Enzyme-Specific Activation versus Leaving Group Ability

Article abstract of DOI:10.1002/cbic.201200227

Enzyme-specific activation and the substrate mimetics strategy are effective ways to circumvent the limited substrate recognition often encountered in protease-catalyzed peptide synthesis. A key structural element in both approaches is the guanidinophenyl (OGp) ester, which enables important interactions for affinity and recognition by the enzyme-at least, this is usually the explanation given for its successful application. In this study we show that leaving group ability is of equal or even greater importance. To this end we used both experimental and computational methods: 1) synthesis of close analogues of OGp, and their evaluation in a dipeptide synthesis assay with trypsin, 2) molecular docking studies to provide insights into the binding mode, and 3) ab initio calculations to evaluate their electronic properties.

Full text of DOI:10.1002/cbic.201200227

DOI: 10.1002/cbic.201200227
Enzyme-Specific Activation versus Leaving Group Ability
Roseri J. A. C. de Beer,[a] Berry Bçgels,[a] Gijs Schaftenaar,[b] Barbara Zarzycka,[b]
Peter J. L. M. Quaedflieg,[c] Floris L. van Delft,[a] Sander B. Nabuurs,[b] and
Floris P. J. T. Rutjes*[a]
Enzyme-specific activation and the substrate mimetics strategy
are effective ways to circumvent the limited substrate recogni-
tion often encountered in protease-catalyzed peptide synthe-
sis. A key structural element in both approaches is the guanidi-
nophenyl (OGp) ester, which enables important interactions for
affinity and recognition by the enzyme—at least, this is usually
the explanation given for its successful application. In this
study we show that leaving group ability is of equal or even
greater importance. To this end we used both experimental
and computational methods: 1) synthesis of close analogues of
OGp, and their evaluation in a dipeptide synthesis assay with
trypsin, 2) molecular docking studies to provide insights into
the binding mode, and 3) ab initio calculations to evaluate
their electronic properties.
Introduction
These days enzymes are commonly used in organic synthesis.[1]
The benefits of enzymatic reactions, including their generally
excellent regio- and enantioselectivity, are widely recognized.
Because of the usually mild reaction conditions, enzymatic
conversions are often regarded as a green alternative to classic
organic reactions. However, the limited substrate scope of
many enzymes remains a big disadvantage.
group ability may be an important factor.[5] In this study, the
contribution of these components was investigated by both
experimental and computational methods. Several analogues
of OGp were designed, synthesized, and docked to trypsin to
provide insight in the binding mode. Subsequently, their effec-
tiveness in dipeptide formation was experimentally deter-
mined, and an attempt was made to increase the activity by
further variation of one of the analogues. An ab initio study
provided insight into the electronic properties of the ana-
logues under investigation.
One of the areas of application of enzymes is peptide syn-
thesis. In this field, proteases are employed to form the pep-
tide bonds (which they would natively hydrolyze) by exploiting
the reversibility of chemical reactions. A prerequisite for this
enzymatic activity, irrespective of whether aqueous media or
organic solvents are used, is that specific amino acids are rec-
ognized.[2] This problem of recognition can be circumvented
by applying the “substrate mimetics” strategy as previously de-
scribed for trypsin and other proteases.[3] The guanidinophenyl
(OGp) ester, which in essence resembles the naturally recog-
nized side chain of arginine, is claimed to serve as a recognition
moiety for trypsin, thereby making recognition independent of
the side chain of the amino acid and thus broadening sub-
strate scope. This approach is typically applied under aqueous
conditions; but because OGp also functions as a leaving
group, the commonly occurring secondary hydrolysis is pre-
vented, as the product formed becomes unrecognizable for
the enzyme.
Results and Discussion
Prediction of the binding mode of OGp analogues to trypsin
To distinguish between the effects of affinity for the enzyme
and leaving group ability, we evaluated a set of close ana-
logues of OGp that differed slightly in both properties. We
opted for trypsin as the model system, because this enzyme is
highly specific for arginine, in contrast to papain, which exhib-
its broad substrate specificity, with only a slight preference for
arginine. Moreover, the catalytic mechanism of the serine pro-
[a] R. J. A. C. de Beer, B. Bçgels, 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
A similar solution to limited substrate acceptance was found
for papain, that is, to overcome enzyme-specific activation.[4]
Based on docking studies, the OGp group is predicted to bind
to papain in a different orientation than the natural substrate,
arginine. By taking advantage of this alternative recognition,
papain is able to catalyze dipeptide formation without being
restricted to specific amino acid residues. We noticed these
versatile applications of the OGp moiety and wondered what
was the reason for these remarkable properties. Additional re-
search, in which OGp was replaced with simpler esters, indicat-
ed that besides enzyme recognition and affinity, the leaving
[b] Dr. G. Schaftenaar, 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)
[c] Dr. P. J. L. M. Quaedflieg
DSM Innovative Synthesis BV
P.O. Box 18, 6160 MD Geleen (The Netherlands)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cbic.201200227.
ChemBioChem 0000, 00, 1 – 7
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

F. P. J. T. Rutjes et al.
tease trypsin,[6] also with respect to substrate mimetics,[7] is
well known.
The analogues were designed in such a way that they close-
ly resemble OGp in structure, although the leaving group char-
acter was varied considerably (Scheme 1).
Figure 1. Molecular modeling of OGp analogues in trypsin. Hydrogen bond-
ing interactions to functionally important amino acids in the trypsin active
site are shown for A) arginine, B) OGp, C) OGb, D) O3G, and E) OAb. The
space occupied by the arginine side chain is indicated in blue.
be located in the oxyanion hole while simultaneously retaining
interaction between the amidinium group and Asp189. The ali-
phatic Z-Gly-O3G analogue (Figure 1D) perfectly mimicked the
side chain of arginine, thus achieving hydrogen-bond inter-
actions with Asp189, Ser190, and Tyr217. The oxyanion hole
residues Gly193 and Ser195 were in the correct position to sta-
bilize the carbonyl of the ester. Z-Gly-NGp showed a binding
mode similar to that of Z-Gly-OGp.
Scheme 1. Structures of OGp analogues. OGp: p-guanidinophenyl ester;
O3G: 3-guanidinopropyl ester; NGp: p-guanidinobenzyl amide; OAb: p-ami-
dinobenzyl ester; OGb: p-guanidinobenzyl ester.
Two benzylic variants (OAb and OGb) were synthesized as
the corresponding Z-Gly-OH esters,[5] as well as an aliphatic an-
alogue (O3G) and an OGp equivalent where the ester bond
was replaced by an amide bond (NGp). The substrates were re-
stricted to glycine esters in order to rule out any influence of
the side chain in the coupling reactions. The anticipated analo-
gy of these four compounds was evaluated by a computational
docking study with the flexible docking program Fleksy.[8] The
results were visualized and analyzed by using the YASARA pro-
gram.[9]
Experimentally determined activity of OGp analogues
The OGp analogues were evaluated experimentally in an enzy-
matic assay with trypsin. HꢀPheꢀNH2 was used as the acyl
acceptor because of its distinctive UV properties at 254 nm,
which simplifies HPLC analysis. The enzymatic reaction was
monitored for three hours. The identity of the products was
confirmed by chemical synthesis of reference compounds and
LC-MS analysis. Table 1 presents either the time to reach full
conversion, or the extent of conversion after three hours. Back-
ground hydrolysis of the analogues was determined from
a blank reaction (no trypsin). This indicated that the ratio of
enzymatic synthesis to hydrolysis remained constant over time
(measured at 24 h unless stated otherwise).
The three-dimensional structure of trypsin has been previ-
ously solved by crystallographic studies.[10] The secondary
structure of this globular enzyme consists of b-sheets organ-
ized into two densely packed hydrophobic b-barrels. Five sub-
pockets (S3 through S2’)[11] are important for specific binding of
the substrate. A crucial interaction occurs at the carboxylate
moiety of Asp189 at the bottom of subpocket S1, which pri-
marily determines the specificity of trypsin for positively
charged side chains. The catalytic triad comprises Ser195,
His57, and Asp102. The adjacent backbone amides of Gly193
and Ser195 create the oxyanion hole.
Z-Gly-OGp is readily converted by trypsin, although the syn-
thesis to hydrolysis ratio is not favorable (Table 1). The remain-
ing analogues did not show any activity at the same enzyme
concentration, except Z-Gly-OGb, for which there was slight ac-
tivity. A 100-fold increase in trypsin concentration gave some
differentiation: Z-Gly-OGb was almost completely consumed in
two hours, whereas Z-Gly-OAb and Z-Gly-O3G were 43 and
38% converted, respectively, in three hours. The lower activity
with Z-Gly-OAb might be explained by its somewhat distorted
fit in the active site of trypsin. The low activity with Z-Gly-O3G,
however, was absolutely unexpected, as this analogue ap-
peared to be a perfect mimetic. Z-Gly-NGp was the only ana-
logue that remained completely inactive under these condi-
tions. In all cases, the leaving group characteristics of the ana-
logues were reduced compared to OGp. We conclude that
although all the analogues were expected to be fairly similar in
terms of affinity for the enzyme, this contribution to activity is
The Z-Gly-OGp ester (Figure 1B) could be easily docked in
the active site of trypsin, in a similar way to the arginine side
chain (Figure 1A), in agreement with previous findings.[7] The
guanidino group made the equivalent crucial interaction with
Asp189 in the S1 pocket and, additionally, formed hydrogen
bonds with Ser190 and Trp215. Furthermore, the substrate car-
bonyl was nicely located in the oxyanion hole. The docking
poses of the benzylic esters Z-Gly-OGb and Z-Gly-OAb were
less optimal, even though the carbonyl groups were located in
the oxyanion hole. Figure 1C shows that OGb slightly extend-
ed beyond the volume occupied by the arginine side chain as
a result of the additional carbon atom, whereas OAb (Fig-
ure 1E) had to adopt a somewhat distorted conformation to
&2
&
www.chembiochem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 7
ÝÝ
These are not the final page numbers!

Enzyme-Specific Activation versus Leaving Group Ability
Table 1. Various Z-Gly-Act compounds tested experimentally.[a]
Experimental
Background[c]
Z-Gly-OH [%]
Enzymatic
Z-Gly-Phe-NH2 [%]
Act
t [min]
Conv [%]
Z-Gly-OH [%]
1
2
3
4
5
6
7
8
OGp[b]
OAb
OGb
NGp
15
180
120
180
180
180
180
90
100
43
99
–
38
73
100
100
2.7
2.0
3.3
–
0.9
5.0
2.4
5.7
22.5
23.4
72.5
–
29.6
54.0
76.3
73.6
74.8
17.6
23.2
–
O3G
7.5
O3Gr
O3G=
OTfe
14.0
21.3
20.7
[a] Conditions: 2 mm Z-Gly-Act, 15 mm PheꢀNH2, 160 mm trypsin, 0.2m HEPES (pH 8.0), 0.2m NaCl, 20 mm CaCl2, 10% (v/v) DMF. [b] 1.6 mm trypsin, which is
100 times less than for the remaining entries. [c] Only spontaneous hydrolysis was observed as background reaction.
rather small. Rather, the results imply a large influence of the
leaving group ability of the ester.
O3G variants with increased activity
If electronic properties indeed play an important role, it should
be possible to increase the activity of the analogues by trans-
forming them into better leaving groups. O3G was selected for
this, as it appeared to be a near-perfect mimic of the natural
substrate and its activity was surprisingly low compared to
OGp.
Inspection of the molecular model of trypsin with Z-Gly-O3G
revealed that there is space in the binding pocket for the intro-
duction of a small substituent (Figure 2), such as a methylene
(O3G=) or two fluorides (O3GF2), which are weak and strong
inductively electron-withdrawing groups, respectively. How-
ever, these modifications also create additional van der Waals
interactions with the pocket, which by themselves can be a rea-
son for increased affinity of the ester for the enzyme. To assess
this effect, a cyclopropyl group was included (O3Gr), which
was shown to fit in the binding pocket to make these addi-
tional interactions without altering the electronic properties.
The O3G variants were then experimentally evaluated in a
trypsin enzymatic assay. As the synthesis of Z-Gly-O3GF2 failed
(synthesis details are provided in the Supporting Information),
the trifluoroethyl ester (OTfe) was included as an alternative, as
it also nicely fits in the active site of trypsin (Figure 2D) and it
is known to act as a good leaving group. Both O3G variants
showed considerably superior activity over O3G (Table 1). As
anticipated, addition of the steric cyclopropyl group (O3Gr)
increased the activity (from 38 to 73% conversion in 3 h),
whereas introduction of the slightly inductively electron-with-
Figure 2. Structures of O3G variants and their docking poses. A) O3G=,
B) O3Gr, C) O3GF2, and D) OTfe.
drawing methylene (O3G=) improved the activity even further
(100% conversion in 3 h). Surprisingly, of all the analogues
tested, Z-Gly-OTfe yielded the fastest reaction (100% conver-
sion in 90 min), despite the absence of a cationic recognition
element. This again supports the idea that the leaving group
ability is an important contributor to the suitability of the ester
for enzymatic peptide synthesis.
Ab initio calculations
An appropriate computational technique to study electronic
events within or between molecules is ab initio calculations. As
the computational requirements are large for these calcula-
ChemBioChem 0000, 00, 1 – 7
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
&
3
&
ÞÞ
These are not the final page numbers!

F. P. J. T. Rutjes et al.
tions, four compounds were selected: Z-Gly-OGp, Z-Gly-O3G,
Z-Gly-NGp, and Z-Gly-OTfe. Furthermore, the system was dras-
tically simplified, in that the enzyme was represented by only
the hydroxyl connected to a carbon of the active site serine. In
addition, part of the protecting group of the ester was not
taken into consideration. This situation is displayed as the
white regions of Figure 3A. The first step of the reaction com-
is principally indicative of the leaving group ability of the vari-
ous alcohols and amine, showed more variation but with a sim-
ilar trend. The enzyme cooperatively stabilizes the negative
charge that develops on the leaving group during the collapse
of the tetrahedral intermediate (simulated by a proton). How-
ever, the differences between the leaving groups are mainly
determined by the extent to which they can stabilize the de-
veloping negative charge. The
low activation energy (TS2) of
OGp reflects its ability to meso-
merically stabilize the negative
charge, whereas the high activa-
tion energy of NGp, also capable
of mesomeric stabilisation, can
be explained by the lower elec-
tronegativity of nitrogen com-
pared to oxygen. Similarly, but
to a lesser extent, the inductively
electron-withdrawing effect of
the OTfe group is helpful in sta-
bilizing the negative charge. In
Figure 3. Reaction structural pathway of Z-Gly-Act with trypsin.
the case of O3G, the negative
prises the formation of a tetrahedral intermediate, which is sta-
bilized by the oxyanion hole (Figure 3B). The subsequent col-
lapse of the intermediate liberates the alcohol (or amine in the
case of NGp) from the complex (Figure 3C). This was the end-
point of our calculations, as we were interested in differences
in leaving group ability. Obviously, the enzyme still needs to
be deacylated by a nucleophile to complete the catalytic cycle,
which would also proceed by a tetrahedral intermediate.
We set out to locate the tetrahedral intermediate for the
OGp compound. After exhaustive partial optimizations (in
which the CꢀO bond connecting OGp to the central carbon of
the tetrahedral intermediate was fixed), we came to the con-
clusion that nowhere along this internal coordinate did a sta-
tionary point exist. No optimum or saddle point could be
found. Removing the restraint of the fixed CꢀO bond always
resulted in dissociation upon optimization. A stable intermedi-
ate was found, however, when a hydrogen atom was added to
the oxygen carrying the negative charge. This is in line with
the commonly accepted hypothesis[12] that, in the natural pro-
tein environment, hydrogen bonding in the oxyanion hole sta-
bilizes the tetrahedral intermediate. As an equivalent amount
of atoms and charge was required throughout the calculations,
a hydrogen atom was added to the leaving group too. This
corresponds with the accepted mechanism, in which the leav-
ing group is protonated upon formation of the first tetrahedral
intermediate.[13] With the stable intermediates identified, the
energy diagram (Figure 4) and the data (Table 2) were pro-
duced.
Figure 4. Energy diagram for R=OGp (green), O3G (purple), NGp (red), and
OTfe (blue) derived from ab initio calculations. A, B and C correspond to the
structures depicted in Table 2.
charge is isolated on the oxygen, without possibilities for fur-
ther stabilisation, which might also account for the high
energy level at C.
In comparing these computational results with the experi-
mental outcomes, one should bear in mind that only the acyla-
tion step was computationally studied. The experiments with
trypsin provide insight in the efficiency of the complete cata-
lytic cycle, including deacylation of the enzyme. In spite of this,
the ranking OGp>OTfe>O3G>NGp from the ab initio calcu-
lations is in agreement with the experimentally determined ac-
tivities of Z-Gly-OGp and its analogues. In addition, the marked
contrast between the highly energetic tetrahedral intermediate
and TS2 of NGp, and the favorable energies for OGp, seem to
be consistent with the observation that acylation is the rate-
determining step in amide hydrolysis,[14] in contrast to deacyla-
tion as rate-limiting step in OGp ester hydrolysis.[15] However,
In the diagram, the relative energies are compared; for con-
venience, the starting point for each of the reaction paths is ar-
bitrarily set to zero. The activation energy (TS1) for the forma-
tion of tetrahedral intermediate (B) was similar for all com-
pounds under investigation, although the energy level of B dif-
fered for each analogue. The next transition state (TS2), which
&4
&
www.chembiochem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 7
ÝÝ
These are not the final page numbers!

Enzyme-Specific Activation versus Leaving Group Ability
We hypothesized that a major
contribution was the leaving
group ability of the ester; this
was supported by ab initio cal-
culations that showed that OGp
is a good leaving group whereas
NGp and O3G are not. Further-
more, we were able to design
improved O3G variants based on
this hypothesis. Increasing activi-
ty was observed in the order
O3G<O3Gr<O3G= <OTfe,
the latter being the most effec-
tive because, although no cat-
ionic recognition element is
present in this molecule, it is
a good leaving group.
Table 2. Ab initio computed energies.
Compound (XR)
OGp
Reaction coordinate[a]
E-B3LYP [Hartree]
ZPVE [Hartree]
E relative [kcalmolꢀ1
]
A
ꢀ1023.822519
ꢀ1023.768428
ꢀ1023.827721
ꢀ1023.796406
ꢀ1023.836929
ꢀ910.799298
ꢀ910.745545
ꢀ910.793383
ꢀ910.744129
ꢀ910.770669
ꢀ1003.978561
ꢀ1003.923725
ꢀ1003.970149
ꢀ1003.913334
ꢀ1003.981686
ꢀ964.639842
ꢀ964.582210
ꢀ964.638998
ꢀ964.588930
ꢀ964.639257
0.318313
0.317477
0.323082
0.318871
0.319910
0.326420
0.324387
0.330697
0.323695
0.325304
0.333122
0.330513
0.337313
0.330891
0.332390
0.194069
0.193946
0.200064
0.193792
0.195890
0
TS (A!B)
33.5
ꢀ0.2
16.8
ꢀ8.0
0
32.5
6.4
32.9
17.3
0
32.8
7.9
39.5
ꢀ2.4
0
34.2
2.4
29.9
ꢀ0.4
B
TS (B!C)
C
O3G
NGp
OTfe
A
TS (A!B)
B
TS (B!C)
C
A
TS (A!B)
B
TS (B!C)
Experimental Section
C
A
Synthesis: See the Supporting In-
formation for a detailed descrip-
tion of the synthetic procedures
and product characterization of
the O3G variants. The details for
the compounds in Scheme 1 can
be found in a previous article.[5]
TS (A!B)
B
TS (B!C)
C
[a] See Figure 4. TS, transition state.
Molecular modeling of trypsin–
OGp analogue complexes: All de-
according to Menger et al. the reaction kinetics of proteases
towards esters are highly dependent of the nature of the ester,
as was demonstrated by comparing the p-nitrophenyl ester
with the ethyl ester.[16] These authors argued that the “p-nitro-
phenyl ester syndrome” can be attributed to excellent electro-
philic assistance to the departing entity. When reasoning by
this analogy, it follows that the O3G ester must be inferior to
OGp.
scribed molecular docking studies were performed by using the
flexible docking program Fleksy.[8,17] The crystal structure of trypsin
in complex with bovine pancreatic trypsin inhibitor (BPTI),[18]
solved at 1.5 ꢁ resolution, was used for the receptor structure (PDB
ID: 3FP6). The structure was prepared for docking by removing
BPTI and all water molecules, then hydrogen atoms were added
and their positions were optimized by using the YASARA pro-
gram.[9] In the applied docking protocol, only those docking poses
in which the scissile bond of the docked substrate mimetic aligned
to the scissile bond of the natural peptide substrate were taken
forward. Otherwise, default parameters[8] were applied.
Conclusions
General procedure for the enzymatic reactions: Enzymatic acyl
transfer reactions were performed at 258C in a total volume of
375 mL of HEPES (0.2m, pH 8.0), NaCl (0.2m), CaCl2 (20 mm), and
DMF (10%), with para toluene sulfonic acid (2 mm) as an internal
standard. Stock solutions of Z-Gly-Act compounds (50 mm) in DMF
and H-Phe-NH2 (30 mm) in buffer were prepared. The final concen-
trations of acyl donor and acyl acceptor were 2 mm and 15 mm, re-
spectively. The latter was calculated as free, Na-unprotonated nu-
cleophile concentration, [HN]0, according to the Henderson–Hassel-
balch equation: [HN]0 =[N]0/(1+10pKꢀpH). MilliQ water (1 mL, Mil-
lipore) was added to trypsin (9.6 mg), the solution was stirred, and
aliquots were stored for a maximum of one month at ꢀ208C. Fol-
lowing thermal equilibration of the assay mixture, the enzymatic
reaction was started by addition of trypsin (1.6 or 160 mm). Blank
reactions were run in parallel (no trypsin). Spontaneous ester hy-
drolysis was determined from the control reaction, as well as non-
enzymatic aminolysis of the acyl donor esters (the latter could be
ruled out). At regular intervals, aliquots (20 mL) were withdrawn
and quenched with glacial acetic acid (20 mL). The reactions were
monitored for three hours by HPLC, and checked once more for
Various methods were employed to determine the properties
to which the success of the OGp ester as substrate mimetic
and enzyme-specific activating ester can be attributed. Al-
though a direct experimental approach to determine solely the
affinity of a substrate is not available, a computational docking
study of closely related OGp analogues was insightful.
Z-Gly-OGb and Z-Gly-OAb showed that a worse fit in the
active site could be directly linked to a decrease in activity in
the enzymatic assay. This was anticipated, as a good fit in the
active site can be considered as an indication for affinity. To
our complete surprise, however, it was also demonstrated that
Z-Gly-O3G, although a perfect analogue of the natural sub-
strate arginine according to docking studies, was barely active
compared to Z-Gly-OGp. Apparently, recognition of the ester
by the enzyme is alone insufficient for activity, in contrast to
what has been suggested for substrate mimetics in the litera-
ture.
ChemBioChem 0000, 00, 1 – 7
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
&
5
&
ÞÞ
These are not the final page numbers!

F. P. J. T. Rutjes et al.
[1] M. T. Reetz, Angew. Chem. 2011, 123, 144–182; Angew. Chem. Int. Ed.
2011, 50, 138–174.
[2] F. Bordusa, Chem. Rev. 2002, 102, 4817–4867.
[3] F. Bordusa, Curr. Protein Pept. Sci. 2002, 3, 159–180.
[4] 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.
[5] R. J. A. C. de Beer, B. Zarzycka, M. Mariman, H. I. V. Amatdjais-Groenen,
M. J. Mulders, P. J. L. M. Quaedflieg, F. L. van Delft, S. B. Nabuurs, F. P. J. T.
Rutjes, ChemBioChem 2012, 13, 1319–1326.
changes in reaction mixture composition after 24 h. The values
reported are the averages of at least two separate experiments.
The identities of the formed peptide products were established by
chemical synthesis of reference compounds and LC-MS.
HPLC analysis: Samples were analyzed by a LC 2010 analytical
HPLC system (Shimadzu, Kyoto, Japan) equipped with a Varian
Inertsil RP C18 column (ODS-3, 5 mm, 150ꢂ4.6 mm; Agilent) and
eluted with various mixtures of acetonitrile/water containing tri-
fluoroacetic acid (0.1%) under isocratic and gradient conditions
(flow rate 1.0 mLminꢀ1, detection 254 nm). Product yields were cal-
culated from peak areas of the substrate esters and the hydrolysis
and aminolysis products.
[6] a) L. Polgꢃr, Cell. Mol. Life Sci. 2005, 62, 2161–2172; b) H. P. Kasserra,
K. J. Laidler, Can. J. Chem. 1969, 47, 4031–4039.
[7] M. Thormann, S. Thust, H.-J. Hofmann, F. Bordusa, Biochemistry 1999,
38, 6056–6062.
[8] S. B. Nabuurs, M. Wagener, J. de Vlieg, J. Med. Chem. 2007, 50, 6507–
6518.
[9] E. Krieger, G. Koraimann, G. Vriend, Proteins: Struct. Funct. Bioinf. 2002,
47, 393–402.
[10] R. Huber, W. Bode, Acc. Chem. Res. 1978, 11, 114–122.
[11] I. Schechter, A. Berger, Biochem. Biophys. Res. Commun. 1967, 27, 157–
162.
[12] R. C. Wilmouth, K. Edman, R. Neutze, P. A. Wright, I. J. Clifton, T. R.
Scheider, C. J. Schofield, J. Hajdu, Nat. Struct. Biol. 2001, 8, 689–694.
[13] B. Liu, C. J. Schofield, R. C. Wilmouth, J. Biol. Chem. 2006, 281, 24024–
24035.
Ab initio calculations: Standard LCAO-MO-SCF calculations were
performed with the program Gamess-US[19] by employing restricted
Hartree–Fock (RHF) procedures and density functional theory
(DFT). All DFT calculations were carried out with the B3LYP ex-
change-correlation function.[20] The geometries of the isomers were
determined by using analytical gradient and numerical second-de-
rivative optimization procedures with the 6–31G** basis set. The
relative energies were corrected for the contribution of zero-point
vibrational energies (ZPVE). The ZPVEs were calculated for the 6–
31G** optimized geometries by employing RHF and the 6–31G**
basis set.
[14] D. M. Blow, Acc. Chem. Res. 1976, 9, 145–152.
[15] H. Sekizaki, K. Itoh, E. Toyota, K. Tanizawa, Chem. Pharm. Bull. 1996, 44,
1577–1579.
[16] F. M. Menger, M. Ladika, J. Am. Chem. Soc. 1987, 109, 3145–3146.
[17] M. Wagener, J. de Vlieg, S. B. Nabuurs, J. Comput. Chem. 2012, 33,
1215–1217.
Acknowledgements
[18] E. Zakharova, M. P. Horvath, D. P. Goldenberg, Proc. Natl. Acad. Sci. USA
2009, 106, 11034–11039.
[19] M. W. Schmidt, K. K. Baldridge, J. A. Boatz, S. T. Elbert, M. S. Gordon, J. H.
Jensen, S. Koseki, N. Matsunaga, K. A. Nguyen, S. Su, T. L. Windus, M.
Dupuis, J. A. Montgomery, Jr., J. Comput. Chem. 1993, 14, 1347–1363.
[20] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
This research was performed under the ACTS-NWO Integration of
Biosynthesis & Organic Synthesis (IBOS) program. DSM (Geleen,
The Netherlands) is kindly acknowledged for financial support.
S.B.N. is supported by a VENI grant (700.58.410) from the Nether-
lands Organization for Scientific Research (NWO). Dr. Martin C.
Feiters (Radboud University Nijmegen) assisted with the interpre-
tation of the ab initio results, which is highly appreciated.
Keywords: enzyme catalysis · enzyme-specific activation ·
guanidinophenyl ester · papain · peptides
Received: April 2, 2012
Published online on && &&, 0000
&6
&
www.chembiochem.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 0000, 00, 1 – 7
ÝÝ
These are not the final page numbers!

FULL PAPERS
Activation by the enzyme or just
a good leaving group? The guanidino-
phenyl (OGp) ester is a key activating
group in substrate mimetics and che-
moenzymatic peptide bond formation.
We show that the leaving group ability
of the ester is at least as important in
this process as the specific interactions
of the activating group with the
enzyme.
R. J. A. C. de Beer, B. Bçgels,
G. Schaftenaar, B. Zarzycka,
P. J. L. M. Quaedflieg, F. L. van Delft,
S. B. Nabuurs, F. P. J. T. Rutjes*
&& – &&
Enzyme-Specific Activation versus
Leaving Group Ability
ChemBioChem 0000, 00, 1 – 7
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chembiochem.org
&
7
&
ÞÞ
These are not the final page numbers!