Investigations on the Enzyme Specificity of Clostripain: A New Efficient Biocatalyst for the Synthesis of Peptide Isosteres

Article abstract of DOI:10.1021/jo991302q

To explore the ability of the cysteine protease clostripain as a biocatalyst for the synthesis of peptide isosteres, the S′-subsite specificity of this enzyme toward unnatural substrates was investigated. First, the function of clostripain for acylating aliphatic noncyclic and cyclic amines varying in chain length and ring size was analyzed using a standard acyl donor. Additionally, this series was expanded by use of aromatic amines, amino alcohols, derivatives of non-α-amino carboxylic acids, and symmetric and asymmetric diamines, respectively. The results obtained give a detailed picture of the unique reactivity of clostripain toward synthetic substrates, allowing insights into the basic enzyme-substrate interactions. Furthermore, the data provide a guideline for the use of clostripain as a biocatalyst for synthesis of peptide isosteres. The study was completed by the utilization of a model substrate mimetic enabling clostripain to react with noncoded and non-amino acid-derived amines as well as nonspecific acyl moieties. The results of this study indicate that this approach may extend the application range of clostripain as a biocatalyst outside of peptide synthesis.

Full text of DOI:10.1021/jo991302q

1672
J . Org. Chem. 2000, 65, 1672-1679
In vestiga tion s on th e En zym e Sp ecificity of Clostr ip a in : A New
Efficien t Bioca ta lyst for th e Syn th esis of P ep tid e Isoster es†
Robert Gu¨nther,§ Anja Stein,§ and Frank Bordusa*,‡,§
Max-Planck Society, Research Unit “Enzymology of Protein Folding”, Weinbergweg 22,
D-06120 Halle/ Saale, Germany, and Department of Biochemistry, Faculty of Biosciences, Pharmacy and
Psychology, University of Leipzig, Talstrasse 33, D-04103 Leipzig, Germany
Received August 17, 1999
To explore the ability of the cysteine protease clostripain as a biocatalyst for the synthesis of peptide
isosteres, the S′-subsite specificity of this enzyme toward unnatural substrates was investigated.
First, the function of clostripain for acylating aliphatic noncyclic and cyclic amines varying in chain
length and ring size was analyzed using a standard acyl donor. Additionally, this series was
expanded by use of aromatic amines, amino alcohols, derivatives of non-R-amino carboxylic acids,
and symmetric and asymmetric diamines, respectively. The results obtained give a detailed picture
of the unique reactivity of clostripain toward synthetic substrates, allowing insights into the basic
enzyme-substrate interactions. Furthermore, the data provide a guideline for the use of clostripain
as a biocatalyst for synthesis of peptide isosteres. The study was completed by the utilization of a
model substrate mimetic enabling clostripain to react with noncoded and non-amino acid-derived
amines as well as nonspecific acyl moieties. The results of this study indicate that this approach
may extend the application range of clostripain as a biocatalyst outside of peptide synthesis.
In tr od u ction
Peptide isosteres have become important for the syn-
thesis of pharmacologically active and proteolytically
stable peptide derivatives4 or protease inhibitors.5 Hence,
there is a strong need for efficient, selective, and envi-
ronmentally friendly methods for their synthesis. Thus,
the screening of naturally occurring proteases is funda-
mental to find suitable biocatalysts possessing appropri-
ate catalytic properties. Our results found for clostripain-
catalyzed couplings of trifluoromethyl-substituted
peptides2g indicate a high reactivity of this cysteine
protease toward synthetically modified amino acid-
containing acyl acceptors. Motivated by these findings,
the S′-subsite specificity (nomenclature according to ref
6) of clostripain was studied in detail by acyl transfer
experiments to libraries of unnatural amino components
It is well-documented that proteases catalyze in vitro
the hydrolysis of peptides and proteins as well as the
reverse of hydrolysis.1 On the basis of this activity, these
enzymes can be used as biocatalysts for peptide bond
formations apparently opposite to their original in vivo
function. Due to the mild synthesis conditions and the
high degree of stereo- and regiospecificity of such reac-
tions, enzymatic peptide coupling is considered to be an
attractive alternative for solution and solid phase peptide
synthesis. However, despite these undisputed advan-
tages, the high substrate specificity of proteases seriously
limits the amino acid residues between which a peptide
bond can be synthesized. Due to the original specificity
of proteases for proteinogenic amino acids, the coupling
of modified and noncoded amino acids or non-amino acid-
derived derivatives is especially difficult. Therefore, only
a few papers report the protease-mediated synthesis of
peptides containing artificial amino acids or non-amino
acid-derived functionalities.2 But even in these cases the
product yields usually drop with an increasing degree of
modifications of both the amino acid side chain and the
backbone structure particularly in the region of synthe-
sis.3
(2) (a) Widmer, F.; Breddam, K.; J ohansen, J . T. Carlsberg Res.
Commun. 1981, 46, 97. (b) Cerovsky, V.; J akubke, H.-D. Int. J . Pept.
Protein Res. 1994, 44, 466. (c) Margolin, A. L.; Tai, D.-F.; Klibanov, A.
M. J . Am. Chem. Soc. 1987, 109, 7885. (d) Schuster, M.; Munoz, B.;
Yuan, W.; Wong, C.-H. Tetrahedron Lett. 1993, 34, 1247. (e) Wong,
C.-H.; Schuster, M.; Wang, P.; Sears, P. J . Am. Chem. Soc. 1993, 115,
5893. (f) Schuster, M.; Wang, P.; Paulson, J . C.; Wong, C.-H. J . Am.
Chem. Soc. 1994, 116, 1135. (g) Bordusa, F.; Dahl, C.; J akubke, H.-
D.; Burger, K.; Koksch, B. Tetrahedron: Asymmetry 1999, 10, 307.
(3) (a) Chen, S. T.; Chen, S. Y.; Wang, K. T. J . Org. Chem. 1992,
57, 6960. (b) Moree, W. J .; Sears, P.; Kawashiro, K.; Witte, K.; Wong,
C.-H. J . Am. Chem. Soc. 1997, 119, 3942. (c) Kitaguchi, H.; Klibanov,
A. M. J . Am. Chem. Soc. 1989, 111, 9272. (d) Cantacuzene, D.;
Guerreiro, C. Tetrahedron 1989, 45, 741.
* Corresponding author. E-mail: bordusa@rz.uni-leipzig.de. Tel.:+49
341 9736918. Fax.:+49 341 9736998.
(4) (a) J akubke, H.-D. Peptide, Chemie und Biologie; Spektrum
Akademischer Verlag: Heidelberg, Berlin, Oxford, 1996. (b) Valle, G.;
Crisma, M.; Toniolo, C.; Polinelli, S.; Boesten, W. H. J .; Schoemaker,
H. E.; Meijer, E. M.; Kamphius, J . Int. J . Peptide Protein Res. 1991,
37, 521. (c) Benedetti, E. Biopolymers 1996, 40, 3. (d) Karle, I. L.
Biopolymers 1996, 40, 157. (e) Giannis, A.; Kolter, T. Angew. Chem.
1993, 105, 1303; Angew. Chem., Int. Ed. Engl. 1993, 32, 1244.
(5) (a) Rich, D. H. In Comprehensive Medicinal Chemistry; Sammes,
P. G., Ed.; Pergamon: Oxford, 1990; Vol. 2, p 391. (b) Alewood, P. F.;
Brinkworth, R. I.; Dancer, R. J .; Garnham, B.; J ones, A.; Kent, S. B.
H. Tetrahedron Lett. 1992, 33, 977. (c) Scholz, D.; Billich, A.; Charpiot,
B.; Ettmayer, P.; Lehr, P.; Rosenwirth, B.; Schreiner, E.; Gstach, H.
J . Med. Chem. 1994, 37, 3079. (d) Slee, D. H.; Laslo, K. L.; Elder, J .
H.; Ollmann, I. R.; Gustchina, A.; Kervinen, J .; Zdanov, A.; Wlodawer,
A.; Wong, C.-H. J . Am. Chem. Soc. 1995, 117, 11867.
† Abbreviations: Boc, tert-butyloxycarbonyl; Bz, benzoyl; DCC, N,N′-
dicyclohexylcarbodiimide; DMAP, 4-(dimethylamino)pyridine; DMF,
dimethylformamide; DTT, DL-dithiothreitol; HEPES, (N-[2-hydroxy-
ethyl]piperazine-N′-[2-ethanesulfonic acid]); OEt, ethyl ester; OGp,
4-guanidinophenyl ester; Z, benzyloxycarbonyl.
‡ Max-Planck Society.
§ University of Leipzig.
(1) (a) Schellenberger, V.; J akubke, H.-D. Angew. Chem. 1991, 103,
1440; Angew. Chem., Int. Ed. Engl. 1991, 30, 1437. (b) Wong, C. H.
Science 1989, 244, 1145. (c) J ackson, D. Y.; Burnier, J .; Quan, C.;
Stanley, M.; Tom, J .; Wells, J . A. Science 1994, 266, 243. (d) Wong,
C.-H.; Whitesides, G. M. Enzymes in Synthetic Organic Chemistry;
Pergamon: Oxford, 1994; pp 41-130 and references therein.
10.1021/jo991302q CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/02/2000

Enzyme Specificity of Clostripain
J . Org. Chem., Vol. 65, No. 6, 2000 1673
duced analogously to the definition of the Michaelis
constant KM (eq 1).10
k3Kn
k4
k
d[Ac-OH]
d[Ac-N]
p )
[HN] )
+
5[HN] ) p0 + pn[HN]
k4
(1)
According to eq 1, the p value is defined as the
nucleophile concentration at which the rates of both
reactions are equal corresponding to 50% yield. Conse-
quently, a decrease in the p value is inversely correlated
with an increase in the peptide yield. Using the nucleo-
phile in excess, the partition value can be determined
from the product ratios according to eq 2, where [HN] is
F igu r e 1. Schematic structure of substrate mimetics com-
pared to common acyl donor components.
Sch em e 1. Kin etic Mod el of P r otea se-Ca ta lyzed
Acyl Tr a n sfer Rea ction a
[Ac-OH]
[Ac-N]
p )
[HN]
(2)
the initial nucleophile concentration and [Ac-OH] and
[Ac-N] represent the product concentrations.
Acyl Tr a n sfer Stu d ies. The ability of clostripain to
catalyze acylations of a wide variety of nucleophilic
components was explored using Bz-Arg-OEt as the acyl
donor and about 40 amines of different structure as acyl
acceptors. At first, the function of clostripain for coupling
of aliphatic noncyclic and cyclic amines varying in chain
length or ring size was investigated. Additionally, this
series was expanded by the use of two aromatic amines
(compounds 9 and 10). The resulting partition values of
coupling reactions are summarized in Table 1. The data
document that, except for compound 4, all amines show
productive binding at the S′-subsite of clostripain result-
ing in amide product formation. Remarkably, p values
within a range of approximately 1 to 5 mM were evident
for most derivatives corresponding to results found for
most efficient peptidic acyl acceptors.7 This fact also holds
for the acylation of the bulky cyclic amines 5-7 at least
to a ring size of eight methylene groups. Since these
substrates do not have a common peptide structure, this
finding suggests that clostripain does not form close
contacts to the backbone, in particular to the amide
function of P′2 amino acid residues, contrary to most other
proteases.11 Instead of this, the high reactivity toward
these bulky hydrophobic amines indicates a more flexible
and relaxed S′-subsite binding region. Only the use of
cyclododecylamine (compound 8) drops the efficiency of
catalysis indicated by the increase in p. In this case
mainly competitive hydrolysis of the acyl donor was
observed. An increase of the p value was also found for
the branched noncyclic 2-aminopentane (3). In contrast,
the nonbranched analogues 1 and 2 gave more than 1
order of magnitude lower partition values, indicating an
unfavorable effect of a neighboring branch on the CN-
bond formation. Further alkylation of the neighboring
carbon resulting in dialkylated amines causes a complete
loss of reactivity as found for reactions with tert-butyl-
amine (4) as the acyl acceptor. This observation corre-
sponds with the failed acceptance of clostripain for the
a
EH, free enzyme; Ac-X, acyl donor; HX, leaving group; Ac-E,
acyl enzyme complex; Ac-OH, hydrolysis product; HN, acyl
acceptor; Ac-N, aminolysis product.
derived from noncoded amino acids, cyclic, noncyclic, and
aromatic amines, amino alcohols, and symmetric and
asymmetric diamines, respectively. Clostripain, produced
and secreted by proliferating cells of the anaerobic
microorganism Clostridium histolyticum, originally hy-
drolyzes peptide bonds at the carboxyl side of arginine
and, much more slowly, of lysine residues.7 For that
reason, we started our study using the conventional
substrate Bz-Arg-OEt as the acyl donor and expanded it
to the substrate mimetic Bz-Phe-OGp (Figure 1).8 The
latter represents a new type of acyl donors which are
recognized by the enzyme on the basis of specific interac-
tions with the 4-guanidinophenyl ester leaving group
(OGp). This highly specific leaving group binds instead
of the specificity-determining arginine side chain func-
tionality of common acyl donors largely independently
of its C-terminal amino acid residue. As a result, peptide
bond formation occurs independently of the primary
specificity of the enzyme, enabling clostripain to react
also with nonspecific acyl moieties.9
Resu lts a n d Discu ssion
Dea cyla tion Kin etics. In the presence of added
nucleophilic amino components, cysteine and serine
proteases are capable of catalyzing acyl transfers to
nucleophiles simultaneously with the hydrolysis of the
acyl donors (Scheme 1). Consequently, the resulting
hydrolysis product Ac-OH, which possesses a very low
acylation potential, and the resulting peptide product
Ac-N are formed. As an efficiency parameter of the two
competitive reactions, the partition value p was intro-
C
R,R-dialkylated amino acid analogue H-Aib-NH2.2g The
(6) Schechter, I.; Berger, A. Biochem. Biophys. Res. Commun. 1967,
27, 157.
(7) (a) Ullmann, D.; J akubke, H.-D. Eur. J . Biochem. 1994, 223, 865.
(b) Bordusa, F.; Ullmann, D.; J akubke, H.-D. Biol. Chem. 1997, 378,
193.
(8) Thormann, M.; Thust, S.; Hofmann, H.-J .; Bordusa, F. Biochem-
istry 1999, 38, 6056.
(9) Bordusa, F.; Ullmann, D.; Elsner, C.; J akubke, H.-D. Angew.
Chem. 1997, 109, 2583; Angew. Chem., Int. Ed. Engl. 1997, 36, 2473.
comparatively high p value found for aniline (9), which
is similar to that of 2-aminopentane (3), might also be
(10) Ko¨nnecke, A.; Schellenberger; V.; Hofmann, H.-J .; J akubke, H.-
D. Pharmazie 1984, 39, 785.
(11) (a) Segal, D. M.; Powers, J . C.; Cohen, G. H.; Davies, D. R.;
Wilcrox, P. E. Biochemistry 1971, 10, 3728. (b) Bizzozero, S. A.;
Baumann, W. K.; Dutler, H. Eur. J . Biochem. 1982, 122, 251.

1674 J . Org. Chem., Vol. 65, No. 6, 2000
Gu¨nther et al.
Ta ble 1. P a r tition Va lu es p of Clostr ip a in -Ca ta lyzed Acyl Tr a n sfer Rea ction s of Bz-Ar g-OEt w ith Alk yl a n d Ar yl
Am in esa
a
Conditions: 0.025 M Na2CO3/NaHCO3 buffer, pH 10.0, 0.2 M NaCl, 0.01 M CaCl2, 25 °C, [acyl donor] ) 2 mM, [acyl acceptor]0 ) 20
mM, [clostripain] ) 5.0 × 10-7 M.
Ta ble 2. P a r tition Va lu es p of Clostr ip a in -Ca ta lyzed Acyl Tr a n sfer Rea ction s of Bz-Ar g-OEt w ith Va r iou s Am in o
Alcoh olsa
a
Conditions: 0.025 M Na2CO3/NaHCO3 buffer, pH 10.0, 0.2 M NaCl, 0.01 M CaCl2, 25 °C, [acyl donor] ) 2 mM, [acyl acceptor]0 ) 20
mM, [clostripain] ) 5.0 × 10-7 M.
the result of this unfavorable branch-effect. Correspond-
ingly, in the case of benzylamine (10), the shift of the
branch by the additional methylene group leads to a
significant increase in the degree of amine acylation.
Interestingly, a branch with longer aliphatic residues
seems to compensate partly for this unfavorable branch-
effect, as suggested by the high reactivity of leucinol (26)
documented in Table 2. Furthermore, Table 2 illustrates
the influence of additional functionalities, e.g., hydroxyl
groups, within the amino component on clostripain-
catalyzed amide formations. Remarkably, in no case a
competitive formation of an O-acylated product occurred,

Enzyme Specificity of Clostripain
J . Org. Chem., Vol. 65, No. 6, 2000 1675
Ta ble 3. P a r tition Va lu es p of Clostr ip a in -Ca ta lyzed Acyl Tr a n sfer Rea ction s of Bz-Ar g-OEt w ith Sym m etr ic a n d
Asym m etr ic Alk yl Dia m in esa
a
Conditions: 0.025 M Na2CO3/NaHCO3 buffer, pH 10.0, 0.2 M NaCl, 0.01 M CaCl2, 25 °C, [acyl donor] ) 2 mM, [acyl acceptor]0 ) 20
mM, [clostripain] ) 5.0 × 10-7 M.
as also indicated by the failed acceptance of the enzyme
for 1-pentanol (27). A comparison of the partition values
found for the individual amino alcohols with those
observed for the appropriate unsubstituted amines does
not show significant differences (cf. Table 1). Hence,
clostripain-catalyzed reactions seem to be nearly unaf-
fected by additional hydroxyl groups of the acyl acceptors.
This behavior also holds for the acylation of the dihy-
droxylated 3-amino-1,2-propanediol (24). Corresponding
to this, reactions using alaninol (25) result in p values
comparable to that found for the non-hydroxylated
analogue 2-aminopentane (3). Thus, clostripain catalyzes
the acylation of amino alcohols not only very efficiently
like that of unsubstituted amines, but also with a high
selectivity to the amino group of these acyl acceptors.
The selectivity of clostripain catalysis was further
investigated by using various aliphatic diamines (35-
41) as acyl acceptors. The results observed for reactions
with Bz-Arg-OEt are summarized in Table 3. In most
cases, the p values are comparable with those of the
similar single amines. Obviously, as already found for
the extra hydroxyl groups of amino alcohols, further
amino groups also appear to have no significant influence
on the efficiency of enzyme catalysis, thus qualifying
diamines as specific acyl acceptors for clostripain, too.
Only the reactions with 1,6-diaminohexane (38) resulted
in a slightly higher p value, indicating unfavorable
interactions of the extra amino group with the S′-subsite
of clostripain. A partly positive surface charge at the
appropriate enzyme region could explain this phenom-
enon. However, such a positively charged domain should
be strong locally restricted, as shown by the results
observed for the ω-amino carboxylic acids (Table 4). As
indicated by the higher p values found for these amino
components, a negative surface charge should predomi-
nate within the S′-subsite of clostripain. As illustrated
in Figure 2, the strongest negative potential should exist
very near the active site. Therefore, the shortest sub-
strate is also the poorest one, whereas a shift of the
negatively charged carboxylate group to more remote
F igu r e 2. Influence of various functional groups at different
positions of the acyl acceptor of the general structure H2N-
(CH2)nR on log p of clostripain-catalyzed acyl transfer reac-
tions. R: ([) COOCH3; (b) COOH; (9) CONH2; (2) NH2. To
facilitate the understanding of these semilogarithmic plots, the
symbols are connected by dotted lines. Conditions: 0.025 M
Na2CO3/NaHCO3 buffer, pH 10.0, 0.2 M NaCl, 0.01 M CaCl2,
25 °C, [acyl donor]: Bz-Arg-OEt ) 2 mM, [acyl acceptor]0 )
20 mM, [clostripain] ) 5.0 × 10-7 M.
positions increases the rate of acylation. The optimum
of reactivity found for 6-aminohexanoic acid (52) in-
versely correlates with the lower reactivity observed for
1,6-diaminohexane (38), confirming the postulated change
of surface charges in the appropriate S′-subsite region.
The reactivity of the corresponding carboxylic acid amides
(54-58) and esters (59-63) are practically unaffected by
these electrostatic effects. In both cases the reactivity
increases with the number of methylene groups and,
therefore, with the degree of hydrophobicity, whereas the
esters act as the most efficient substrates. Hence, from
a synthetic point of view, the enzymatic acylation of
amino carboxylic acids should be performed using the
appropriate ester analogues which can be subsequently
hydrolyzed to the corresponding free acids.

1676 J . Org. Chem., Vol. 65, No. 6, 2000
Gu¨nther et al.
Ta ble 4. P a r tition Va lu es p of Clostr ip a in -Ca ta lyzed Acyl Tr a n sfer Rea ction s of Bz-Ar g-OEt w ith Non -r-a m in o
Ca r boxylic Acid Der iva tivesa
a
Conditions: 0.025 M Na2CO3/NaHCO3 buffer, pH 10.0, 0.2 M NaCl, 0.01 M CaCl2, 25 °C, [acyl donor] ) 2 mM, [acyl acceptor]0 ) 20
mM, [clostripain] ) 5.0 × 10-7 M.
Analyzing the selectivity of enzymatic acylation of the
diamines, a single acylation was found in almost all cases.
Traces of double-acylated product could only be detected
for ethylendiamine (35) with approximately 3% of the
whole amide product (42). Interestingly, the asymmetric
1,3-diaminopentane (41) was acylated by clostripain
exclusively at the amino group at 1-position, whereas an
acylation of the amino group at 3-position did not occur.
This finding corresponds to the already mentioned ob-
servation that clostripain accepts branched amines less
specifically than nonbranched derivatives. In agreement
with these results, clostripain discriminates between the
two primary amino groups of 41 forming exclusively the
single-acylated isomer at 1-position. From a synthetic
point of view, this distinct selectivity makes clostripain
a useful and easy tool for selective acylations of amino
components with various amino groups differing in the
degree of branching without the need of additional
protection and deprotection steps.
F igu r e 3. Influence of the chain length and concentration of
the acyl acceptor on the amount of single-acylated amine x in
clostripain-catalyzed reactions. From white to black: 1,2-
diaminoethane, 1,4-diaminobutane, 1,6-diaminohexane, 1,8-
diaminooctane. Conditions: 0.025 M Na2CO3/NaHCO3 buffer,
pH 10.0, 0.2 M NaCl, 0.01 M CaCl2, 25 °C, [acyl donor] ) 2
mM, [acyl acceptor]0 ) 20 mM, [clostripain] ) 5.0 × 10-7 M.
It is well-known in organic chemistry that the ratio
between single and double acylation of symmetric di-

Enzyme Specificity of Clostripain
J . Org. Chem., Vol. 65, No. 6, 2000 1677
F igu r e 4. Course of selected clostripain-catalyzed acyl transfer reactions of Bz-Arg-OEt with several acyl acceptors: [a]
pentylamine; [b] 5-aminopentanol; [c] 1,4-diaminobutane. (-9-) acyl donor Bz-Arg-Et, (-2-) aminolysis product, (-b-) hydrolysis
product. Conditions: 0.025 M Na2CO3/NaHCO3 buffer, pH 10.0, 0.2 M NaCl, 0.01 M CaCl2, 25 °C, [acyl donor] ) 2 mM, [acyl
acceptor]0 ) 8 mM, [clostripain] ) 5.0 × 10-7 M.
Ta ble 5. P a r tition Va lu es p of Clostr ip a in -Ca ta lyzed Acyl Tr a n sfer Rea ction s of Bz-P h e-OGp w ith Selected Am in es a n d
Am in o Alcoh olsa
a
Conditions: 0.2 M HEPES buffer, pH 8.0, 0.2 M NaCl, 0.01 M CaCl2, 2.5% DMF, 25 °C, [acyl donor] ) 2 mM, [acyl acceptor]0 ) 20
mM at pH 10.0, [clostripain] ) 3.3 × 10-6 M.
amines depends on the ratio of reactant concentrations.
To evaluate whether this ratio also affects the enzymatic
reaction, acyl transfers using various acyl acceptor
concentrations were performed. Figure 3 illustrates the
amount of single-acylated diamine on the whole amide
product. In analogy to other chemical reactions, it is
obvious that the degree of double acylation increases with
a decrease of the acyl acceptor concentration. Therefore,
the highest amount of double-acylated amide product was
found at equimolar concentrations of both reactants.
However, in contrast to chemical reactions, the enzymatic
approach seems to be additionally affected by the chain
length in the symmetric diamine. Whereas the short 1,2-
diaminoethane shows the highest degree of double acy-
lation, an elongation of the chain length favors the
formation of a single-acylated product. Consequently, the
amount of double-acylated product drops synergistically
both by the increase of acyl acceptor excess and by the
elongation of the chain length of the diamine.
the enzymatic reaction needs an accurate time control
to avoid losses of peptide product. Usually, the specificity
data derived from the synthesis reactions correspond to
the data derived from the native hydrolysis reaction.12
Accordingly, the high acylation rates found for the most
non-amino acid-derived amines indicate a high specificity
of the enzyme toward the formed sequence and, therefore,
equally high rates of hydrolysis. But indeed all synthesis
reactions occur practically irreversible without any sec-
ondary hydrolysis as exemplary illustrated by Figure 4.
Even after reaction times of 1 and 2 days, practically no
cleavage of the synthesis products occurs. This peculiarity
was also found for acyl acceptors whose structure closely
corresponds to that of coded amino acids, as for instance,
leucinol (26). This behavior becomes even more remark-
able since clostripain is well-known for its high amidase
activity which usually results in a fast cleavage of the
newly formed peptide bond.7a Up to now it is unclear
whether this unusual ”substrate”-induced effect is based
The course of the kinetically controlled peptide syn-
thesis by protease catalysis is typically characterized by
an optimum of product formation followed by a secondary
hydrolysis of the newly formed peptide bond. Therefore,
(12) (a) Fersht, A. R.; Blow, D. M.; Fastrez, J . Biochemistry 1973,
12, 2035. (b) Schellenberger, V.; Braune, K.; Hofmann, H.-J .; J akubke,
H.-D. Eur. J . Biochem. 1991, 199, 623. (c) Schellenberger, V.; Siegel,
R. A.; Rutter, W. J . Biochemistry 1993, 32, 4344.

1678 J . Org. Chem., Vol. 65, No. 6, 2000
Gu¨nther et al.
on an incorrect binding of these synthesis products or an
incomplete catalysis. Since these derivatives still ef-
ficiently bind to clostripain (data not shown), this prin-
ciple may be useful for the design of specific as well as
selective inhibitors. From a synthetic point of view, there
is no risk of any proteolytic side reactions and, therefore,
no need of time control of the enzymatic synthesis
reaction.
for the synthesis of peptide isosteres. Remarkably, de-
spite the high activity of clostripain for the synthesis of
peptide isosteres, the enzyme is practically inactive
toward the cleavage of the newly formed amide bonds.
Up to now, the molecular basis for this atypical behavior
is not yet understood. In the case of substrate mimetics,
unwanted proteolytic cleavages are generally prevented
due to the coupling of nonspecific acyl residues. Further-
more, the remarkably high efficiency of reactions using
the substrate mimetic Bz-Phe-OGp as the acyl donor
demonstrated that clostripain is capable of acylating non-
amino acid-derived amines with nonspecific acyl residues.
Since the classical approach seriously limits the use of
clostripain to reactions with Arg- and Lys-containing acyl
donors, this combination considerably expands the syn-
thetic utility of this enzyme. Based on the potential
acceptance of substrate mimetics derived from non-amino
acid derivatives, this approach may open up a new field
of synthetic applications of proteases completely outside
peptide synthesis, achieving both efficient and selective
organic amide bond formations under extraordinarily
mild reaction conditions.
Apart from the broad nucleophile specificity described
here, clostripain possesses a narrow primary specificity
for arginine. Whereas the coupling of lysine at P1-position
is still possible,7b all other amino acids are not accepted
by the enzyme. Consequently, the synthetic utility of this
protease is originally limited to both of these amino acid
residues. Recently, by use of 4-guanidinophenyl esters
of acylated amino acid derivatives, we could show that
clostripain recognizes this special type of substrates even
when derived from nonspecific amino acids.9 As a result,
peptide bond formation occurs independently of the
primary specificity of the enzyme. There is no doubt that
the combination of both features would considerably
extend the application range of this enzyme for synthesis.
Therefore, model reactions with Bz-Phe-OGp as the acyl
donor and selected amino components were performed.
The p values observed for these reactions are given in
Table 5. The results convincingly show that Bz-Phe-OGp
acts as an efficient acyl donor acylating non-amino acid-
derived amino components. Remarkably, clostripain cata-
lyzes acylations with the nonspecific Bz-Phe-residue even
more efficiently than those with the specific arginine one,
as indicated by the mostly 1 order of magnitude lower p
values. Thus, only traces of Bz-Phe-OH were formed by
the enzyme. These results become still more impressive,
since only small amounts of the amino components are
unprotonated at the pH value of 8.0 which is dictated by
the lower intrinsic stability of the guanidinophenyl ester.
Therefore, only a small amount of the whole amine can
serve as a deacylating component (0.2 to 3.7% depending
on the individual pK value). This finding indicates that
the strategy used for these substrate mimetics can be
successfully combined with the broad tolerance of clos-
tripain toward the acyl acceptor, realizing efficient CN-
bond formations under mild pH conditions in aqueous
systems.
Exp er im en ta l Section
Ma ter ia ls. Clostripain (EC 3.4.22.8) was a gift from Fluka
Chemie AG, Switzerland, and had a specific activity of 100 U
mg-1 (1 U ) amount of enzyme that hydrolyzes 1 µmol min-1
NR-Bz-L-arginine-ethyl ester at pH 7.1 and 25 °C). The enzyme
was activated before any applications for 2-3 h in the presence
of 1.0 mM CaCl2 containing 2.5 mM DTT. Amino acids, amines,
4-aminophenol, DCC, DMAP, benzyl chloroformate (Z-chlo-
ride), S-methylisothiourea, and 4-toluenesulfonic acid were
purchased from commercial suppliers. All reagents were of the
highest available commercial purity. Solvents were purified
and dried by the usual methods. Mass spectra were recorded
using thermospray ionization or by MALDI-ToF.
Ch em ica l Syn th eses. Bz-Phe-OGp was prepared by con-
densation of Boc-Phe-OH and 4-[N′,N′′-bis(Z)guanidino]phenol
following the procedure described by Sekizaki et al.13 Because
of the more advantageous preparation and higher reactivity,
N,N′-bis(Z)-S-methylisothiourea was used for the amidination
of 4-aminophenol to synthesize 4-[N′,N′′-bis(Z)guanidino]phe-
nol14 instead of 1-[N,N′-bis(Z)amidino]pyrazole. The benzoyl-
protected ester was prepared, starting with the deprotection
of the NR-amino group with trifluoroacetic acid and subsequent
benzoylation using benzoyl chloride. A final catalytic hydro-
genation of the bis(Z)-protected ester results in the 4-N′,N′′-
deprotected 4-guanidinophenyl ester. The amino acid methyl
esters were synthesized following the procedure described by
Brenner et al.15 The amino acid amides were prepared from
the appropriate methyl esters by ammonolysis with ammonia.
The identity and purity of all final products were checked by
analytical HPLC (220 nm), NMR, thermospray mass spectros-
copy, and elemental analysis. In all cases, satisfactory elemen-
tal analyses data were found ((0.4% for C, H, N).
Con clu sion
The use of proteases as biocatalysts for highly selective
CN-bond formations is strictly connected with the sub-
strate specificity of these enzymes and, therefore, usually
restricted toward coded amino acid residues. In contrast,
the results of this study indicate a completely different
behavior for the cysteine protease clostripain. The en-
zyme is capable of binding and acylating aliphatic amines
widely varying in structure as well as amino alcohols,
non-R-amino carboxylic acids, and symmetric and asym-
metric diamines. Moreover, for most derivatives, acyla-
tion rates could be observed higher than those found for
peptidic acyl acceptors, indicating a high flexible and
relaxed S′-subsite which seems to be predominated by
hydrophobic domains. Whereas the presence of polar
hydroxyl groups or positively charged amino functions
usually does not reduce the specificity of binding, amines
with free carboxylate groups show lower reactivities. On
the basis of these results, this study further provides a
guideline for the application of clostripain as biocatalyst
En zym a tic Syn th eses. Enzymatic reactions with Bz-Arg-
OEt were performed in 0.025 M Na2CO3/NaHCO3 buffer, pH
10.0 containing 0.2 M NaCl and 0.01 M CaCl2 at 25 °C.
Reactions with Bz-Phe-OGp were performed in 0.2 M HEPES
buffer, pH 8.0, containing 0.2 M NaCl, 0.01 M CaCl2, and 2.5%
DMF at 25 °C. Stock solutions of acyl donor esters (4 mM)
were prepared in distilled water. In the case of Bz-Phe-OGp,
5% DMF was added to realize a complete solubility of the ester.
Amino components were dissolved in 0.4 M HEPES buffer (pH
8.0) or 0.05 M Na2CO3/NaHCO3 buffer (pH 10.0) containing
(13) Sekizaki, H.; Itoh, K., Toyota, E.; Tanizawa, K. Chem. Pharm.
Bull. 1996, 44, 1577.
(14) Lal, B.; Gangopadhyay, A. K. Tetrahedron Lett. 1996, 37, 2483.
(15) Brenner, M.; Mu¨ller, H. R.; Pfister, R. W.; Helv. Chim. Acta
1950, 33, 568.

Enzyme Specificity of Clostripain
J . Org. Chem., Vol. 65, No. 6, 2000 1679
0.4 M NaCl and 0.02 M CaCl2. To neutralize hydrochlorides
or hydrobromides, appropriate equivalents of NaOH were
added to the stock solutions of the amino components. If not
otherwise stated, the final acyl donor and nucleophile concen-
tration was 2 mM and 20 mM, respectively. The concentration
of the amino components was calculated as free, NR-unproto-
nated nucleophile concentration [HN]0 according to the formal-
ism of Henderson-Hasselbalch [HN]0)[N]0/(1 + 10pK-pH). For
the reactions with Bz-Phe-OGp, the amines were used in
identical concentrations as used for reactions with Bz-Arg-OEt.
In these cases the actual concentration of [HN]0 within the
reaction mixture was calculated to the lower pH value of 8.0.
After thermal equilibration of the assay mixtures, the reactions
were started by addition of the enzyme, resulting in active
enzyme concentrations of 5.0 × 10-7 M for reactions with Bz-
Arg-OEt and 3.3 × 10-6 M in the case of Bz-Phe-OGp. Reaction
times of 10-120 min led to a complete ester consumption. For
the HPLC analysis aliquots were withdrawn and diluted with
a quenching solution of 50% aqueous methanol containing 1%
trifluoroacetic acid. For each acyl donor and acyl acceptor, an
experiment without enzyme was carried out for determining
the extent of spontaneous ester hydrolysis which was strictly
less than 5%. On the basis of the same control experiments,
nonenzymatic aminolysis of the acyl donor esters was inves-
tigated and could be ruled out. The results reported are the
average of at least three independent experiments. The
identity of the formed peptide and amide products was
validated by thermospray and MALDI-ToF mass spectroscopy,
respectively. NMR measurements were used for the examina-
tion of the enzymatically formed amide bonds.
HP LC An a lyses. Samples were analyzed by analytical
reversed phase HPLC on C18 polymer coated columns [Vydac
218TP54, 5 µm, 300 Å, (25 × 0.4 cm), and Grom Capcell, 5
µm, 300 Å, (25 × 0.4 cm)] and a C4 reversed phase column
[Vydac 214TP, 10 µm, 300 Å, (25 × 0.4 cm)] which were
thermostated at 25 °C and eluted with various mixtures of
water/acetonitrile containing 0.1% trifluoroacetic acid under
gradient conditions. Detection was at 254 nm to monitor the
aromatic chromophores of the acyl donors. Thus, the yields
could be determined from the peak areas of the hydrolysis and
aminolysis products, whereby 4-toluenesulfonic acid served as
an internal standard. In the case of chromophoric amino
components, the yields were calibrated from the lack of
hydrolysis products by at least five independent experiments.
Ack n ow led gm en t. This work was supported by the
Deutsche Forschungsgemeinschaft (Innovationskolleg
“Chemisches Signal und biologische Antwort” and J A
559/9-1) and Fonds der Chemischen Industrie (Liebig
Scholarship, F.B.). We thank Mrs. Doris Haines and
Mrs. Regina Schaaf for skillful technical assistance. The
authors thank ASTA Medica AG and Fluka AG for
special chemicals.
J O991302Q