Reverse proteolysis promoted by in situ generated peptide ester fragments

Article abstract of DOI:10.1021/ja0344213

In this contribution we describe a general synthesis concept for the in situ preparation of protease specific reactants using methyl thioesters as universal precursors. The precursor esters are readily available by standard synthesis procedures and can be

Full text of DOI:10.1021/ja0344213

Published on Web 04/26/2003
Reverse Proteolysis Promoted by in Situ Generated Peptide
Ester Fragments
Nicole Wehofsky,§ Norman Koglin,‡ Sven Thust,† and Frank Bordusa*,§
Contribution from the Max-Planck Society, Research Unit for Enzymology of Protein Folding,
Weinbergweg 22, D-06120 Halle/Saale, Germany, Department of Biochemistry, Faculty of
Biosciences, Pharmacy, and Psychology, UniVersity of Leipzig, Talstrasse 33,
D-04103 Leipzig, Germany, and Department of Organic Chemistry, Faculty of Chemistry,
UniVersity of Leipzig, Johannisallee 29, D-04103 Leipzig, Germany
Received January 30, 2003; E-mail: bordusa@enzyme-halle.mpg.de
Abstract: In this contribution we describe a general synthesis concept for the in situ preparation of protease
specific reactants using methyl thioesters as universal precursors. The precursor esters are readily available
by standard synthesis procedures and can be used directly as reactants for protease-mediated peptide
coupling reactions. Alternatively, they can serve as initial building blocks for the in situ preparation of various
types of substrate mimetics. The synthesis of the latter is achieved by a one-pot spontaneous
transthioesterification reaction of the parent thioester (Y-(Xaa)n-SMefY-(Xaa)n-SR; R: CH2CH2COOH,
CH2C6H5, C6H4NHC(:NH)NH2), which proceeds efficiently in both a sequential manner and parallel to the
subsequent enzymatic reaction. The resulting substrate mimetics act as efficient acyl donor components
and show the typical behavior of substrate mimicry enabling irreversible reactions with originally nonspecific
acyl moieties. Neither a workup of the substrate mimetic intermediate nor changes of the reaction conditions
during the whole synthesis process are required. Model peptide syntheses using trypsin, R-chymotrypsin,
and V8 protease as the biocatalysts proved the function of the approach and illustrated its synthetic value
for protease-mediated reactions and the compatibility of the approach with state-of-the-art solid-phase
peptide ester synthesis methods.
Introduction
sis. The single or combined application of these approaches has
made possible the use of proteases for the synthesis of short
Forced by successful enzyme, medium, and substrate engi-
neering methods, proteases have gained in importance as regio-
and stereospecific catalysts in organic synthesis.2 Especially for
applications that are based on their native hydrolysis activity,
such as regiospecific ester hydrolysis or the kinetic resolution
of racemates, proteases are now generally recognized as normal
bench reagents. In principle, these engineering methods also
allow for the reduction of competitive acyl donor hydrolysis,
the alteration of the enzyme specificity and the minimization
of undesired proteolytic cleavages, which are the main draw-
backs when proteases are used as catalysts for reverse proteoly-
peptides,3 the coupling of longer-chain peptide fragments,4 and
even the synthesis of fully active enzymes,5 nonpeptidic
carboxylic acid amides,6 or peptidoglycans.7 Despite these
successful examples, it is still the enzyme’s stringent specificity
for both the acyl donor and acceptor component that hinders
the universal synthetic use of the enzymatic approach. Sub-
stantial improvements to overcome the specificity problem for
the acyl donor have been attained through the development of
substrate mimetics.8 Contrary to classical acyl donors, in
substrate mimetics the site-specific amino acid moiety is
transferred from the C-terminal of the peptide residue to the
(3) For recent examples see: (a) Matsumoto, K.; Davis, B. G.; Jones, J. B.
Chem. Eur. J. 2002, 8, 4129. (b) Sekizaki, H.; Itoh, K.; Toyota, E.;
Tanizawa, K. J. Peptide Sci. 2002, 8, 521. (c) Gu¨nther, R.; Thust, S.;
Hofmann, H.-J.; Bordusa, F. Eur. J. Biochem. 2000, 267, 3496. (d) Rival,
S.; Aulnier, J.; Wallach, J. Biocatal. Biotrans. 2000, 17, 417.
(4) For recent examples see: (a) Liu, C.-F.; Tam, J. P. Org. Lett. 2001, 3,
4157. (b) Cerovsky, V.; Kockska¨mper, J.; Glietsch, H. G.; Bordusa, F.
ChemBioChem 2000, 2, 126. (c) Low, D. W.; Yang, G.; Winkler, J. R.;
Gray, H. B. J. Am. Chem. Soc. 1997, 119, 4094.
§ Max-Planck Society, Research Unit for Enzymology of Protein Folding.
‡ Department of Biochemistry, Faculty of Biosciences, Pharmacy, and
Psychology, University of Leipzig.
† Department of Organic Chemistry, Faculty of Chemistry, University
of Leipzig.
(1) Abbreviations: AM, amino methyl; Boc, tert-butoxycarbonyl; Bz, benzoyl;
DIEA, N,N-diisopropylethylamine; DMF, dimethylformamide; Fmoc,
9-fluorenylmethoxycarbonyl; Hepes, N-[2-hydroxyethyl]piperazine-N′-[2-
ethanesulfonic acid]; NEM, 4-ethylmorpholine; NMM, 4-methylmorpholine;
OGp, 4-guanidinophenyl ester; PyBOP, (1H-benzotriazol-1-yloxy)tripyr-
rolidinophosphonium hexafluorophosphate; SBn, benzyl thioester; SCe,
carboxyethyl thioester; SGp, 4-guanidinophenyl thioester; SMe, methyl
thioester; TBTU, (N,N,N′,N′-tetramethyl-O-benzotriazol-1-yl)uronium-tet-
rafluoroborate; THF, tetrahydrofuran; TIS, triisopropylsilane; TLCK, NR-
p-tosyl-L-lysine chloromethyl ketone; TPCK, NR-p-tosyl-L-phenylalanine
chloromethyl ketone; V8 protease, Glu-specific endopeptidase from Sta-
phylococcus aureus strain V8; Xaa, amino acid residue; Z, benzyloxycar-
bonyl.
(5) Jackson, D. Y.; Burnier, J.; Quan, C.; Stanley, M.; Tom, J.; Wells, J. A.
Science 1994, 266, 243.
(6) Gu¨nther, R.; Bordusa, F. Chem. Eur. J. 2000, 6, 463.
(7) Wehofsky, N.; Lo¨ser, R.; Buchynskyy, A.; Welzel, P.; Bordusa, F. Angew.
Chem., Int. Ed. 2002, 41, 2735.
(8) (a) Schellenberger, V.; Jakubke, H.-D.; Zapevalova, N. P.; Mitin, Y. V.
Biotechnol. Bioeng. 1991, 38, 104. (b) Sekizaki, H.; Itoh, K.; Toyota, E.;
Tanizawa, K. Chem. Pharm. Bull. 1996, 44, 1585. (c) Bordusa, F.; Ullmann,
D.; Elsner, C.; Jakubke, H.-D. Angew. Chem., Int. Ed. Engl. 1997, 36, 2473.
(d) Thormann, M.; Thust, S.; Hofmann, H.-J.; Bordusa, F. Biochemistry
1999, 38, 6056.
(2) For recent review see: Bordusa, F. Chem. ReV. 2002, 102, 4817.
9
6126
J. AM. CHEM. SOC. 2003, 125, 6126-6133
10.1021/ja0344213 CCC: $25.00 © 2003 American Chemical Society

In Situ Generated Peptide Esters
A R T I C L E S
Scheme 1. Basis Architecture of Common Acyl-Donor
Components versus Substrate Mimetics
groups and should provide stable thioesters. Furthermore, it had
to undergo an efficient thiol exchange when exposed to thiol-
containing compounds and should be easily removable from
the reaction mixture after transesterification. Conforming to
these requirements, we chose methyl mercaptan as an interesting
candidate whose respective methyl thioesters could be consid-
ered the smallest thioesters of all. Furthermore, methyl mer-
captan should, due to its high vapor pressure, evaporate
spontaneously from the reaction mixture after its liberation by
transthioesterification. This should facilitate the purification of
the synthesis products as well as mediate a rapid and complete
thiol exchange. Suitable substrate mimetic leaving groups were
selected from the scientific literature with respect to the three
main important specificity classes of proteases, i.e., enzymes
specific for Glu/Asp, Tyr/Phe, and Arg/Lys. For Glu/Asp-
specific enzymes, a number of potential substrate mimetic
thioesters have already been reported, from which we selected
empirically the carboxyethyl thioester (SCe) type.10 In the case
of Arg/Lys-imitating substrate mimetics, the 4-guanidinophenyl
ester (OGp) moiety is commonly used,8,9 although it does not
undergo efficient transesterification due to its nonthioester
nature. Therefore, we used its structurally related 4-guanidino-
mercaptophenol (SGp) analogue, for which we expected not
only a similar kind of behavior as a substrate mimetic, but also
a sufficient transesterification activity. Correspondingly, we have
replaced the already established phenyl ester (OPh)-type sub-
strate mimetics for Phe/Tyr-specific enzymes4b,11 with the
respective benzyl thioesters (SBn). The use of the structurally
more related thiophenyl esters was avoided due to their high
chemical reactivity that may cause undesired spontaneous
acylations. Scheme 2 summarizes the ester structures selected
and illustrates their formation from the parent methyl thioester
by transthioesterification. The proof of concept was performed
with five NR-Z-protected amino acid methyl thioesters (Z-Xaa-
SMe, Xaa: Ala, Phe, Pro, Lys, Glu) differing in the nature of
their individual amino acid moiety. The synthesis of the esters
can be easily achieved by well-known standard methods from
readily available starting materials with excellent yields. To
evaluate the transthioesterification behavior of the Z-Xaa-SMe
esters, the influence of the nature and concentration of the added
thiol on the rate and completeness of thiol exchange was
investigated. Figure 1A shows the effect of the thiol concentra-
tion on the rate of spontaneous transthioesterification, using the
example of Z-Ala-SMe with 3-mercaptopropionic acid and
resulting in the formation of Z-Ala-SCe. Complete thiol
exchange was achieved with a 10-fold excess of 3-mercapto-
propionic acid over Z-Ala-SMe within 1 h, while lower
concentrations of the added thiol led to a decrease in the rate
of transthioesterification. Nevertheless, in all cases quantitative
conversions could be reached as indicated by long-time experi-
ments. By contrast, control reactions in which the evaporation
of the released methyl mercaptan was restricted showed
somewhat reduced reaction rates. Furthermore, no complete
transesterifications could be observed in these cases which
finally prove the importance of the evaporation of released
methyl mercaptan for shifting the equilibrium toward the product
ester. A shift to complete consumption of the parent thioester
ester leaving group (Scheme 1). This replacement is ac-
companied by a shift in the enzyme activity, enabling proteases
to couple nonspecific acyl moieties in an irreversible manner.9
On the other hand, the approach requires the synthesis of
specifically designed substrates bearing ester leaving groups that
are adapted to the respective enzyme’s S1 subsite specificity.
This requirement inevitably limits the choice of useable pro-
teases to those which have similar S1 subsite specificities. Thus,
one must know in advance, which protease would be the most
suitable catalyst for an intended synthesis, since the later
replacement of an enzyme with another having a different S1
specificity but better meets the demands of synthesis, requires
a complete new chemical synthesis of the respective type of
acyl donor. This general drawback complicates the practicability
of the enzymatic approach and makes the screening for the most
suitable biocatalyst inefficient and time-consuming. In addition,
some of the substrate mimetic types developed are somewhat
bulky in their fully protected forms, which can hamper their
preparation especially on solid support.
This contribution reports on a novel general synthesis concept
for both substrate mimetics and classical-type acyl donor esters
that enhances the flexibility of the enzymatic approach as well
as simplifies the chemical synthesis of the acyl donor and its
use in enzymatic reactions. The concept is based on a single
and uniform parent substrate thioester that itself can be used as
a classical acyl donor component or, alternatively, can serve as
a starting building block for the in situ preparation of different
types of substrate mimetics. The synthesis of the latter is
achieved by a one-pot spontaneous transthioesterification reac-
tion of the parent thioester, which proceeds efficiently in both
a sequential manner and parallel to the subsequent enzymatic
reaction. Hence, neither a workup of the resulting substrate
mimetics nor changes of the reaction conditions during the
whole synthesis process are required. Model peptide couplings
with three different enzymes, the parent substrate ester, and in
situ generated substrate mimetics show no significant decrease
in the reaction rate or the yield of synthesis due to the additional
transesterification step. The universality of the approach was
further demonstrated by model peptide fragment condensations,
illustrating that the new synthesis concept is of high compat-
ibility with state-of-the-art solid-phase peptide ester synthesis
methods.
Results and Discussion
We initially selected a suitable parent thioester with respect
to the following demands. To ensure efficient chemical syn-
thesis, the thiol should be small, without additional functional
(9) For recent examples see: (a) Sekizaki, H.; Itoh, K.; Toyota, E.; Tanizawa,
K. Chem. Pharm. Bull. 1998, 46, 846. (b) Wehofsky, N.; Kirbach, S. W.;
Haensler, M.; Jakubke, H.-D.; Wissmann, J.-D.; Bordusa, F. Org. Lett. 2000,
2, 2027. (c) Gu¨nther, R.; Stein, A.; Bordusa, F. J. Org. Chem. 2000, 65,
1672. (d) Sekizaki, H.; Itoh, K.; Toyota, E.; Tanizawa, K. Amino Acids
2001, 21, 175.
(10) (a) Wehofsky, N.; Bordusa, F. FEBS Lett. 1999, 443, 220. (b) Wehofsky,
N.; Wissmann, J.-D.; Alisch, M.; Bordusa, F. Biochim. Biophys. Acta 2000,
1479, 114.
(11) Cerovsky, V.; Bordusa, F. J. Pept. Res. 2000, 55, 325.
9
J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003 6127

A R T I C L E S
Wehofsky et al.
Scheme 2. In Situ Generation of Different Types of Substrate Mimetics by Transthioesterification of Acyl Methyl Thioesters and Their Use
for Enzymatic Peptide Synthesisa
a 1, acyl methyl thioester (SMe); 2, acyl-4-guanidinophenyl thioester (SGp); 3, acyl benzyl thioester (SBn); 4, acyl carboxyethyl thioester (SCe); 5,
peptide product. Thiols for transthioesterification: i, 4-guanidinomercaptophenol; ii, benzyl mercaptan; iii, 3-mercaptopropionic acid. iv, C-terminal acceptor
peptide.
Figure 1. Effect of the concentration of 3-mercaptopropionic acid (A) and V8 protease (B) on the rate of spontaneous transthioesterification of Z-Ala-SMe
to Z-Ala-SCe. Conditions: 0.2 M Hepes buffer, pH 8.0, 2.5% DMF, 37 °C, [Z-Ala-SMe]: 4 mM. (A) Concentration of 3-mercaptopropionic acid: (-b-), 0
mM; (-1-), 4 mM; (-9-), 12 mM; (-[-), 20 mM; (-2-), 40 mM. (B) Concentration of V8 protease: (-b-), 0 µM; (-1-), 1 µM; (-9-), 3 µM; (-[-), 5 µM;
(-2-), 10 µM; concentration of 3-mercaptopropionic acid: 4 mM. X ) concentration of Z-Ala-SCe.
can also be achieved by adding the enzyme to the transthioes-
terification mixture. As shown in Figure 1B for the Glu-specific
V8 protease, the rate of thiol exchange increases with rising
enzyme concentrations, enabling a nearly complete transthioes-
terification and subsequent hydrolysis under equimolar quantities
of thiol and Z-Ala-SMe in about 5 h. By contrast, no hydrolysis
of Z-Ala-SMe by V8 protease was found in control reactions
in which 3-mercaptopropionic acid was lacking. This finding
clearly proves that in the parallel approach a transthioesterifi-
cation reaction also takes place and that the formation of Z-Ala-
SCe is crucial for mediating the subsequent enzymatic reaction.
Analysis of the kinetics of Z-Ala-SCe formation reveals however
that only a negligible low concentration of the intermediate ester
is virtually detectable during the course of reaction. This
indicates an enzymatic hydrolysis of Z-Ala-SCe immediately
after its formation, suggesting that the transthioesterification
reaction is the slowest step. By contrast, the correlation between
the rate of the whole reaction and the enzyme concentration
indicates that the rate of Z-Ala-SCe formation is directly
controlled by the rate of its own consumption by the enzyme
that finally makes it likely that the enzymatic reaction is actually
rate-limiting. In general, besides Z-Ala-SMe, similar results were
found for 3-mercaptopropionic acid-mediated transthioesterifi-
cations of all other Z-Xaa-SMe esters. Despite the structural
diversity, this also holds for reactions with benzyl mercaptan
and 4-guanidinomercaptophenol that used R-chymotrypsin and
trypsin as the biocatalyst, respectively. Differences were limited
solely to the rate of spontaneous (nonenzyme-coupled) tran-
sthioesterifications with 4-guanidinomercaptophenol, which
proceeded about 5-fold slower than those with 3-mercaptopro-
pionic acid and benzyl mercaptan. On one hand, these findings
prove that beside the carboxyethyl thioesters the two aromatic
ester types also behave as substrate mimetics; on the other hand,
they suggest that the function of the transthioesterification
approach is rather insensitive toward the acyl donor’s sequence,
the structure of the added thiol and the nature of the enzyme as
well.
Next we have investigated the utility of the transthioesteri-
fication concept for enzymatic peptide synthesis. For this
purpose, the general course of the sequential and parallel
transthioesterification approach was first studied using the
synthesis of Z-Ala-Leu-Gly-NH2 from Z-Ala-SMe/Z-Ala-SCe
and the dipeptide H-Leu-Gly-NH2 catalyzed by V8 protease as
an example reaction. The two syntheses were performed with
identical reactant, thiol and enzyme concentrations and differed
only in the time point of enzyme addition. Furthermore, two
sets of control reactions, one without thiol and the other without
enzyme were analyzed, which generally gave no products. The
timelines of the enzyme- and thiol-containing reactions are
illustrated in Figure 2, A and B, for the sequential and parallel
9
6128 J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003

In Situ Generated Peptide Esters
A R T I C L E S
Figure 2. Course of the V8 protease-catalyzed coupling of Z-Ala-SMe/Z-Ala-SCe with H-Leu-Gly-NH2 by using the sequential (A) and parallel (B)
transthioesterification approach. Conditions: 0.2 M Hepes buffer, pH 8.0, 2.5% DMF, 37 °C, [Z-Ala-SMe]: 4 mM, [3-mercaptopropionic acid]: 4 mM,
[H-Leu-Gly-NH2]: 20 mM, [V8 protease]: 1 × 10-5 M. (-9-), Z-Ala-SMe; (-b-), Z-Ala-SCe; (-2-), Z-Ala-Leu-Gly-NH2; (-[-), Z-Ala-OH. X ) relative
concentration.
approach, respectively. In the case of the sequential approach,
the parent Z-Ala-SMe ester was preincubated with 3-mercap-
topropionic acid, allowing spontaneous formation of Z-Ala-SCe.
After about 24 h the thiol exchange and, hence, the formation
of the substrate mimetic was found to be complete. The enzyme
was then added, which resulted in the formation of the desired
tripeptide product in about 70% yield after complete ester
consumption in 6 h. Although this result proves the general
function of the sequential approach to peptide synthesis, the
reaction requires 30 h in total for completion, which is mainly
the result of the rather slow spontaneous transthioesterification
step. Acceleration of the latter can be achieved, however, by
using an excess of the added thiol, as already shown above for
the hydrolysis reactions (cf. Figure 1A), or simply by applying
the parallel approach, as illustrated in Figure 2B. In fact, adding
the enzyme parallel to the thiol at the beginning of the reaction
results in a synthesis course that is virtually identical to that of
the sequential approach but lacks the 24 h transthioesterification
phase. Accordingly, the whole synthesis process is completed
after 6 h and, hence, proceeds as fast as the reaction beginning
with Z-Ala-SCe directly (cf. Figure 2A). In the same way, the
product yields of the sequential and parallel approach are
substantially identical. These findings indicate that for either
approach, the additional transthioesterification step has no
influence on the reaction yield. Moreover, in the case of the
parallel approach, no effect of the preceding transthioesterifi-
cation on the total rate of the synthesis reaction became evident,
which in the end confirms that the thiol exchange is not rate-
limiting. A further advantage of the parallel over the sequential
synthesis is the ease of practicability since it requires only the
control of the coupling reaction itself without the need for
analyzing the progress of the preceding transthioesterification
step.
known secondary specificities of the respective enzymes.12 For
reasons of comparison, analogous reactions were carried out
using both the parallel and sequential approach. Although less
efficient from a practical point of view, the latter allows for the
separate quantification of the transthioesterification and the
subsequent peptide coupling reaction. Hence, conclusions can
be made on the influence of the transthioesterification step on
the efficiency of the parallel approach to peptide synthesis
simply by comparison of the two resulting data sets. To exclude
potential interferences by spontaneous reactions, additional
control experiments without thiol and enzyme were performed,
which again gave no hints of any spontaneous product forma-
tion. The results obtained for the enzymatic reactions are
summarized in Table 1. In general, the data document that with
all enzyme/substrate mimetic systems and Z-Xaa-SMe esters,
the expected peptide products could be synthesized and that
the sequential and parallel approach gave identical yields.
Peptide product formations were also found for reactions using
the specific amino acid-containing esters, demonstrating the
general capability of methyl thioesters to act as suitable
conventional-type acyl donors in peptide synthesis. Differences
in product yields of about 35%, caused by the individual amino
acid moiety of the acyl donor, and of less than 10%, caused by
the acyl acceptors used, were also found in previous studies
for reactions with preformed substrate mimetics and, thus, are
not typical for in situ generated substrate mimetics.3c,8a,10a On
analysis of the effect of the additional transthioesterification step
on the rate of reactions performed via the parallel approach, in
neither case could a significant decrease compared to the
corresponding sequential reactions be observed. In conclusion,
these findings confirm the preliminary results already discussed
above for the synthesis of Z-Ala-Leu-Gly-NH2 and in the end
indicate the general practicability of the transthioesterification
concept for enzymatic peptide synthesis.
As a final step, we investigated the function of the approach
to the ligation of elongated peptide fragments. The preceding
chemical synthesis of the two starting fragments was achieved
by using standard solid-phase Fmoc-peptide synthesis protocols.
These utilized either standard Wang-resin13 for the preparation
To investigate the universality of the transthioesterification
approach to peptide synthesis, we expanded our studies to the
next Z-Xaa-SMe esters and the already specified enzyme/
substrate mimetic systems. In addition, a respective specific
amino acid-containing Z-Xaa-SMe ester was used for each
enzyme to estimate whether the parent SMe esters could act
directly as classical acyl donors without a preceding transthioes-
terification. As the acyl acceptors, several amino acid amides
and peptides were exemplarily selected according to the already
(12) (a) Schuster, M.; Aaviksaar, A.; Schellenberger, V.; Jakubke, H.-D. Biochim.
Biophys. Acta 1990, 1036, 245. (b) Schellenberger, V.; Siegel, R. A.; Rutter,
W. L. Biochemistry 1993, 32, 4344. (c) Schellenberger, V.; Turck, C. W.;
Hedstrom, L.; Rutter, W. L. Biochemistry 1993, 32, 4349.
(13) Wang, S. S. J. Am. Chem. Soc. 1973, 95, 1328.
9
J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003 6129

A R T I C L E S
Wehofsky et al.
Table 1. Peptide Coupling of in Situ Generated Substrate Mimetics and Specific Amino Acid-Containing Z-Xaa-SMe Esters Catalyzed by
V8 Protease, Trypsin, and R-Chymotrypsin Using the Sequential and Parallel Transthioesterification Approacha
acyl donor
acyl acceptor
enzyme
product
yield [%]b
Z-Glu-SMec
Leu-Gly-NH2
Leu-Ala-Ala-Ala-Gly
Leu-Gly-NH2
Leu-Ala-Ala-Ala-Gly
Leu-Gly-NH2
Leu-Ala-Ala-Ala-Gly
Leu-Gly-NH2
Leu-Ala-Ala-Ala-Gly
Leu-Gly-NH2
Leu-Ala-Ala-Ala-Gly
Leu-NH2
Met-Ala-Ala-Ala-Gly
Leu-NH2
Met-Ala-Ala-Ala-Gly
Leu-NH2
Met-Ala-Ala-Ala-Gly
Leu-NH2
Met-Ala-Ala-Ala-Gly
Leu-NH2
Met-Ala-Ala-Ala-Gly
Met-NH2
Arg-Ala-Ala-Ala-Gly
Met-NH2
Arg-Ala-Ala-Ala-Gly
Met-NH2
Arg-Ala-Ala-Ala-Gly
Met-NH2
Arg-Ala-Ala-Ala-Gly
Met-NH2
Arg-Ala-Ala-Ala-Gly
V8d
V8d
V8d
V8d
V8d
V8d
V8d
V8d
V8d
V8d
tpe
tpe
tpe
tpe
tpe
tpe
tpe
tpe
tpe
tpe
ctf
Z-Glu-Leu-Gly-NH2
Z-Glu-Leu-Ala-Ala-Ala-Gly
Z-Pro-Leu-Gly-NH2
Z-Pro-Leu-Ala-Ala-Ala-Gly
Z-Phe-Leu-Gly-NH2
Z-Phe-Leu-Ala-Ala-Ala-Gly
Z-Lys-Leu-Gly-NH2
Z-Lys-Leu-Ala-Ala-Ala-Gly
Z-Ala-Leu-Gly-NH2
Z-Ala-Leu-Ala-Ala-Ala-Gly
Z-Lys-Leu-NH2
Z-Lys-Met-Ala-Ala-Ala-Gly
Z-Pro-Leu-NH2
Z-Pro-Met-Ala-Ala-Ala-Gly
Z-Phe-Leu-NH2
Z-Phe-Met-Ala-Ala-Ala-Gly
Z-Glu-Leu-NH2
Z-Glu-Met-Ala-Ala-Ala-Gly
Z-Ala-Leu-NH2
Z-Ala-Met-Ala-Ala-Ala-Gly
Z-Phe-Met-NH2
Z-Phe-Arg-Ala-Ala-Ala-Gly
Z-Pro-Met-NH2
Z-Pro-Arg-Ala-Ala-Ala-Gly
Z-Lys-Met-NH2
Z-Lys-Arg-Ala-Ala-Ala-Gly
Z-Glu-Met-NH2
Z-Glu-Arg-Ala-Ala-Ala-Gly
Z-Ala-Met-NH2
Z-Ala-Arg-Ala-Ala-Ala-Gly
8510a
92
59
56
89
84
56
55
67
67
76
83
82
91
79
86
75
82
68
71
81
92
91
94
87
92
89
93
81
87
Z-Glu-SMec
Z-Pro-SMe/SCe
Z-Pro-SMe/SCe
Z-Phe-SMe/SCe
Z-Phe-SMe/SCe
Z-Lys-SMe/SCe
Z-Lys-SMe/SCe
Z-Ala-SMe/SCe
Z-Ala-SMe/SCe
Z-Lys-SMec
Z-Lys-SMec
Z-Pro-SMe/SGp
Z-Pro-SMe/SGp
Z-Phe-SMe/SGp
Z-Phe-SMe/SGp
Z-Glu-SMe/SGp
Z-Glu-SMe/SGp
Z-Ala-SMe/SGp
Z-Ala-SMe/SGp
Z-Phe-SMec
Z-Phe-SMec
ctf
Z-Pro-SMe/SBn
Z-Pro-SMe/SBn
Z-Lys-SMe/SBn
Z-Lys-SMe/SBn
Z-Glu-SMe/SBn
Z-Glu-SMe/SBn
Z-Ala-SMe/SBn
Z-Ala-SMe/SBn
ctf
ctf
ctf
ctf
ctf
ctf
ctf
ctf
a Reactions were performed in 0.2 M Hepes buffer, pH 8.0, 2.5% DMF at a concentration of 4 mM acyl donor, 20 mM acyl acceptor, 8 mM thiol (basic
c
conditions). b Identical yields for the sequential and parallel approach were obtained in all cases. Reactions were performed under kinetic control and were
analyzed at the product optimum before the secondary hydrolysis of the products started. d V8, V8 protease; specific conditions: 37 °C, enzyme concentration
between 3 and 10 x 10-6 M for substrate mimetics and 1 × 10-8 M for Z-Glu-SMe, thiol: 3-mercaptopropionic acid, reaction time of the parallel approach:
e
2-8 h, reaction time of transthioesterification: 16-24 h. tp, trypsin; specific conditions: 0.1 M NaCl, 10 mM CaCl2, 25 °C, concentrations of enzyme:
1.5 × 10-5 M for substrate mimetics and 1 × 10-8 M for Z-Lys-SMe, thiol: 4-guanidinomercaptophenol, reaction time of the parallel approach: 4-10 h,
f
reaction time of transthioesterification: 48-120 h. ct, R-chymotrypsin; specific conditions: 0.1 M NaCl, 10 mM CaCl2, 25 °C, concentrations of enzyme:
2 × 10-5 M for substrate mimetics and 1 × 10-7 M for Z-Phe-SMe, thiol: benzyl mercaptan, reaction time of the parallel approach: 2-6 h, reaction time
of transthioesterification: 16-24 h.
of the C-terminal peptide fragment or Kenner’s 4-sulfamylbu-
tyryl aminomethyl safety catch-resin14 for synthesizing the
N-terminal peptide methyl thioester. The latter allows for elegant
esterification of the peptide’s C-terminal carboxylate after safety
catch-linker activation simultaneously to the release of the
peptide from the solid support (Scheme 3). Because the
esterification reaction is based on a nucleophilic attack on an
electrophilic carbonyl-carbon atom, amines and thiols were
found to give the highest product yields.15 Our attempts to
release the peptide with methyl mercaptan used in its solid
sodium methanethiolate form, however, were hampered by the
low solubility of the sodium salt in DMF, which only led to
inefficient ester synthesis. By contrast, the respective amino acid
methyl thioesters are completely soluble and gave quantitative
yields of peptide methyl thioesters in preliminary model
syntheses. Therefore, we preferred the use of a presynthesized
amino acid methyl thioester as the nucleophile for peptide
release in the synthesis of the target peptide ester. Scheme 3
specifies the peptide fragments selected and illustrates the
general course of the reactions. To demonstrate the benefits of
the transthioesterification method, the same N- and C-terminal
peptide fragment was used for all ligation reactions, differing
only in the nature of the added thiol and enzyme, respectively.
Such a constellation can be considered as typical within
screening studies for investigating the most suitable biocatalyst
for an intended synthesis. Solid-phase synthesis, peptide release,
deprotection, and purification led to high yields of the two
peptide fragments 6 and 10, which were subsequently used for
the enzymatic ligation reactions. The latter were performed as
described above, using the parallel as well as the sequential
transthioesterification approach. For synthesis-economy reasons,
only the concentration of the acyl acceptor (10) was decreased
from 20 to 10 mM, which corresponds to a ratio between 6 and
10 of 1 to 2.5. In the case of the parallel approach, after 2 h in
the reaction with V8 protease, 3 h in that of trypsin and 1 h in
the synthesis with R-chymotrypsin, complete consumption of
6 accompanied by the formation of the ligation product 11 was
found. Figure 3 shows selected HPLC profiles that illustrate
the well-defined course of catalysis, the example being the
R-chymotrypsin-mediated synthesis. As for the efficiency of
ligation reactions, with V8 protease 64%, with trypsin 68% and
with R-chymotrypsin 75% of peptide 11 could be obtained,
reflecting the distinct suitability of the individual enzymes for
this ligation reaction. Nonquantitative coupling yields are based
exclusively on competitive hydrolysis of 6, while no further
side products could be detected. Reactions using the sequential
(14) Kenner, G. W.; McDermott, J. R.; Sheppard, R. C. Chem. Commun. 1971,
636.
(15) (a) Backes, B. J.; Ellmann, J. A. J. Am. Chem. Soc. 1994, 116, 11171. (b)
Backes, B. J.; Virgilio, A. A.; Ellman, J. A. J. Am. Chem. Soc. 1996, 118,
3055. (c) Backes, B. J.; Ellman, J. A. J. Org. Chem. 1999, 64, 2322. (d)
Ingenito, R.; Bianchi, E.; Fattori, D.; Pessi, A. J. Am. Chem. Soc. 1999,
121, 11369. (e) Shin, Y.; Winans, K. A.; Backes, B. J.; Kent, S. B. H.;
Ellmann, J. A.; Bertozzi, C. R. J. Am. Chem. Soc. 1999, 121, 11684.
9
6130 J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003

In Situ Generated Peptide Esters
A R T I C L E S
Scheme 3. Peptide Fragment Ligations via Combination of the Transthioesterification and Alkanesulfonamide Safety Catch Solid-Phase
Peptide Synthesis Approach Using V8 Protease, R-Chymotrypsin and Trypsin as the Ligation Catalystsa
a Conditions of ligation reaction for V8 protease: 0.2 M Hepes buffer, pH 8.0, 2.5% DMF, 37 °C; for trypsin and R-chymotrypsin: 0.2 M Hepes buffer,
pH 8.0, 0.1 M NaCl, 10 mM CaCl2, 2.5% DMF, 25 °C. Concentration of reactants: [acyl donor]: 4 mM, [acyl acceptor]: 10 mM, [thiol]: 8 mM, [V8
protease]: 4 × 10-6 M, [R-chymotrypsin]: 2 × 10-5 M, [trypsin]: 1.5 × 10-5 M.
Alternatively, they allow for efficient transthioesterification in
sequential as well as parallel to the subsequent enzymatic
coupling reaction. Independent of the transthioesterification
approach used, the resulting spontaneously formed substrate
mimetics act as efficient acyl donor components for all enzymes
tested and show the typical behavior of substrate mimicry
enabling the irreversible coupling of nonspecific native peptide
bonds. In no case could an effect of the preceding transthioes-
terification reaction on the final product yield be observed. On
the contrary, significant differences in the rate of reactions were
found. While in sequential syntheses the thiol exchange is the
rate-limiting step of reaction, the parallel approach enables
reaction rates, which are virtually as fast as those starting from
the respective preformed substrate mimetic directly. The
consumption of the substrate mimetic by the enzyme after its
in situ formation, which can be considered the driving force of
the accelerated transthioesterification, simultaneously shortens
the substrate mimetic’s lifetime. This should allow for the use
of stronger activated and, thus, higher reactive substrate mimetic
types in this system, which would undergo a rapid hydrolysis
under conventional conditions. From a synthetic point of view,
the use of stronger activated esters should raise the synthesis
rate and, furthermore, could allow for the efficient coupling of
acyl donors containing originally slow-reacting acyl moieties.
Despite the synthesis of higher activated esters, the need for
synthesizing only one precursor ester generally reduces the
synthetic efforts to a minimum and simplifies the screening for
novel and more efficient acyl donor ester types significantly.
Similarly, it allows for a fast screening of the most suitable
biocatalyst for an intended synthesis as well, which is time-
consuming and inefficient when conventional synthesis ap-
proaches are used. Besides the proteases selected, the spectrum
of suitable enzymes can be easily enlarged, for example, by
further wild-type enzymes such as subtilisin and thrombin,8c
Figure 3. Analysis of the R-chymotrypsin-catalyzed coupling of Bz-Pro-
Thr-Ile-Gly-Gln-Val-Ser-Ala-Leu-Gly-SMe/-SBn with H-Leu-Ala-Leu-Ala-
Ser-Ala-Ser-Ala-Thr-Gly-OH by HPLC. (a) Before addition of enzyme; (b)
after 1 h. (1) Benzyl mercaptan; (2) Bz-Pro-Thr-Ile-Gly-Gln-Val-Ser-Ala-
Leu-Gly-SMe; (3) Bz-Pro-Thr-Ile-Gly-Gln-Val-Ser-Ala-Leu-Gly-OH; (4)
Bz-Pro-Thr-Ile-Gly-Gln-Val-Ser-Ala-Leu-Gly-Leu-Ala-Leu-Ala-Ser-Ala-
Ser-Ala-Thr-Gly-OH. A254 ) absorbance at 254 nm. Conditions: 0.2 M
Hepes buffer, pH 8.0, 0.1 M NaCl, 10 mM CaCl2, 2.5% DMF, 25 °C, [acyl
donor]: 4 mM, [acyl acceptor]: 10 mM, [benzyl mercaptan]: 8 mM,
[R-chymotrypsin]: 2 × 10-5 M.
transthioesterification approach led to identical product yields
in all three cases, while the elongated reaction times due to the
preceding spontaneous transthioesterification were again the only
difference between the two approaches.
In summary, we have described an efficient synthesis concept
for the in situ preparation of substrate mimetics using amino
acid and peptide methyl thioesters as the universal precursors.
The latter are readily available by standard chemical peptide
synthesis procedures and can be used directly as classical acyl
donor components for protease-catalyzed peptide synthesis.
9
J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003 6131

A R T I C L E S
Wehofsky et al.
General Procedure for the Preparation of Peptides. The peptides
MAAAG, RAAAG, LAAAG, and LALASASATG were synthesized
with a semiautomatic batch synthesizer SP 650 (Labortech AG) using
p-alkoxybenzyl alcohol resin, according to Wang13 and standard Fmoc
chemistry. Peptides were precipitated with dry diethyl ether. The identity
and purity of all derivatives were verified by analytical HPLC (220
nm), NMR, thermospray mass spectroscopy, and elementary analysis.
In all cases, satisfactory analytical data were found ((0.4% for C, H,
N).
Protocol for the Synthesis of 4-Guanidinomercaptophenol. 4-Guani-
dinomercaptophenol was prepared from 4-[N′,N′′-bis(Boc)guanidino]-
mercaptophenol by deprotection of the guanidino functionality with
TFA and precipitation with dry diethyl ether. The latter was synthesized
from N,N′-bis(Boc)-N-(trifluoromethylsulfonyl)guanidine (1 equiv) and
4-aminomercaptophenol (2 equiv) in absolute THF at room temperature.
After complete consumption of the guanidine, the reaction mixture was
evaporated, and the resulting crude product was purified by flash column
chromatography on silica gel using petroleum ether/ethyl acetate (3/1)
as the eluent.
4-[N′,N′′-Bis(Boc)guanidino]mercaptophenol: 1H NMR (300 MHz)
δ 1.43 (s, 9H), 1.53 (s, 9H), 5.46 (s, 1H), 7.30/7.44 (m/m, 4H), 9.96
(s, 1H), 11.43 (s, 1H); elementary analysis calcd (%) for C17H25N3O4S
(367.5): C 55.57, H 6.86, N 11.43, found: C 55.13, H 6.72, N 11.79;
MS, m/z: 368 [M + H]+.
4-Guanidinomercaptophenol × 1 TFA: 1H NMR (300 MHz) δ
5.66 (s, 1H), 7.15/7.38 (m/m, 4H), 7.52 (s, 3H), 9.85 (s, 1H); elementary
analysis calcd (%) for C9H10F3N3O2S (281.3): C 38.43, H 3.58, N 14.94,
found: C 38.98, H 3.41, N 14.67; MS, m/z calcd for C7H9N3S (167.1):
168 [M + H]+.
Preparation of Bz-Pro-Thr-Ile-Gly-Gln-Val-Ser-Ala-Leu-Gly-
SMe. The peptide methyl thioester was synthesized using the alkane-
sulfonamide safety catch-linker method.14 The first amino acid, Fmoc-
Leu-OH, was loaded to commercially available 4-sulfamylbutyryl
aminomethyl resin by one standard PyBOP/DIEA coupling step
resulting in a loading yield of 83%. All remaining amino acids were
coupled by stepwise solid-phase method using PyBOP/NMM activation
protocols. At the end of synthesis the NR-amino group of the resin-
bound peptide was deprotected with piperidine/DMF and subsequently
benzoylated using benzoic acid. Alkylation of the linker’s sulfonamide
functionality was achieved using iodoacetonitrile according to the
procedure of Backes et al.15b,c leading to the respective activated N,N-
cyanomethylacylalkanesulfonamide ester. The peptide was liberated
from the resin by adding a 5-fold excess of TFA‚H-Gly-SMe, providing
the fully protected peptide methyl thioester. The amino acid ester H-Gly-
SMe was prepared from Boc-Gly-SMe by deprotection of the NR-amino
group with TFA and precipitation with dry diethyl ether. Neutralization
of the amino acid trifluoroacetate was achieved by adding of NMM (1
equiv). Final deprotection of the peptide’s side-chain functionalities
by TFA/TIS/water treatment and purification of the crude product by
preparative HPLC resulted in the NR-Bz-protected decapeptide methyl
thioester. Its identity and purity was checked by analytical HPLC (220
nm), elementary analysis and mass spectroscopy.
Bz-Pro-Thr-Ile-Gly-Gln-Val-Ser-Ala-Leu-Gly-SMe: elementary
analysis calcd (%) for C49H77N11O14S (1076.3): C 54.68, H 7.21, N
14.32, found: C 53.99, H 7.01, N 13.90; MS, m/z: 1076 [M + H]+.
Transthioesterification Reactions. Spontaneous thiol exchange
reactions were performed at 37 °C using an assay mixture containing
0.2 M Hepes buffer, pH 8.0 and 2.5% DMF which was added to mediate
complete solubility of the methyl thioesters. The final concentration
of esters was 4 mM and that of 3-mercaptopropionic acid varied
between 4 and 40 mM. After thermal equilibration of the assay mixture,
the reactions were started by addition of the thiol. Model V8 protease-
catalyzed transthioesterification reactions were carried out in analogous
reaction mixtures. After incubation of the assay mixtures for several
minutes at 37 °C, the reactions were initiated by enzyme addition,
resulting in V8 protease concentrations between 1 and 10 µM. The
by enzyme variants such as subtiligase, which is known for the
acceptance of benzyl thioesters,16 or by our recently optimized
trypsin variants,17 which react efficiently with SGp-esters.
Furthermore, it can be expected that the approach is not only
useful for enzymatic peptide synthesis but also for the kinetic
resolution of racemic compounds. For the latter use, a significant
influence of the nature of the ester leaving group on the
enantioselectivity of proteases was found,18 which could easily
be specified with the approach presented. These characteristics
qualify the method as a useful and broadly applicable synthesis
concept in biocatalysis, addressing the main conflict of high
specificity and limited universality of enzymatic reactions.
Experimental Section
Materials. TPCK-treated bovine trypsin (EC 3.4.21.4., product code
(pc): T-1426), TLCK-treated bovine R-chymotrypsin (EC 3.4.21.1.,
pc: C-3142) and V8 protease (EC 3.4.21.19., pc: 45172) were obtained
from Fluka or Sigma. Proteases were used without further purification.
Amino acid derivatives, amides, 4-aminomercaptophenol, benzoic acid,
benzyl mercaptan, coupling reagents, 3-mercaptopropionic acid, thiophe-
nol, N,N′-bis-(tert-butoxycarbonyl)-N-(trifluoromethylsulfonyl)guani-
dine, sodium methanethiolate, and 4-sulfamylbutyryl AM resin were
products of Bachem, Fluka, Merck, Aldrich or Novabiochem. If not
otherwise stated, all reagents were of the highest available commercial
purity. Solvents were purified and dried by usual methods.
Chemical Syntheses of Amino Acid Methyl Thioesters. NR-
Protected amino acid methyl thioesters were synthesized by coupling
of the respective N-terminal and side-chain protected amino acid with
sodium methanethiolate using the mixed anhydride method (isobutyl-
chloroformiate/NEM).19 Following removal of the side-chain protection,
the products were precipitated and washed with dry diethyl ether. The
yields of recovered esters ranged between 80 and 90%.
Z-Ala-SMe: 1H NMR (300 MHz) δ 1.26 (d, 3H), 2.20 (s, 3H), 4.18
(m, 1H), 5.07 (s, 2H), 7.35 (m, 5H), 8.06 (d, 1H); elementary analysis
calcd (%) for C12H15NO3S (253.3): C 56.90, H 5.97, N 5.53, found:
C 57.10, H 5.77, N 5.52; MS, m/z: 252 [M - H]+.
Z-Glu-SMe × 0.7 H2O: 1H NMR (300 MHz) δ 2.19 (s, 3H), 2.30
(m, 4H), 4.18 (m, 1H), 5.04 (s, 2H), 7.35 (m, 5H), 8.05 (d, 1H), 12.10
(s, 1H); elementary analysis calcd (%) for C14H18,4NO5,7S (324.0): C
51.90, H 5.72, N 4.32, found: C 52.09, H 5.83, N 4.11; MS, m/z calcd
for C14H17NO5S (311.1): 312 [M + H]+.
1
Z-Lys-SMe × 1 TFA × 1 H2O: H NMR (300 MHz) δ 1.53 (m,
6H), 2.20 (s, 3H), 2.74 (m, 2H), 4.08 (m, 1H), 5.08 (d, 2H), 7.34 (m,
5H), 7.68 (s, 2H), 8.04 (d, 1H); elementary analysis calcd (%) for
C17H25F3N2O6S (442.5): C 46.15, H 5.70, N 6.33, found: C 45.50, H
5.57, N 6.59; MS, m/z calcd for C15H22N2O3S (310.1): 309 [M - H]+.
1
Z-Phe-SMe: H NMR (300 MHz) δ 2.23 (s, 3H), 2.94 (m, 2H),
4.35 (m, 1H), 5.00 (s, 2H), 7.29 (m, 10H), 8.13 (d, 1H); elementary
analysis calcd (%) for C18H19NO3S (329.4): C 65.63, H 5.81, N 4.25,
+
found: C 65.70, H 5.62, N 4.12; MS, m/z: 328 [M - H]
.
1
Z-Pro-SMe: H NMR (300 MHz) δ 2.05 (m, 7H), 3.47 (m, 2H),
4.46 (m, 1H), 5.11 (s, 2H), 7.34 (m, 5H); elementary analysis calcd
(%) for C14H17NO3S (279.4): C 60.19, H 6.13, N 5.01, found: C 59.50,
H 6.01, N 5.04; MS, m/z: 278 [M - H]+.
1
Boc-Gly-SMe: H NMR (300 MHz) δ 1.34 (s, 9H), 2.19 (s, 3H),
3.90 (d, 2H), 7.99 (t, 1H); elementary analysis calcd (%) for C8H15-
NO3S (205.3): C 46.81, H 7.37, N 6.82, found: C 47.02, H 7.21, N
6.59; MS, m/z: 206 [M + H]+.
(16) Braisted, A. C.; Judice, J. K.; Wells, J. A. Methods Enzymol. 1997, 289,
298.
(17) (a) Xu, S.; Rall, K.; Bordusa, F. J. Org. Chem. 2001, 66, 1627. (b) Rall,
K.; Bordusa, F. J. Org. Chem. 2002, 67, 9103.
(18) Broos, J.; Engbersen, J. F. J.; Verboom, W.; Reinhoudt, D. N. Recl. TraV.
Chim. Pays-Bas. 1995, 114, 255.
(19) Bodanszky, M.; Bodanszky, A. The Practice of Peptide Synthesis; Akademie
Verlag: Berlin, 1985; p 170.
9
6132 J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003

In Situ Generated Peptide Esters
A R T I C L E S
rate of transthioesterification reactions was analyzed by RP-HPLC
determining the disappearance of the methyl thioesters. For this purpose,
at defined time intervals aliquots were withdrawn, diluted with a
quenching solution containing 50% aqueous methanol and 5% trifluo-
roacetic acid, and immediately analyzed. The transthioesterification
reactions with benzyl mercaptan and 4-guanidinomercaptophenol were
performed analogously by using a slightly different buffer system
adapted to the enzyme requirements (0.2 M Hepes buffer, pH 8.0, 0.1
M NaCl, 10 mM CaCl2, 2.5% DMF) at 25 °C. Similar conditions were
used for the respective enzymatic reactions in which trypsin and
R-chymotrypsin were used in a concentration range between 0.2 and
20 µM.
Enzymatic Syntheses. Enzymatic peptide synthesis reactions were
performed in 0.2 M Hepes buffer, pH 8.0 containing 0.1 M NaCl and
10 mM CaCl2 at 25 °C (trypsin and R-chymotrypsin) and 0.2 M Hepes
buffer, pH 8.0 at 37 °C (V8 protease). Stock solutions of acyl donor
esters (16 mM) were prepared in distilled water containing 10% (v/v)
DMF as cosolvent. Acyl acceptor components (stock solution 40 mM)
and thiol additives (stock solution 100 mM) were dissolved in 0.4 M
Hepes buffer, pH 8.0, 0.2 M NaCl, 20 mM CaCl2 (trypsin and
R-chymotrypsin), and 0.4 M Hepes buffer, pH 8.0 (V8 protease). To
adjust a pH value of 8.0, appropriate equivalents of NaOH were added
to the stock solutions of the acyl acceptors and thiols. If not otherwise
stated, the final concentration of acyl donor, thiol component and acyl
acceptor were 4, 8, and 20 mM, respectively. The latter was calculated
as free, NR-unprotected species according to the Henderson-Hassel-
balch equation [HN]0 ) [N]0/(1 + 10pK-pH). The pK values of the
R-amino group of the acyl acceptor components were determined by
inflection point titration on a Video titrator VIT 90 (Radiometer,
Denmark). After thermal equilibration of assay mixtures, reactions using
the parallel approach were initiated by addition of the respective enzyme
at final concentrations between 3 and 10 µM for V8 protease, 15 µM
for trypsin, and 20 µM in the case of R-chymotrypsin. For parallel
syntheses, reaction times between 1 and 10 h led to complete ester
consumption. In sequential syntheses, the methyl thioester was prein-
cubated with the respective thiol. After complete transthioesterification,
acyl acceptor and enzyme were added to the reaction mixture. Elongated
reaction times between 17 h and up to 5 days led to complete ester
consumption. Progress of all reactions was analyzed by RP-HPLC. For
this purpose, at certain time intervals aliquots were withdrawn and
diluted with a quenching solution containing 50% aqueous methanol
and 5% trifluoroacetic acid. The control reactions lacking either the
thiol or the enzyme were performed under identical conditions. The
identities of the formed thioester intermediates and the resulting peptide
products were established by thermospray mass spectroscopy. The data
reported are the average of at least three independent reactions.
HPLC Analyses. Enzymatic reactions were analyzed by analytical
reversed phase HPLC using RP C18 and C8 columns (Merck, 5 µm,
25 cm × 0.4 cm, Grom Capcell, 5 µm, 25 cm × 0.4 cm) and eluted
with various mixtures of acetonitrile/water containing 0.1% trifluoro-
acetic acid under gradient conditions at flow rates of 1.0 mL min-1
.
Detection was carried out at 254 nm. Reaction rates and product yields
were calculated from peak areas of acyl donor esters, hydrolysis and
aminolysis products, respectively. Peptide purification was performed
on a HPLC system with 220 nm UV detection, using a preparative RP
C8 column (Merck, 7 µm, 25 cm × 2.5 cm) and gradient elution at a
flow rate of 15 mL min-1
.
Acknowledgment. Dedicated to Professor H.-D. Jakubke on
the occasion of his 70th birthday. This work has been supported
by the Deutsche Forschungsgemeinschaft (SFB 610 A3 and BO
1770/1-1) and the Fonds der Chemischen Industrie. We are
grateful to Dr. G. Jahreis for his help in peptide synthesis and
to Professor G. Fischer for his continuing support.
JA0344213
9
J. AM. CHEM. SOC. VOL. 125, NO. 20, 2003 6133