
180
180
180
180
180
15
2.5
2.0
15.6
14.6
10.1
12.6
12.7
8.3
11.1
14.9
2.5
180
180
45
15
180
1.8
1.5
6.7
9.7
[a] Conditions: Z-Gly-Act (2 mm), H-Phe-NH2 (15 mm), papain (3.5 mm), HEPES buffer [pH 8.0, 0.2m, NaCl (0.2m),
CaCl2 (20 mm), DMF (10%, v/v)].
ChemBioChem 2012, 13, 1319 – 1326
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
1321

F. P. J. T. Rutjes et al.
Figure 1. Molecular modeling of a selection of esters in papain. Hydrogen bonding interactions to functionally important amino acids in the papain active
site are shown for A) Z-Gly-OGp, B) Z-Gly-OGb, C) Z-Gly-OBn, D) Z-Gly-ODmap, E) Z-Gly-O3G, F) Z-Gly-OCam, and G) Z-Gly-OTfe.
Docking of activating esters into the active site of papain
effects of hydrogen bonding interactions between the enzyme
and the substrate.[19] Similarly, the OTfe group (Figure 1G) ex-
We recently developed a molecular model of the interactions
of Z-XAA-OGp esters with the active site of papain.[12] A similar
molecular modeling study was performed for each of the
remaining 18 substrates with the aid of the flexible docking
program Fleksy.[17] Results were visualized and analyzed by use
of the YASARA program. Several representative examples are
shown in Figure 1 and discussed below.
actly fills a hydrophobic cavity in the peptide groove, interact-
ing with Trp26, Gly65. and Ala160.
These results imply that besides a good fit in the active site
with sufficient interactions to the peptide groove and proper
stabilization by the oxyanion hole, the leaving group ability of
the activating ester is also of significant importance. Clearly,
the relatively poor leaving group abilities of the aliphatic
esters cannot be compensated for by the enzyme, whereas the
differences between the much better phenolic leaving groups
can be explained on the basis of their interactions with papain.
The results also show that the good leaving groups OCam and
OTfe are to some extent even further activated by the enzyme.
The outcome with the benzyl ester is remarkable in our view,
because the inferior leaving group ability is largely overcome
by the enzyme-specific activation.
Z-Gly-OGp (Figure 1A) fits tightly in the papain active site,
which is located in a groove between two domains.[12,18] The
nucleophilic cysteine is as such positioned to attack the car-
bonyl carbon, whereas the carbonyl oxygen is situated in the
oxyanion hole, which is made up of the side chain NH2 group
of Gln19 and the backbone amide of Cys25. The guanidino-
phenyl group is positioned in the groove, participating in sev-
eral hydrogen bonding interactions.
In general, all the compounds discussed have the same N-
to-C directionality as Z-Gly-OGp and fit the oxyanion hole.
The most noticeable difference between Z-Gly-OGp and Z-
Gly-OGb (Figure 1B) is that the guanidinobenzyl group is now
pointing out of the groove as a consequence of the extra
carbon atom. As a result, the hydrogen bonding interactions
with the groove are lost, which might explain the significant
change in activity. Similarly, in Z-Gly-OBn (Figure 1C) the
benzyl group is also oriented towards the solvent, but it can
favorably interact with Gly65 at the edge of the groove. Z-Gly-
ODmap (Figure 1D), a phenolic ester, is nicely located in the
groove and interacts with Tyr67 and Val133. Although the ali-
phatic Z-Gly-O3G (Figure 1E) seems to be reasonably well situ-
ated in the groove, its reaction time is significantly longer than
those of the phenolic esters. The small OCam moiety (Fig-
ure 1F) participates in additional interactions with Trp26,
Asp158, and Ala160 and fills the lower part of the peptide
groove. High reactivity for OCam esters with a-chymotrypsin
had already been reported to be attributable to orientation
Scope of amino acid donor in Z-XAA-OBn and Z-XAA-ODmap
The high levels of conversion, combined with low hydrolysis
rates, observed with the readily accessible ODmap and OBn
esters led us to select them for further exploration. In particu-
lar, in order to probe the more general applicability of the con-
cept of enzyme-specific activation, we investigated whether
the relatively broad range of amino acid donors determined
for OGp[12] would be maintained with ODmap and OBn. Z-XAA-
ODmap and Z-XAA-OBn esters were synthesized in a procedure
analogous to that followed for OGp esters, from a representa-
tive set of amino acids with appropriate acid-labile side chain
protection and evaluated in the enzymatic assay. In the case of
OBn esters, the amount of papain was doubled to compensate
for the slower reaction rate that had been observed with Z-
Gly-OBn. The identities of the products in the enzymatic reac-
tions were confirmed by chemical synthesis of reference com-
pounds and LC-MS analysis.
1322
www.chembiochem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2012, 13, 1319 – 1326

Papain-Specific Activating Esters
Z-b-Ala-OH. Increasing the reaction time for the cor-
responding OBn esters did not result in significantly
higher conversions, but dipeptide synthesis occurred
when the quantities of papain were simultaneously
increased by a further factor of ten (Table 2).
The scope of the amino acid donor is comparably
broad for ODmap and OGp esters. The OBn ester
group also enables the papain-catalyzed synthesis of
various dipeptides, provided that the amount of
enzyme is increased. This seems an acceptable trade-
off given the fact that OBn esters are readily accessi-
ble, even with an enzymatic approach.
Synthesis of two selected dipeptides on prepara-
tive scale
The applicability of the enzyme-specific activation
strategy was illustrated by means of preparative scale
syntheses of two bioactive peptides: Z-Gly-Phe-NH2,
with Z-Gly-ODmap as activating ester, and H-Glu-Trp-
OH starting from enzymatically synthesized Z-l-Glu-
OBn. The first dipeptide is a substrate for metallo-
endoproteases and blocks the exocytotic release of
histamine and catecholamines from mast cells and
adrenal chromaffin cells.[21] Furthermore, it interferes
with insulin processing and inhibits glucose transport
in adipocytes.[22] The second is also known as thymo-
Figure 2. Product distributions for papain-catalyzed dipeptide synthesis for the three
&
&
ester types OGp (green), ODmap (blue), and OBn (red). : spontaneous hydrolysis, : en-
&
zymatic synthesis, : enzymatic hydrolysis, : side product. Results are clustered per
amino acid. Conditions: Z-XAA-Act (2 mm), H-Phe-NH2 (15 mm), papain (3.5 mm), HEPES
buffer (pH 8.0, 0.2m), NaCl (0.2m), DMF (10%, v/v). For reactions with Z-XAA-OBn double
quantities of papain were used (7.0 mm). The dipeptide products Z-IF-NH2 and Z-NF-NH2
precipitated during the reaction, so the yields were estimated, Z-RF-NH2 underwent sec-
ondary hydrolysis after 1 min to give Z-R-OH+H-F-NH2, and Z-FF-NH2 was enzymatically
converted into Z-FF-OH in the OBn case.[20]
The distribution of background hydrolysis, enzymatic synthe-
Table 2. Various Z-XAA-OBn esters experimentally tested with
a high
sis, enzymatic hydrolysis, and in some cases side product for-
mation for each of the studied esters is shown in Figure 2 (the
corresponding tables are included in the Supporting Informa-
tion). The background hydrolysis was determined from a blank
reaction with no papain present. The time indicated above
each bar is either that required to reach 100% conversion, or
that after which regular monitoring was stopped (3 h). The in-
dicated product distribution remained constant over time, as
measured after 24 h, unless stated otherwise.
papain concentration.[a]
Amino Experimental Background
Enzymatic
acid
t [h] Conv [%] Z-XAA-OH [%] Z-XAA-Phe-NH2 [%] Z-XAA-OH [%]
1
2
3
l-Ile
6
100
–
0.8
–
30.9 (85.5)[b]
51.7
14.5
47.5
0.6
d-Ala 10 100
b-Ala 24 26
25.4
[a] Conditions: Z-XAA-OBn (2 mm), H-Phe-NH2 (15 mm), papain (70 mm),
HEPES buffer (pH 8.0, 0.2m), NaCl (0.2m), CaCl2 (20 mm), DMF (10%, v/v).
[b] Dipeptide product precipitates during reaction; estimated yield is
given in parentheses.
For ease of comparison, the results for Z-XAA-OGp[12] are also
included in the diagram, and the data are clustered per amino
acid. Some trends can be distinguished: generally, the percen-
&
tages of enzymatic synthesis ( ) over hydrolysis ( ) are compa-
rable for each individual amino acid independent of the ester.
This was to be expected because the binding pocket for the
amino acid did not change. A remarkable exception in this re-
spect is the complete inactivity of Z-l-Thr-OBn towards papain,
which might be due to steric effects.
gen, oglufanide, or IM862, a naturally occurring immunomodu-
lator, which is a potent anti-angiogenic agent and normalizes
the immune system function of immunocompromised individ-
uals.[1a,23]
Firstly, the small-scale experiment with Z-Gly-ODmap and H-
Phe-NH2 (Table 3, entry 1) was repeated with 100 mg acyl
donor instead of 0.25 mg, resulting in a longer reaction time
and a slightly increased amount of hydrolysis as determined
by HPLC analysis of a sample from the reaction mixture
(entry 2). The desired dipeptide was isolated in 74% yield after
column chromatography. Next, the excess of H-Phe-NH2 was
decreased to arrive at equimolar amounts of acyl donor and
acceptor (entry 3). At this point, additional DMF was required
to keep all compounds in solution, and the papain concentra-
tion was increased tenfold to compensate for the anticipated
&
The spontaneous background hydrolysis ( ) decreased on
going from the OGp ester to the ODmap and OBn esters,
which correlates well with the decreasing leaving group abili-
ties of these esters. The nonenzymatic cyclic side product for-
&
mation ( ; a piperidone from Arg and a succinimide from Asn)
dropped drastically in this order for the same reason. All of the
OBn esters reacted considerably more slowly than the OGp
and ODmap esters, despite the double quantities of papain.
This difference in reactivity became even more apparent for
the more challenging amino acids Z-l-Ile-OH, Z-d-Ala-OH, and
ChemBioChem 2012, 13, 1319 – 1326
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
1323

F. P. J. T. Rutjes et al.
loss of activity. These conditions resulted in a faster reaction, a
somewhat better S/H ratio, and an isolated yield of 79%. Final-
ly, this reaction was performed on a gram scale (entry 4). De-
spite the use of a ten times higher papain concentration, the
reaction was not complete after 75 min, but the product was
purified nevertheless. The isolated yield of 74% was equal to
the yield calculated from the level of conversion indicated by
HPLC.
Conclusion
We have shown that papain-catalyzed dipeptide synthesis can
also be performed successfully with activating esters that do
not resemble arginine. Out of a set of potentially activating
esters, both the ODmap and OBn esters appeared to be suita-
ble replacements for OGp, with the added benefits of 1) being
simpler in structure and therefore cheaper and easy to synthe-
size, and 2) giving smaller amounts of undesired hydrolysis
products. The scopes of the ODmap and OBn acyl donors were
compared with that of the OGp esters, and it was found that
the synthesis to hydrolysis ratios are variable for the different
amino acids, but rather were essentially independent of the
ester involved. The reactions with OBn esters required in-
creased quantities of papain to afford similar reaction rates.
The applicability of the ODmap and OBn esters was validated
in preparative-scale syntheses of two biologically active dipep-
tides.
The synthesis of the second dipeptide was envisioned as de-
picted in Scheme 3. The chemoenzymatic method developed
by Nuijens was employed to benzylate Z-Glu-OH selectively at
the a-position and the activating ester was obtained in 96%
yield.[13a] This synthesis clearly emphasizes the advantage of
OBn as an activating ester over phenolic esters, which cannot
be enzymatically prepared. H-Trp-OBn was selected as the nu-
cleophile to enable the simultaneous removal of both protect-
ing groups, but with the risk of repeated coupling due to the
continued presence of the OBn group, which is recognized by
papain.
The computational model of papain, which was originally
built for OGp, was effectively applied in a docking study to
provide insight into the differences in reactivity of all the po-
tentially activating esters. This exercise showed that besides
a proper fit in the active site, the leaving group ability is also
important. To shed light on this intricate relationship, further
studies, both computational and experimental, are currently
underway. The balance between a good leaving-group ability
to enable the enzymatic reaction, a stable ester that does not
undergo rapid spontaneous hydrolysis, and the extent to
which the ester is activated by the enzyme seems to be deli-
cate.
Test experiments on preparative scales indicated that multi-
ple addition of H-Trp-OBn did indeed occur, but that it could
be minimized by stopping the reaction in time. These experi-
ments also showed that the reaction proceeds much more rap-
idly but with a worse S/H ratio than in the case of the previous
dipeptide. This led us to conduct a 500 mg scale experiment
with a 1:1 donor/acceptor ratio and a papain concentration of
8.8 mm. Under these conditions, HPLC analysis showed full con-
version within 45 min and an S/H ratio of 66:27; additionally, a
7% yield of tripeptide was formed. The desired dipeptide was
isolated in pure form in 60% yield after column chromatogra-
phy. Subjection of the protected dipeptide to hydrogenolysis
conditions provided biologically active H-Glu-Trp-OH in 87%
yield.
Experimental Section
Synthesis: See the Supporting In-
formation for a detailed descrip-
Table 3. Reaction conditions and yields of preparative-scale syntheses of Z-Gly-Phe-NH2.
tion of the synthetic procedures
for the activating esters and prod-
uct characterization.
Enzymatic synthesis of dipepti-
des on preparative scale: The
acyl donor, the acyl acceptor, and
pTSA (as an internal standard)
were placed in a round-bottomed
Scale
[mg]
D/A[a]
[mm]
DMF
[%, v/v]
Papain
[mm]
Time
[min]
Conv.[b]
[%]
S/H ratio[c]
[%]
Yield[d]
[%]
1
2
3
4
0.25
100
100
2:15
2:15
15:15
15:15
10
10
20
20
3.5
3.5
35
10
120
60
100
97
99
98:2
85:12
90:9
n.a.
74
79
flask and dissolved in
a small
amount of DMF, to which was
added HEPES buffer (pH 8, with
a final concentration of 0.2m) con-
taining NaCl (0.2m) and CaCl2
(20 mm). Finally, papain, activated
with dithiothreitol (DTT) in phos-
phate buffer (0.1m, pH 6.5) con-
taining EDTA (2.5 mm), was added
to start the reaction. The mixture
was stirred at 258C until total con-
version of the starting material
was observed, which was deter-
mined by HPLC. The reaction mix-
ture was then extracted with
EtOAc (3ꢁ). The combined organic
1000
350
75
89
74:15
74
[a] Donor to acceptor ratio Z-Gly-ODmap/H-Phe-NH2. [b] By HPLC. [c] Synthesis over hydrolysis ratio Z-Gly-Phe-
NH2/Z-Gly-OH. [d] Isolated yield of dipeptide after purification by flash column chromatography.
Scheme 3. Synthesis of bioactive dipeptide H-Glu-Trp-OH. Step 1: Z-Glu-OH, BnOH, Alcalase-CLEA, THF, molecular
sieves (3 ꢂ), 150 rpm, 378C. Step 2: Z-Glu-OBn (5 mm), H-Trp-OBn (5 mm), papain (8.8 mm), HEPES buffer (pH 8.0,
0.2m), NaCl (0.2m), CaCl2 (20 mm), DMF (15%, v/v), 258C. Step 3: H-Cube, Pd/C (10%), H2 (10 bar), flow rate
1 mLminꢁ1, RT.
1324
www.chembiochem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2012, 13, 1319 – 1326

Papain-Specific Activating Esters
150 rpm for 16 h in the presence
of molecular sieves (3 ꢂ). After fil-
tration, the enzyme was washed
by resuspension in THF (3ꢁ20 mL)
followed by filtration. The com-
bined organic layers were concen-
trated in vacuo and the resulting
oil was purified by column chro-
matography with aminomethyl
resin (TFA in CH2Cl2, 0!2.5%). The
solution was concentrated, TFA
Table 4. Enzymatic synthesis of dipeptides on preparative scales.
Acyl donor
Acyl acceptor
[mg]
pTSA
[mg]
DMF
[mL]
Buffer
[mL]
Milli-Q
[mL]
Papain
[mg]
[mg]
Z-Gly-ODmap
99
99
985
557
H-Phe-NH2
H-Phe-NH2
H-Phe-NH2
H-Trp-OBn
443
59
59
57
7.6
76
15
4
40
45
75
10
100
150
60
6
60
12
16
160
60
Z-Gly-ODmap
Z-Gly-ODmap
Z-Glu-OBn
496
114
105
layers were washed with brine, dried over sodium sulfate, and con-
centrated to dryness. Table 4 summarizes the results.
was removed by co-evaporation with toluene (2ꢁ20 mL) and
CHCl3 (2ꢁ20 mL), and the product was lyophilized from MeCN/H2O
(3:1) to give Z-Glu-OBn(1.24 g, 96%) as a white solid. Spectral data
were consistent with those reported in the Supporting Information
for the chemically synthesized compound.
Na-Cbz-Glycine-phenylalanine
amide (Z-Gly-Phe-NH2): The product
was obtained as a white solid after
purification by column chromatogra-
Glutamate-tryptophan (H-Glu-Trp-
phy (MeOH in CH2Cl2, 2!5%). M.p.
OH):
Z-Glu-Trp-OBn
(121 mg,
143.08C; Rf =0.45 (10% MeOH in
0.22 mmol, 1 equiv) was dissolved in
H2O in MeOH (20%, 12 mL) and hy-
drogenated in the presence of Pd/C
(10%) in the H-Cube (ThalesNano) at
room temperature, 1 mLminꢁ1, and
10 bar H2 pressure. The product was
obtained as a slightly pink solid after
evaporation of the solvent in vacuo
(63 mg, 87%). M.p.=159.28C; Rf =
0.27 (CHCl3/MeOH/NH4OH 65:45:20);
CH2Cl2); [a]2D0 =+7.1 (c=2.60, MeOH);
1H NMR (CD3CN, 400 MHz): d=7.42–
7.18 (m, 10H), 6.88 (d, J=7.7 Hz;
NH), 6.44 (brs; NH), 5.98 (brt; NH), 5.83 (brs; NH), 5.07 (s, 2H), 4.52
(ddd, J=5.2, 8.3, 8.3 Hz, 1H), 3.67 (dd, J=6.1. 17.0 Hz, 1H), 3.62
(dd, J=6.0. 16.9 Hz, 1H), 3.13 (dd, J=5.1, 14.0 Hz, 1H), 2.89 ppm
(dd, J=8.5, 14.0 Hz, 1H); 13C NMR (CD3CN, 75 MHz): d=173.8,
170.1, 157.7, 138.4, 137.9, 130.2, 129.4, 129.2, 128.9, 128.7, 127.5,
67.3, 54.9, 44.9, 38.2 ppm; IR (film): n˜ =3460, 3299, 3062, 3032,
1712, 1681, 1644, 1563, 1529, 1284, 1250, 1219, 1172, 1045, 992,
728, 697 cmꢁ1; HRMS (ESI) m/z calcd for C19H21N3NaO4: 378.1430
[M+Na]+, found: 378.1443.
[a]2D0 =+22.4
(c=0.31,
DMSO);
1H NMR (CD3OD/D2O, 400 MHz): d=7.64 (d, J=7.8 Hz, 1H), 7.37 (d,
J=8.1 Hz, 1H), 7.18 (s, 1H), 7.11 (ddd, J=1.1, 7.1, 8.1 Hz, 1H), 7.04
(ddd, J=1.1, 7.1, 8.0 Hz, 1H), 4.61 (dd, J=4.8, 8.7 Hz, 1H), 3.82 (dd,
J=5.2, 7.4 Hz, 1H), 3.39 (dd, J=4.8, 14.7 Hz, 1H), 3.18 (dd, J=8.8,
14.7 Hz, 1H), 2.52–2.36 (m, 2H), 2.16–1.95 ppm (m, 2H); 13C NMR
(CD3OD/D2O, 75 MHz): d=179.5, 177.5, 169.8, 137.7, 128.7, 124.7,
122.4, 119.8, 119.4, 112.4, 111.5, 56.9, 54.3, 33.3, 28.6, 28.4 ppm; IR
(film): n˜ =3213, 3067, 1693, 1665, 1597, 1548, 1483, 1455, 1414,
1339, 740 cmꢁ1; HRMS (ESI) m/z calcd for C16H20N3O5: 334.1403
[M+H]+; found: 334.1391.
Na-Cbz-glutamate-tryptophan
benzyl ester (Z-Glu-Trp-OBn): The
product was obtained as a white
solid after purification by column
chromatography (AcOH/MeOH/EtOAc
1:5:94) and lyophilization from diox-
ane. M.p.=48.78C; Rf =0.50 (AcOH/
MeOH/EtOAc 1:25:474); [a]2D0 =ꢁ13.8
(c=1.00, MeOH); 1H NMR (CD3CN,
Computational studies: a detailed description of the computation-
400 MHz): d=9.17 (s; NH), 7.51 (d,
J=7.9 Hz; NH), 7.41–7.19 (m, 11H),
al analyses can be found in a previous article on this subject.[12]
7.15–7.10 (m, 1H), 7.08–7.01 (m, 3H),
5.99 (d, J=7.4 Hz; NH), 5.09–4.99 (m, 4H), 4.75–4.69 (m, 1H), 4.13–
4.05 (m, 1H), 3.29–3.15 (m, 2H), 2.30 (t, J=7.5 Hz, 2H), 1.82–
1.71 ppm (m, 2H); 13C NMR (CD3CN, 75 MHz): d=174.7, 172.4,
172.3, 157.1, 137.3, 136.8, 129.4, 129.1, 129.0, 128.8, 128.6, 128.3,
124.7, 122.5, 120.0, 119.2, 118.2, 112.2, 110.3, 67.6, 67.2, 55.1, 54.2,
30.4, 28.2, 28.1 ppm; IR (film): n˜ =3312, 3058, 3031, 2963, 2918,
1714, 1664, 1528, 1455, 1342, 1213, 743, 698 cmꢁ1; HRMS (ESI) m/z
calcd for C31H31N3O7: 558.2240 [M+H]+, found: 558.2240.
Acknowledgements
This research was performed under the ACTS-NWO Integration of
Biosynthesis & Organic Synthesis (IBOS) program. DSM Innovative
Synthesis B.V. (Geleen, the Netherlands) is gratefully acknowl-
edged for financial support. S.B.N. is supported by the Nether-
lands Organization for Scientific Research (NWO) through a VENI
grant (700.58.410). Furthermore, Timo Nuijens (DSM) and Martin
Broxterman are gratefully acknowledged for their help with the
enzymatic synthesis of Z-Glu-OBn.
Chemoenzymatic synthesis of Z-Glu-OBn: Before use, Alcalase-
CLEA (3 g, Type OM, CLEA-Technologies, 580 Ugꢁ1) was suspended
in tBuOH (100 mL) and crushed with a spatula. After filtration, the
enzyme was resuspended in methyl tert-butyl ether (MTBE; 50 mL),
followed by filtration. Large enzyme particles were removed with
a sieve (0.5 mm pore size). Molecular sieves (Acros, 3 ꢂ, 8 to 12
mesh) were activated (2008C under vacuum overnight), crushed,
and sieved (0.5 mm pore size) to remove large particles.
Keywords: enzymatic peptide synthesis
activation · molecular modeling · papain · peptides
· enzyme-specific
Alcalase-CLEA (509 mg) was added to a solution of Z-l-Glu-OH
(977 mg, 3.48 mmol, 1 equiv) in THF (9 mL) and benzyl alcohol
(1 mL, 9.66 mmol, 2.8 equiv). The mixture was shaken at 378C at
[1] a) P. Vlieghe, V. Lisowski, J. Martinez, M. Khrestchatisky, Drug Discovery
Today 2010, 15, 40–56; b) D. Hans, P. R. Young, D. P. Fairlie, Med. Chem.
2006, 2, 627–646; c) R. E. W. Hancock, H.-G. Sahl, Nat. Biotechnol. 2006,
ChemBioChem 2012, 13, 1319 – 1326
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
1325

F. P. J. T. Rutjes et al.
24, 1551–1557; d) A. A. Zompra, A. S. Galanis, O. Werbitzky, F. Albericio,
Future Med. Chem. 2009, 1, 361–377.
[15] S. M. A. Salam, K. Kagawa, T. Matsubara, K. Kawashiro, Enzyme Microb.
Technol. 2008, 43, 537–543.
[2] a) V. A. Tutel’yan, V. K. Khavinson, V. V. Malinin, Bull. Exp. Biol. Med. 2003,
135; b) R. Hartmann, H. Meisel, Curr. Opin. Biotechnol. 2007, 18.
[3] a) M. P. Lupo, A. L. Cole, Dermatol. Ther. 2007, 20, 343–349; b) B. S.
Oricha, Afr. J. Pharm. Pharmacol. 2010, 4, 127–129.
[4] a) D. Agyei, M. K. Danquah, Biotechnol. Adv. 2011, 29; b) V. Marx, Chem.
Eng. News 2005, 83, 17–24; c) L. Andersson, L. Blomberg, M. Flegel, L.
Lepsa, L. Nilsson, M. Verlander, Biopolymers 2000, 55, 227–250.
[5] F. Albericio, Curr. Opin. Chem. Biol. 2004, 8, 211–221.
[6] J. Y. Lee, D. Bang, Biopolymers 2010, 94, 441–447.
[7] S. B. H. Kent, Chem. Soc. Rev. 2009, 38, 338–351.
[8] I. Gill, R. Lꢃpez-FandiÇo, X. Jorba, E. N. Vulfson, Enzyme Microb. Technol.
1996, 18, 162–183.
[9] F. Guzmꢄn, S. Barberis, A. Illanes, J. Biotechnol. 2007, 10, 279–314.
[10] F. Bordusa, Chem. Rev. 2002, 102, 4817–4867.
[16] S.-T. Chen, K. T. Wang, J. Chem. Soc. Chem. Commun. 1988, 327–328.
[17] S. B. Nabuurs, M. Wagener, J. de Vlieg, J. Med. Chem. 2007, 50, 6507–
6518.
[18] E. Schroder, C. Phillips, E. Garman, K. Harlos, C. Crawford, FEBS Lett.
1993, 315, 38–42.
[19] M. Capellas, M. D. Benaiges, G. Caminal, G. Gonzalez, J. Lopez-Santin, P.
Clapꢅs, Biotechnol. Bioeng. 1996, 50, 700–708.
[20] This is in line with literature findings that the second amino acid (occu-
pying the P2 position) is important for the recognition by papain: a) I.
Schechter, A. Berger, Biochem. Biophys. Res. Commun. 1967, 27, 157–
162; Other amino acids that are recognized by papain in that position
include Val and Tyr, but these were not present in our selection: b) Y.
Choe, F. Leonetti, D. C. Greenbaum, F. Lecaille, M. Bogyo, D. Brçmme,
J. A. Ellman, C. S. Craik, J. Biol. Chem. 2006, 281, 12824–12832.
[21] D. I. Mundy, W. J. Strittmatter, Cell 1985, 40, 645–656.
[22] a) L. P. Aiello, M. Wessling-Resnick, P. F. Pilch, Biochemistry 1986, 25,
3944–3950; b) A. Jochen, P. Berhanu, Biochem. Biophys. Res. Commun.
1987, 142, 205–212.
[11] F. Bordusa, Curr. Protein Pept. Sci. 2002, 3, 159–180.
[12] R. J. A. C. de Beer, B. Zarzycka, H. I. V. Amatdjais-Groenen, S. C. B. Jans, T.
Nuijens, P. J. L. M. Quaedflieg, F. L. van Delft, S. B. Nabuurs, F. P. J. T.
Rutjes, ChemBioChem 2011, 12, 2201–2207.
[13] a) T. Nuijens, C. Cusan, J. A. W. Kruijtzer, D. T. S. Rijkers, R. M. J. Liskamp,
P. J. L. M. Quaedflieg, Synthesis 2009, 5, 809–814; b) T. Nuijens, C.
Cusan, J. A. W. Kruijtzer, D. T. S. Rijkers, R. M. J. Liskamp, P. J. L. M. Quaed-
flieg, J. Org. Chem. 2009, 74, 5145–5150.
[23] J. A. Bergeon, Y.-N. Chan, B. G. Charles, I. Toth, Biopolymers 2008, 90,
633–643.
[14] a) M. Fitꢅ, G. Alvaro, P. Clapꢅs, J. Lꢃpez-Santin, M. D. Benaiges, G. Cami-
nal, Enzyme Microb. Technol. 1998, 23, 199–203; b) P. Stehle, H.-P. Bah-
sitta, B. Monter, P. Fꢆrst, Enzyme Microb. Technol. 1990, 12, 56–60; c) A.
Lang, C. Hatscher, P. Kuhl, Tetrahedron Lett. 2007, 48, 3371–3374.
Received: January 8, 2012
Published online on May 21, 2012
1326
www.chembiochem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2012, 13, 1319 – 1326