вход по аккаунту


Basic Principles of Protease-Catalyzed Peptide Bond Formation.

код для вставкиСкачать
[I] R. M. Wenger, Transplant. Proc. 15 No. 4, Suppl. l(1983) 2230.
[2] J. F. Borel, C. Feurer, H. U. Gubler, H. Stahelin, Agents Actions 6 (1976)
131 J . F. Borel, C. Feurer, C. Magnee, Immunology 31 (1976) 631.
[4] J. F. Borel, C. Feurer, C. Magnee, H. Stahelin, Immunology 32 (1977)
151 J. F. Borel, D. Wiesinger, H. U. Gubler, Eur. J. Rheumatol. Inflammation
I (1978) 97.
[6] D. Wiesinger, J. F. Borel, Immunobiolog. 156 (1979) 454.
[7] B. E. Bueding, J. Hawkins, Y. N . Cha, Agents Actions 11 (1981) 380.
[8] K. Thommen-Scott, Agents Actions 1 1 (1981) 770.
191 R. J. Calne, D. J. White, S. Thiru, D. B. Evans, P. McMaster, D. C.
Dunn, G. N. Craddock, D. B. Pentlow, K. Rolles, Lancet ii 1978, 1323.
[lo] R. L. Powles, A. J. Barrett, H. Clink, H. E. M. Kay, J. Sloane, T. J. McEIwain. Lancer ii 1978, 1327.
[ I I] R. M. Wenger in D. J. G. White: Cyclosporin A , Biomedical Press, Amsterdam 1982, p. 19.
[I21 R. M. Wenger, Helu. Chim. Acta 67(1984) 502.
[I31 A. Ruegger, M. Kuhn, H. Lichti, H. R. Loosli, R. Huguenin, C. Quiquerez, A. von Wartburg, Helv. Chim. Acta 59 (1976) 1072.
1141 M. Dreyfuss, E. Harri, H. Hofman, H. Kobel, W. Pache, H. Tscherter,
Eur. J . Appl. Microhiol. 3 (1976) 125.
1151 W. Gams, Per.soonia 6 (1971) 185.
[I61 H. Kobel, R. Traber, Eur. J. Appl. Microbiol. Biotechnol. 14 (1982) 237.
(171 R. Traber, M. Kuhn, A. Ruegger, H. Lichti, H. R. Loosli, A. von Wartburg, Helu. Chim. Acta 60 (1977) 1247.
1181 R. Traber, M. Kuhn, H. R. Loosli, W. Pache, A. von Wartburg, Helu.
Chim. Acta 60 (1977) 1568.
1191 R. Traber, H. R. Loosli, H. Hofman, M. Kuhn, A. von Wartburg, Helu.
Chrm. Acta 65 (1982) 1655.
1201 H. Kobel, R. Traber, Eur. J. Appl. Microbiol. Biolechnol. I4 (1982) 237.
1211 T. J. Petcher, H. P. Weber, A. Ruegger, Helu. Chim. Acta 59 (1976)
1221 H. R. Loosli, H. Kessler, H. Oschkinat, H. P. Weber, T. J. Petscher, A.
Widmer, Helu. Chim. Acta 68 (1985), in press.
[23] P. Chou, G. D. Fasman, J. Mol. Biol. I15 (1977) 135.
1241 J. A. Smith, L. G. Pease, CRC Crit. Reu. Biochem. 8 (1980) 316.
I251 J. S. Richardson, Ado. Prorein Chem. 34 (1981) 167.
1261 H. Kobel, H. R. Loosli, R. Voges, Experientra 39 (1983) 873.
1271 R. Zocher, H. Kleinkauf, Eiochem. Biophys. Res. Commun. 81 (1978)
1281 R. Zocher, N. Madry, H. Peeters, H. Kleinkauf, Phytochemistry 23 (1984)
(291 R. M. Wenger, Helv. Chim. Actn 66 (1983) 2308.
1301 R. S. Cahn, C. Ingold, V. Prelog, Angew. Chem. 78 (1966) 413: Angew.
Chem. I n t . Ed. Engl. 5 (1966) 385.
(311 D. Seebach, E. Hungerbuhler in R. Scheffold: Modern Synthetic Merhods, Salle & Sauerlander, FrankfurtIAarau 1980, p. 93.
1321 M. Schlosser, K. F. Christmann, Justus Liebigs Ann. Chem. 708 (1967) 1.
[33] D. A. Seeley, H. McElwee, J. Org. Chem. 38 (1973) 1691.
1341 R. W. Herr, D. M. Wieland, C. R. Johnson, J. Am. Chem. SOC.92 (1970)
[35] K. E. Pfitzner, J. G. Moffatt, 1. Am. Chem. Sor. 87(1965) 5670.
1361 R. M. Wenger, Helu. Chim. Acta 66 (1983) 2672.
1371 J. R. McDermott, N. L. Benoiton, Can. J. Chem. 51 (1973) 2551, 2562.
[38] M. Zaoral, Collect. Czech. Chem. Commun. 27 (1962) 1273.
1391 W. Konig, R. Geiger, Chem. Ber. 103 (1970) 788.
1401 B. Castro, J. R. Dormoy, J. G. Evin, C. Selve, Tetrahedron Lett. 1975,
1411 H. Wissmann, H. J. Kleiner, Angew. Chem. 92 (1980) 129; Angew. Chem.
Int. Ed. Engl. 19 (1980) 133; J . Kovacs in E. Gross, J. Meienhofer: The
Peprides, Vol. 2. Academic Press, New York 1980, p. 485.
Basic Principles of Protease-Catalyzed Peptide Bond Formation
By Hans-Dieter Jakubke*, Peter Kuhl, and Andreas Konnecke
The stereospecific formation of peptide bonds under mild conditions and without side reactions is still a formidable task in peptide synthesis. One approach that springs to mind,
namely the use of the naturally occurring catalyst involved in the biosynthesis of proteins,
the ribosomal peptidyl transferase, cannot be realized in practice. The fact, however, that
the natural cleavage of proteins is carried out by other enzymes, namely the proteases, together with the reversibility of these cleavage reactions in principle, has led to an interesting
synthetic concept. Proteases normally catalyze the enzymatic degradation of proteins and
peptides by hydrolytic cleavage of the peptide bond in an exergonic reaction. The use of
physicochemical principles in order to influence the equilibrium, the concentration of products, and the kinetic parameters of the reaction results in the successful application of the
catalytic properties of proteases to peptide synthesis. The purpose of this review is to describe and summarize the methods used in such approaches and to attempt a systematic categorization. The principles are applied to the synthesis of such practically relevant products as aspartame and human insulin.
1. Introduction
Peptide synthesis has made a significant contribution
over the last 30 years to progress in peptide re~earchl'-~!
Besides the total synthesis of native active peptides and
their analogues, which play an important role in the eluci['I Prof. Dr. H.-D. Jakubke, Dr. P. Kuhl, Dr. A. Kdnnecke
Sektion Biowissenschaften der Karl-Marx-Universitat,
Wissenschaftsbereich Biochemie
Talstrasse 33, DDR-7010 Leipzig (Deutsche Demokratische Republik)
Angew. Cheni. I n / . Ed. Engl. 24 (1985) No. 2
dation of structure-activity relationships, the semisynthesis
of proteins[41as well as the chemical modification of peptides and proteins, in order to modulate physiological or
pharmacological effects, have been the focus of attention.
Ideally, the formation of the peptide bond should proceed fast and without racemization or other side reactions
and, using equimolar amounts of carboxy and amino components, should result in quantitative yields. This goal, however, is not reached by any of the more than 130 variations
of chemical peptide bond formation. Even the use of pro-
0 VCH Verlaqcgrsellcrhafi mhH. 0-6940 Weinheim, 1985
0570-0833/85/0202-0085 $ 02.50/0
ven synthetic concepts in combination with well-tested
coupling methods, which in model systems d o not result in
any racemization, does not guarantee the absolute avoidance of racemization when applied to the condensation of
peptide segments. Accordingly, in order to improve peptide synthesis, the avoidance of undesired side reactions
plays an important role. Based on such considerations, the
use of appropriate enzymes to form the peptide bond
springs to mind, since most enzymatic reactions are highly
stereospecific. Unfortunately, one cannot use the ribosomal peptidyl transferase to such an end, since it can only
act as an integral component of the ribosome in coordination with further factors that are involved in elongation of
the peptide
Based on considerations concerning chemical equilibria,
van’t Hoff already discussed in 1898 the possibility of using proteases, which normally catalyze the hydrolysis of
peptide bonds, to catalyze their formation[61. About 40
years later Bergmann and Fraenkel-Conrat presented experimental proof that the reversibility of protease-catalyzed reactions can be used to form peptide bond^^^.''.
Bergmann and his students, in particular Fruton, came to
the right conclusions about enzymatic peptide synthesis,
based on results o n the hydrolytic action of proteases as
well as their substrate specificity and stereospecificity. Enzymatic peptide synthesis, however, lost its importance
with the elucidation of the central principles of protein
synthesis in vivo. Apart from subsequent basic research, it
was only unambiguously shown in the mid-seventies that
proteases could be used successfully as biocatalysts in the
synthesis of peptides on a preparative scale. The use of
biocatalysts in peptide synthesis lies within the general
trend to employ enzymes for organic synthesis[91,and the
numerous reviews dealing with this
reflect the
wide interest in protease-catalyzed peptide synthesis over
the last 10 years.
It is in principle possible to influence the enzymatic formation of peptide bonds either thermodynamically or kinetically. Based o n energetic considerations, significant
differences emerge. Taking the basic principles of peptide
bond formation using proteases as catalysts into account,
an attempt will be made to illustrate systematically the
present possibilities and limitations of this novel synthetic
2. Thermodynamically Controlled Peptide Synthesis
The thermodynamically controlled formation of peptide
bonds represents the direct reversal of the catalytic cleavage of peptides by pro tease^"^-'^]. Since, however, concentrations are used in the description below instead of activities, this is not a n exact thermodynamic treatment. In contrast to the hydrolysis, the synthesis of a peptide bond is an
endergonic process, i.e. proceeds with loss of entropy and
is energetically so unfavorable that the equilibrium constant KIynfor the coupling of two unprotected amino acids
is < lo-’. Since the ionic forms of the two substrates are
unreactive, for the thermodynamic approach two equilibria have to be taken into account:
R - C O O ~+ H?~.-RJ
H,N-R, &
+ HzO
Preceding the “inversion equilibrium” K , , , between the
uncharged substrates and the product is an “ionization
equilibrium” K,,,. Taking the concentration of water into
the equilibrium constant, the total process is given by
K s ~ n= K l o n
Kin, = [R-CO-NH-R‘l
For any given pair of substrates and known pH, K,,,and
K , , , are fixed. Therefore, the peptide concentration for the
inversion equilibrium is dependent on the concentrations
of the uncharged substrates that are formed in the preceding ionization equilibrium. These concentrations in turn
are determined by their p K values. pK values of N- and Cprotected amino acids as well as those for longer segments
are substantially different from those of free amino acids;
therefore, sufficiently high concentrations of uncharged
substrates are present in the ionization equilibrium to give
measurable concentrations of peptide in the inversion
The only function of the protease is to accelerate the attainment of the equilibrium for formation of the peptide.
Its effectiveness is greater the more specific the protease is
for the substrates. Therefore, reaction conditions should be
chosen that ensure a high catalytic activity of the protease.
The pH optimum of the synthesis lies-apart from pepsincatalyzed couplings-in the pH range between the pK of
the a-carboxyl group and that of the amino group of the
substrates, i.e. normally between pH 6 and 7 . The time necessary to finally attain the inversion equilibrium is determined by the concentration and the catalytic activity of the
protease as well as the affinity of the protease for the substrates; this can be several minutes up to several days.
Thermodynamically controlled peptide synthesis can be
employed for all proteases in conjunction with carboxy
components unprotected at the C-terminus. However, if
nonspecific proteases are used to condense segments, undesired peptide cleavages may occur. Even for highly specific proteases, such side reactions can take place if the sequence of one segment contains any amino acids that correspond to the substrate specificity of the protease.
There are two principal ways by which one can further
influence thermodynamically controlled peptide formation:
-increasing K,,, by alteration of the p K values of the substrates, and
-increasing the concentration of the peptide product via
manipulations based on the law of mass action.
Both possibilities can be achieved in quite different
ways; in practice, however, one generally prefers to influence the equilibria in favor of synthesis by a combination of both factors. For didactic reasons the different possibilities of control are discussed separately below.
2.1. Increase in Kion
For a given set of reaction conditions there are two ways
to decrease the difference in pK values between amino and
Angew. Chem. int. Ed. Engl. 24 (1985) No. 2
carboxy components in order to increase K,,,.
in turn, to an increase in Ksyn= Kion.K,,,.
This leads,
2. I . 1. Use of Water-Miscible Organic
Water-miscible organic solvents decrease the acidity of
the a-carboxy group of the carboxy component, whereas
they only marginally influence the pK value of the amino
group of the nucleophile. For example, the pK value of
acetylglycine (Ac-Gly) in water is 3.60, while in 80% (v/v)
dimethylsulfoxide it is 6.93; the pK values of Gly-NH,,
8.20 and 8.10, respectively, remain almost constant. Since
pK,,, values measured in organic solvents have no great
relevance, pK values determined under the same conditions should bear greater significance. In water ApK for
acetylglycine/Gly-NH2 is 4.60,while in 80% (v/v) DMSO
it is only 1 . 1 1 . A variety of other cosolvents can be used to
change the pK value by one to two units, resulting in a significant increase in Ksyn.For the coupling of 1 mM Z-Trp
with 100 m M Gly-NH2 at p H 6.7 using chymotrypsin, Ksyn
is 0.45 M
in water, 2.12 M - ' in 60% (v/v) glycerol and
38 M - ' in 85% (v/v) 1,4-butanediol.
This approach, however, is problematic insofar as the
catalytic activity of proteases decreases with increasing
concentration of the cosolvent; thus, the time required to
reach the inversion equilibrium increases. Using 50% (v/v)
ethanol, dimethylformamide (DMF), DMSO, dioxane,
acetone, o r acetonitrile results in inactivation of chymotrypsin, and no synthesis occurs. Only polyalcohols, which
act as enzyme stabilizers, may be used at higher concentrations without inactivation of the enzyme["], but even in
these cases one finds a negative effect on the catalytic acti~ity[**-'~~.
It was shown that no significant alteration of
the substrate specificity occurs[zo1.
were based on the limited solubility of the products, which
thus are removed from the inversion equilibrium and accumulate["]. The product can also be removed from the
equilibrium by extraction or specific complexation.
2.2.1. Formation of Insoluble Products
If sufficiently high concentrations of substrates are employed, such that in the equilibrium Kin, it is possible to
have a peptide concentration that lies above the maximal
saturation concentration, precipitation occurs and thus the
product of the synthesis accumulates. This can be explained using the law of mass action (see also [231). If two
substrates A and B of equal concentration react to form a
product C, the equilibrium A + B + C exists in solution, the
equilibrium constant of which is given by Equation (1).
K = [C]([A].[B])-'
= [C].[A]-2
If part of the product C forms a precipitate, Equation (2)
where [C], is the maximal concentration of the product in
solution and [C], the concentration of the insoluble product with respect to the total volume.
Using the starting concentration [A],, the mass balance
results in Equation (3)
with [A] as the equilibrium concentration of the substrate
A in solution. Insertion of Equation (3) into Equation (2)
results in
2.1.2. Use of Biphasic Systems
In systems consisting of an aqueous phase and a nonmiscible phase (nonpolar organic solvent), pK values are
influenced in such a way that K,,, increasesr231.Depending
on the volume ratio a of organic and aqueous phase and
the partition coefficient P, an increase in the pK values of
the carboxy components and a decrease in the pK values
of the amino components by Ig(1 + a - P ) is observed.
Since the enzyme is localized in the aqueous phase, the
activity can be influenced only by the saturation concentration of the organic solvent in water, and the enzyme is
therefore inhibited far less than by solvents miscible with
~ a t e r ' ~ ~ .This
* ~ ' .advantage, however, is counteracted by
the prolonged time required to reach equilibrium, and additional partition equilibria are most likely the rate-determining steps. Solubility of the substrates in the nonpolar
organic phase limits the general use of biphasic systems for
the enzymatic peptide synthesis.
Substitution of [A]= 2
from Equation ( I )
[C]=[C],into the above equation gives, after rearrangement, Equation (5).
Thus, the apparent equilibrium constant for a reaction in
which some of the product precipitates due to its limited
solubility in the system is greater the higher the starting
concentration of substrates and the lower the solubility of
the product in this system,
Using one component in excess may result in almost
quantitative reaction of the other substrate. This method is
very popular in practice since the condition of low solubility of the product compared to that of the substrates frequently holds.
2.2. Influence on Product Formation
Based on the Law of Mass Action
2.2.2. Extraction of Products
The first positive experimental results for the potential
use of a reversal of protease-catalyzed peptide hydrolysis
In biphasic systems the product is removed from the
equilibrium if, due to a favorable position of the partition
Anfew Chem. h t . Ed EngJ. 24 (198.7) No. 2
equilibrium, it is extracted and thus accumulates in a nonpolar phase that is not miscible with water. Martinek et
al.l2'] derived the following equation for a reaction
A + B . + C + D [Eq.(6)].
mation (Fig. 1) before the slower hydrolysis of the product
starts to become important.
where a = Vorg/Vaq and P is the partition coefficient. They
showed that K b , p h a s (shows
a maximum for a defined ratio of partition coefficients. This means that the effective
equilibrium constant may be smaller or larger than that in
either of the two phases alone.
In most cases, however, the product is only marginally
soluble in the organic phase; it precipitates and is thus removed from the e q ~ i l i b r i u m [ ' ~ - ~This
~ l . can also be
achieved by the use of particular solvent combinations
which allow the adjustment of the saturation concentration
of one substrate, so that the product with the longer peptide chain is less soluble in that particular
2.2.3. Specific Complexation of the Product
If compounds are available that can form specific complexes with the product, the latter can be removed from the
equilibrium by adding such a complex-forming material
and the complex can be accumulated[291(see also Section
Fig. I . Characteristic reaction pathway of a kinetically controlled peptide
synthesis using an N-protected amino acid ester or peptide alkyl ester as acyl
The area in which kinetically controlled synthesis can be
used in practice is limited to serine and cysteine proteases
with preferably acylamino acid esters as the carboxy component. The reactions are characterized by short reaction
times and low enzyme requirements. In contrast to the
thermodynamically controlled equilibration, here the ester
substrate only occurs once as the acyl-enzyme intermediate, which is the decisive step for the ratio of peptide formation to hydrolytical cleavage.
3.1. Nucleophile Specificity of Proteases
3. Kinetically Controlled Syntheses
Investigation of the catalytic mechanism of serine and
cysteine proteases using chymotrypsin and papain as examples revealed that, in the presence of nucleophiles, acylenzyme intermediates R-CO-E are deacylated competitively by water and the nucleophile H2N-R"30-321. If
the nucleophile is an amino acid or peptide derivative,
then a new peptide bond is formed during the aminolytic deacylation (Scheme 1). If k2>k3+k4 and
k4[H2N-R'] > k3[H,O] = k;, then a kinetically controlled
accumulation of the peptide product occurs, with a concentration greater than the thermodynamic equilibrium
concentration. Especially suitable carboxy components for
this type of enzymatically catalyzed peptide synthesis are
acylamino acid alkyl esters if they match the substrate specificity of the particular protease, since in general they fulfill the condition k, > k3 k,. Furthermore, kz of the ester
substrate is substantially greater than k2 of the peptide
f o ~ m e d [ ~ ~which
- ~ ~ Iresults
in a maximum for product for-
- X-OH
Scheme I . Peptide synthesis catalyzed by serine and cysteine proteases
(H-E) with R-COOX as the acyl donor (carboxy component) and H2N-R'
as C-protected amino acid or peptide (amino component); X=alkyl.
Substrate binding sites in the active centers of proteases
exhibit differentiated specificity towards nucleophiles such
as amino acid and peptide derivatives. This influences primarily k , and thus the success of kinetically controlled
peptide synthesis[".".'h-4'l . A n acyl-enzyme breaks down
via hydrolysis and aminolysis with reaction velocities u3
and u,, respectively. From Scheme 1, Equation (7) is obtained.
This simplifies to Equation (8)
P _
with p = k$/k:. The partition constant p corresponds to the
d[R-COOH]= d[R-CO-NH-R'][401; i.e., partition of the
acyl-enzyme into peptide and acylamino acid occurs to
50% each. Table 1 contains some partition constants. These
values indicate the possible use of p as an efficiency parameter for various nucleophiles. They also show that p is
not a universal constant, since it is influenced by the acyl
part of the acyl-enzyme. Nevertheless, it is possible to deduce the order of magnitude of the concentration of nucleophile required for a particular peptide synthesis directly from the value ofp. Equation (8) is strictly valid only
Angen'. Chem. Int. Ed. Engl. 24 (19851 N o . 2
Table 1 . Partition constants p of some acylchyrnotr>jhiils [a]
Ac-Phe-CT [b]
p [rnol.L-'.lO-']
Ac-Tyr-CT [cJ
Mal-Phe-CT 1401
[a] CT= chymotrypsin; Ma1 = maleyl-(3-~arboxyacryloyl-).[b] 6-CT; calculated according to [32]. [cl a-CT; calculated according to 1391.
for the case where one finds a linear relationship between
the concentration of nucleophile and the partition in the
range of interest. This is true in the majority of cases investigated. However, deviations from linearity have been
The mathematical treatment of these cases is
rather complicated since p becomes a function of the concentration of nucleophile.
3.2. Influence of the Reaction Medium
Since only the non-protonated form of the nucleophile
reacts in the aminolytical deacylation of the acyl-enzyme,
it is necessary to take the effective concentration of nucleophile instead of the total concentration into account. This
can be calculated from the pK value of the nucleophile
and the pH of the medium. Since pK values of a-amino
groups of amino acid and peptide derivatives lie around 8,
it is advisable to carry out kinetically controlled peptide
synthesis at p H L 8 . The partition constant p is also dependent on pH. For the proteases trypsin and chymotrypsin p is reduced in alkaline medium, which results in increased synthesisf3". Organic solvents and ionic substances
as well as the reaction temperature can influence the partition constant p .
4. Selected Examples
Compared to the well-established chemical coupling
reactions, a synthetic strategy based on proteases as catalysts is naturally still in its infancy. The majority of studies
with respect to the usefulness of proteolytic enzymes for
the formation of peptide bonds were initially model studies. In general, therefore, the methodological aspects were
more important than any practical applications and substrates were chosen in such a way as to correspond to the
specificity of the proteases employed. Recently, however,
examples of the syntheses of biologically active peptides
using coupling steps exclusively or partially catalyzed by
proteases have increased. The area of semisynthesis of
proteins has seen considerable progress based on the use
of p r o t e a s e ~ [ ~ . ~ * I .
The following selected examples illustrate the strategy
and tactics as well as the present possibilities for the use of
proteolytic enzymes in the synthesis of biologically active
peptides and proteins.
4.1. Aspartame
The precursor (Z-Asp-Phe-OMe) for the peptide sweetener aspartame (Asp-Phe-OMe) is obtained by a thermoAnyew. Chem. In!. Ed. Enql. 24 (1988) No. 2
lysin-catalyzed coupling of Z-Asp with Phe-OMe in a thermodynamically controlled reaction. An insoluble salt is
formed with the enantiomer of the amino component, ZAsp-Phe-OMe. Phe-OMe, which results in an almost quantitative shift of the equilibrium towards the
this reaction the strategy of minimal protection is fortuitously complemented by the salt formation. Enzymatic
synthesis is thus superior to chemical methods since the
latter require the use of @-protected aspartate derivatives.
4.2. Leucine-Enkephalin
The kinetically controlled synthesis of Leu-enkephalin,
Tyr-Gly-Gly-Phe-Leu, was achieved by stepwise elongation
of the C-terminal end of the peptide via catalysis with the
serine exoprotease carboxypeptidase Y (CPD-Y)[441.Amino acid amides were used as nucleophiles since free amino acids produce only low yields and amino acid esters
give rise to side reactions which are difficult to control[451.
Thus, the seemingly complicated route of selectively removing the C-terminal amide group with CPD-Y followed
by esterification of the peptide had to be resorted to before
it was possible to use the peptide in the next coupling step
as the substrate. The protecting group for the a-amino
function was Bz-Arg, which can easily be removed by tryptic ~ I e a v a g e [ ' ~ . ~ ~ l .
4.3. Dynorphin(1-8)
Dynorphin(l-8), Tyr-Gly-Gly-Phe-Leu-Arg-Arg-Ile,
Leu-enkephalin that is extended at the C-terminal end, was
synthesized by a combination of thermodynamicafiy and
kinetically controlled couplings with papain, chymotrypsin, and trypsin as bio~atalysts'~'~.
To this end, four dipeptides were synthesized enzymatically and subsequently
condensed to yield two tetrapeptide segments. Coupling of
the two tetrapeptides Boc-Tyr(Bz1)-Gly-Gly-Phe-OEt
Leu-Arg-Arg-Ile-NH-NH-ChH, with chymotrypsin results
in a partially protected octapeptide, which after deblocking yields the biologically active dynorphin( 1-8) (Fig. 2).
The protecting group for the carboxy function is in general
phenylhydrazide, which can be introduced via papain catalysis into Boc-amino
and can easily be removed
by oxidation or can be converted into an ester1491.
Fig. 2. Synthesis of dynorphin(1-8) catalyzed by proteases [47]; CT=chymotrypsin, P = papain: T= trypsin; EzI = henzyl, Boc=terzbutoxycarbonyl.
This example demonstrates two major advantages of enzymatic peptide synthesis : condensation of segments containing a variety of amino acids can be achieved without
the risk of racemization and it is not necessary to protect
the guanidino group of arginine.
The C-terminal dipeptide could also be obtained by a
thermolysin-catalyzed reaction in which the equilibrium of
the thermodynamically controlled condensation was
shifted in the direction of products by extraction of the
product Z-Leu-Met-NH, into the organic phase.
4.4. BzlGlyZ’(Acm)Cys3’1EGF(21-31)NHEt
4.6. Human Insulin
The advantages of the combined use of chemical and enzymatic methods are illustrated by the synthesis of segments of the mouse epidermal growth factor (EGF)[”]. All
couplings catalyzed by proteases were carried out under
kinetic control (Fig. 3) since this ensures a short reaction
time, even when low concentrations of enzyme are used.
Enzyme-labile protecting groups were Bz-Arg and BzPhe[17,461,
which were also used to regulate the hydrophobicity and hydrophilicity of the partial sequences.
Using thermodynamically controlled reactions it is possible in a variety of ways to obtain semisynthetic human
insulin and analogues from porcine insulin, which is available from natural sources (Fig. 5). The control of the thermodynamic equilibrium is always achieved by high concentrations of the appropriate nucleophilic amino component and frequently also by a large volume of water-miscible organic cosolvents. The simplest way to obtain the human insulin ester is in a one-step reaction involving an enzyme-catalyzed exchange of the C-terminal alanine-B30 of
porcine insulin for a threonine e ~ t e r [ ~ ~ - Using
’ ’ ~ . the above
reaction conditions prevents almost completely the possible cleavage of the ArgBZZ-GlyBZ3
bond by trypsin. Such
methods are now used commercially for the production of
human insulin.
Fig. 3. Synthesis of Bz [Gly”(Acm)Cys”] EGF(2I-31)NHEt [SO]; CP=chymopapain, V8 = Staphylococcusaureus protease V8, Y = carboxypeptidase Y,
DCC = dicyclohexylcarbodiimide; Bz = benzoyl, Su = succinimido, Acm =
4.5. Eledoisin (6-ll)-Hexapeptide
H 1 ~ A r g 2 2 - L y s z g
This hexapeptide was obtained by kinetically controlled
couplings using papain and chymotrypsin in the biphasic
system CCl,/buffer[”I (Fig. 4). The biphasic system was
employed not primarily to influence the equilibrium, but
to insure a high catalytic activity of the protease dissolved
in the buffer as well as excellent solubility of the substrates
in the organic phase. Thus, the total reaction volume could
be kept small, ensuring yields that are higher than can be
achieved using water-miscible c o ~ o l v e n t s [ ~ ~ ~ .
Fig. 5. Semisynthetic conversion of porcine insulin into human insulin;
PI = porcine insulin, HI =human insulin; DAI =des-Ala3”-insulin,
DO1 =deo~tapeptide”~~~~’~-insulin,
A =Achromobacter protease, CPA= carboxypeptidase A.
One can also use des-AlaB3”-[54.’8.591
or deoctapeptide B23-B30-insulin[21.601 as substrates which can be coupled to
synthetic peptides and thus provide an easy approach to
Apart from trypsin, a
the synthesis of insulin
lysine-specific Achrornobacter p r o t e a ~ e [ ~ ~ .and
~ ~ . CPD”~
Yrssl have been used. The advantages of the enzymatic
semisynthesis over chemical methods are obvious for the
above mentioned examples since nearly all problems connected with blocking or deblocking are avoided using proteases.
4.7. Ribonuclease S-Peptide(1-15)
Fig. 4. Synthesis of eledoisin (6-1 I)-hexapeptide [51]; MA=mixed anhydride method; Z = benzyloxycarbonyl.
One example for the specific complexation of the peptide formed, which is thus removed from the synthesis
equilibrium, is the coupling of the synthetic ribonuclease
S-peptide segments 1- 10 and 11- 15 catalyzed by clostripain. The resulting peptide 1 - 15 is complexed by native
ribonuclease(2 1-24) S - ~ r o t e i n [ ~ ~ !
Angew. Chem. Int. Ed. Engl. 24 (1985) No. 2
S. Use of Immobilized Proteases
For future applications of the enzymatic peptide synthesis, the use of immobilized proteases will probably be of
technological and economical importance. Basic studies
on the applicability of immobilized proteases have been
carried out under kinetic as well as thermodynamic control
using covalently bound chymotrypsin, trypsin, thermolysin, and
It was shown that the immobilized
biocatalysts have almost the same efficiency as the native
enzymes. Contrary to earlier
it could also be
shown that immobilized papain is ideally suited for kinetically controlled synthesis at high p H values[371.
The effort involved in immobilizing an enzyme is compensated for by the possibility of its repeated use. Depending on the stability of the immobilized protease, this is guaranteed if short reaction times are employed, which is a
characteristic of kinetically controlled reactions. In model
experiments the possibility of repeated use under different
reaction conditions could be d e m o n ~ t r a t e d [ ~ ~. .A~ *fur-~~]
ther notable advantage of immobilized enzymes is the fact
that the products are not contaminated with proteolytically
active or denatured enzymes. Furthermore, the possibility
of a continuous flow application may become a reality.
Of practical importance so far is the synthesis of the
aspartame precursor using ionically bound thermolysin in
a mixture of greater than 99% (v/v) ethyl acetate; this,
however, requires rather long reaction times[651.
The thermodynamically controlled coupling of desAlaB30-insulinwith Thr-OtBu in the presence of high concentrations of water-miscible organic solvents is also effectively catalyzed by covalently bound Achrornobacter protease[661or tryp~in[~’!
6. Criteria for Synthesis Planning
In contrast to the synthesis of di- and tripeptides, it is
necessary for the assembly of longer chains of peptide sequences to carry out an exact synthetic plan, the specific
method of coupling being at first of little importance. The
strategy of a peptide synthesis normally means the order in
which the particular amino acid residues are coupled,
namely either via stepwise addition from the C- or the Nterminal end or via condensations of presynthesized segments (segment condensation). A stepwise chemical synthesis starting from the N-terminal end, analogous to the
ribosomal protein biosynthesis, is prohibited due to the
permanent risk of racemization. Apart from the severe solubility problems in the synthesis of polypeptides with
complete side-chain protection, it is mainly the risk of racemization that leads to the enormous variety of synthetic
strategies for the chemical synthesis of peptides.
Although enzyme-catalyzed coupling reactions take
place stereospecifically (free of racemization) and require
only minimal side-chain protection, thereby solving the
aforementioned problems, the substrate specificity is a limitation, which prevents this method from being universally
applicable. A protease is much more limited in its area of
use compared to chemical coupling methods, which are
only influenced at most by the steric factors of the compoAngew. Chem. Inf. Ed. Enyl. 24 11985) No. 2
nents to be coupled. This disadvantage, however, is compensated for completely by other advantages. These include not only the aforementioned factors (the stereospecific course of the reaction and the dispensing with the
protection of side chains of trifunctional amino acids) but
also mild and ecologically advantageous reaction conditions. If the potential of proteolytic enzymes that are to be
used principally for the formation of peptide bonds is examined, it is seen that exopeptidases should find preferential application to the stepwise synthesis of short peptides
and segments, while condensations to larger blocks should
best be carried out by catalysis using endopeptidases of the
required specificity. Exopeptidases have the great advantage that, in starting a stepwise synthesis either from the Cterminal end (aminopeptidases) or from the N-terminal
end (carboxypeptidases), the internal peptide bonds within
the growing chain can no longer be proteolytically cleaved
by the enzyme employed as the biocatalyst. For preparative purposes aminopeptidases seem less well suited, since
both substrate and product contain free a-amino functions
which complicate the isolation of the product. Furthermore, since the substrate specificity is not strongly determined, there is a possibility of undesired coupling reactions. Carboxypeptidases exhibit superior properties as
biocatalysts for stepwise coupling reactions, especially the
serine protease carboxypeptidase Y (see also Section 4.2).
Since, however, even the application of CPD-Y shows
some preparative limitations, it is advantageous to have
endopeptidases available for the synthesis of short peptides, especially since such enzymes can even be used for
the synthesis of dipeptides if appropriate combinations of
protecting groups are employed. Furthermore, one can use
endopeptidases with broader specificity for stepwise synthesis as was shown for oligopeptide synthesis catalyzed
by papain[681.
Problem-free coupling of segments require endopeptidases with high substrate specificity and the absence of the
specific amino acids in the segments, except for the position where the peptide bond is to be formed. Apart from
the lysine-specific Achromobacter protease, trypsin catalyzes the formation of a peptide bond after lysine and arginine residues. Thus, coupling of segments can only be
carried out under the conditions described above for carboxy components that carry one of the mentioned basic
amino acids in the C-terminal position. This restriction
necessarily leads to the conclusion that one may employ
less-specific proteases to catalyze coupling of segments. In
order to avoid undesirable chain cleavage, manipulations
that favor the synthesis have to be carried out, as described
in Sections 2 and 3. From the literature, examples are
known for the synthesis of biologically active oligopeptides using exclusively protease-catalyzed coupling steps
(see also Sections 4.2 and 4.3). However, without intending
to play down this ambitious synthetic concept, it seems unlikely that in the near future chemical coupling methods
will be generally replaced by the enzymatic approach. Instead, the enzyme-catalyzed couplings will contribute significantly to the methods available to the peptide chemist.
In a given synthesis project it must first be determined if,
for the separation into segments, favorable combinations
of amino acids can be found that permit protease-catalyzed coupling of segments. The synthesis of the segments
should preferably be carried out in combination with
proven chemical coupling reactions. Therefore, simple
chemical coupling reactions without the risk of racemization are preferred by all means over enzymatic reactions,
especially for segments containing a large number of amino acids that d o not need further protecting groups. On
the other hand, for segments with predominantly trifunctional amino acid residues where semipermanent side
chain protection is not desired, enzyme-catalyzed reactions
using proteases of the appropriate substrate specificity are
preferable. In this way, segments are obtained with sufficient solubility to be used in protease-catalyzed block couplings.
For the use of serine and cysteine proteases a further
point needs to be decided; namely, should the carboxy
component be used as the acylamino acid or should a derivative (e.g. alkyl ester) that would favor acylation be
used. Although the kinetically controlled reaction with little enzyme requirement and short reaction time should be
preferable, the decision in each case will be determined by
the overriding total synthetic concept. An unfavorable nucleophile specificity may be better taken care of in a thermodynamically controlled reaction with the appropriate
manipulations of conditions than in a kinetically controlled coupling, where requirements are less flexible.
For the synthesis of biologically active peptides or segments with predominantly hydrophobic amino acid residues it is frequently necessary to overcome the poor solubility of the reactants through the introduction of solubilizing protecting groups[691.The introduction of such solubilizing residues into the classical Nu- and C”-protecting
groups cannot be regarded as the solution to this problem;
for enzyme-catalyzed peptide synthesis totally different
protecting groups are required. Even polyethylene glycol
esters, which have proved useful for chemical synthesis in
aqueous solution‘701, are less useful as carboxy components, whereas amino components blocked in this way can
be reacted without any problem~~’’~.
In general, protecting
groups that can be removed under mild conditions are required, as is possible, for example, in the ease of Bz-Arg or
Bz-Phe as a-amino protecting groups which can easily be
cleaved off by trypsin or ~ h y m o t r y p s i n l ” ~ ~Solubilizing
ester groups attached to the carboxy component increase
the solubility of the respective reactant in kinetically controlled reactions. After the coupling, however, this solubilizing effect is naturally lost. It is possibly preferable to use
non-ionic solubilizing ester groups rather than charged residues.
The improvement of the solubility of reactants and
products is of immediate importance for a continuous flow
synthesis using immobilized proteases. Their application
seems justified if the effort required for the immobilization
is compensated for by the repeated use of the biocatalyst.
Finally, the question arises whether the use of proteases
as catalysts can be of advantage for peptide bond formation via a polymer-based route. Simple model studies using
a polymer-bound amino component show that polyethylene glycol is well suited as a soluble polymer[”]. Silica gel
as the insoluble polymer matrix seems less promising since
the fixed nucleophile is limited in its mobility and because
of the possibility of steric hindrance between the enzyme
and the polymer[731.By the use of appropriate spacers as
well as polymers with a high loading capacity for nucleophiles, these limitations should be successfully surmountable.
7. Summary and Prospects
Based on the classical studies of the reversibility of reactions catalyzed by proteases, it has been possible in the last
ten years to show convincingly that proteolytic enzymes
can in principle be used as biocatalysts. Thus, they can be
employed in peptide bond formation on a preparative
scale and are of practical use in the biotechnological synthesis of biologically active peptides and proteins. The
stereospecificity of proteases guarantees the formation of
stereochemically uniform products and requires only minimal semipermanent protection of side-chain functions in
trifunctional amino acid residues. However, due to the
substrate specificity, the universal applicability of proteases is limited. Therefore, it is of great importance to select
proteases with modified substrate specificity based on strategic and tactical considerations with respect to the various
product sequences. The advantages of both the thermodynamically and the kinetically controlled reactions have to
be optimally exploited. Environmentally acceptable reaction conditions, which are characteristic of biotechnological processes, together with the potential for continuous
flow application using immobilized proteases, should be
an important stimulus to develop and perfect the methodology of enzyme-catalyzed peptide synthesis in order to improve and augment the armamentarium of the synthetic
peptide chemist.
Received: August 21, 1984 [A 5191
German version: Angew. Chem 97 (1985) 79
[I] E. Gross, J. Meienhofer: 7’he Peptrdes: Analysis, Synthesis. Biology, Vol.
1-6. Academic Press, New York 1979-1984.
[2] E. Wiinsch: Synthese uon Peptiden. in Houben-Weyl-Miiller: Methoden
der organischen Chemie. Ed. 15. Teil I and 2, Thieme, Stuttgart 1974.
131 H.-D. Jakubke, H. Jeschkeit: Aminosiiuren, Peptide, Proteine. Verlag
Chemie, Weinheim 1982.
[4] R. E. Offord in K. BIPha, P. Malori: Peptides 1982. de Gruyter, Berlin
1983, p. 31.
[5] H:G. Gassen, Angew. Chem. 94 (1982) 15; Angew. Chem. I n [ . Ed. Engl.
21 (1982) 23.
[6] J. H. van? Hoff, Z. Anorg. Chem. 18 (1898) 1.
171 M. Bergmann, H. Fraenkel-Conrat, 1. Biol. Cheni. 119 (1937) 707.
[S] M. Bergmann, H. Fraenkel-Conrat, J. Biol. Chem. I24 (1938) 1.
191 J. B. Jones, Methods Enzymol. 44 (1976) 831; C. J. Sih, C . 4 . Chen, Angew. Chem. 96 (1984) 556; Angew. Chem. I n t . Ed. Engl. 23 (1984) 570.
[lo] H:D. Jakubke, P. Kuhl, Pharmazie 3 7 (1982) 97.
1111 J. S. Fruton in A. Meister: Adu. Enzymol. Relat. Areas Mol. Biol. 53
(1982) 239.
1121 I . M. Chaiken, A. Komoriya, M. Ohno, F. Widmer, Appl. Biochem. Biotechno/. 7 (1982) 385.
[13] D. D. Petkov, J . Theor. Biol. 98 (1982) 419.
[I41 Y. Isowa, Yuki Gosei Kagaku Kyokai Shi 36 (1978) 195: Chem. Abstr. 89
(1978) 18465.
[IS] F. Brtnik, K. Jost, Chem. Listy 74 (1980) 951.
[I61 K. Morihara, Tanpakushitsu Kakusan Koso 26 (1981) 1979; 29(1984) 49;
Chem. Abstr. 95 (1981) 216416 and 100 (1984) 81610, respectively.
1171 J . D. Glass, Enzyme Microb. Techno/. 3 (1981) 2.
[18] D. Konopinska, F. Muzalewski, Mol. Cell. Biochem. 51 (1983) 165.
[I91 Z. Wang, Shengwu Huaxue Yu Shengwu Wuli Jinzhan 54 (1983) 5 3 ;
Chem. Abstr. 100 (1984) 82 164.
Angew Chem Int. Ed. Engl. 24 (19851 No. 2
[20] G. A. Homandberg, J. A. Mattis, M. Laskowski, Jr., Biochemistrv 17
(1978) 5220.
[21] K. Inouye, K. Watanabe, Y. Tochino, M. Kobayashi, Y. Shigeta, Biopolymers 20 (1981) 1845.
[22] A. Konnecke, V. Schellenberger, H.-D. Jakubke, Z. Chem. 24 (1984)
185: A. Konnecke, M. Hansler, H.-D. Jakubke, unpublished.
[23] K. Martinek, A. N. Semenov, J. Appl. Biochem. 3 (1981) 93; Usp. Khim.
50 (1981) 1376.
[24] P. Kuhl, A. Wilsdorf, H:D. Jakubke, Monatsh. Chem. 114 (1983) 571.
[25] P. Kuhl, J. Walpuski, H.-D. Jakubke, Pharmazie 37 (1982) 766.
(261 P. Kuhl, H.-D. Jakubke, Z . Chem. 22 (1982) 407.
[27] H.-D. Jakubke, P. Kuhl, A. Konnecke, G. Doring, J. Walpuski, A. Wilsdorf, N. P. Zapevalova in K. BIaha, P. Malofi: Peptides 1982. d e Gruyter,
Berlin 1983, p. 43.
[28] G. Doring, P. Kuhl. H.-D. Jakubke, Monatsh. Chem. 112 (1981) 1165.
I291 G. A. Homandberg, A. Komoriya, 1. M. Chaiken, Biochemistry 21 (1982)
I301 M. L. Bender, G. E. Clement, C. R. Gunter, F. J. Kezdy, J. A m . Chem.
SOC.86 (1964) 3697.
(311 L. J. Brubacher, M. L. Bender, Biochern. Biophys. Res. Commun. 27
(1967) 176.
[32] J. Fastrez, A. R. Fersht, Biochemistry 12 (1973) 2025; A. R. Fersht, D. M.
Blow, J. Fastrez, ihid. 12 (1973) 2035.
[33] J. R. Whitaker, M. L. Bender, J. Am. Chem. SOC.87 (1965) 2728.
134) M. L. Bender, F. J. Kezdy, Annu. Rev. Biochem. 34 (1965) 49.
[35] B. Asboth, L. Polgar, Acta Biochim. Biophys. Acad. Sci. Hung. 12 (1977)
[36] S . A. Bizzozero, B. A. Rovagnati, H. Dutler, Helu. Chim. Acta 65 (1982)
[37] V. Schellenberger, A. Konnecke, H.-D. Jakubke in U. Ragnarsson: Peptides 1984, Almquist & Wiksell Stockholm 1984, p. 201.
[38] W. Kullmann, Biochem. J . 220 (1984) 405.
[39] L. Riechmann, V. Kasche, Biochem. Biophys. Res. Commun. 120 (1984)
[40] A. Konnecke, V. Schellenberger, H.-J. Hofmann, H.-D. Jakubke, Pharmazie 19 (1984) 785.
[41] D. D. Petkov, I. Stoineva, Biochem. Biophys. Res. Cornmun. 118 (1984)
[42] H. Zahn, V. K. Naithani, H.-G. Gattner, E. E. Biillesbach, P. M. Tharnm,
Naturwissenschafen 68 (1981) 56; D. J. Saunders, D. Brandenburg, C.
Shang-chuan, W. Chih-chen, Endeavour 6 (1982) 146.
[431 Y. Isowa, M. Ohmori, T. Ichikawa, K. Mori, Y. Nonaka, K. Kihara, K.
Oyama, Tetrahedron Lett. 1979. 261 1.
[441 F. Widmer, K. Breddam, J. T. Johansen in K. Brunfeldt: Peptides 1980,
Scriptor, Kopenhagen 1981, p. 46.
I451 K. Breddam, F. Widmer, J. T. Johansen, Carlsherg Res. Commun. 48
(1983) 231.
[46] C. Meyers, J. D. Glass, Proc. Natl. Acad. Sci. USA 72 (1975) 2196.
Angew. Chem. Int. Ed. Engl. 24 (1985) No. 2
W. Kullmann, J. Urg. Chem. 47 (1982) 5300.
H. B. Milne, F. H. Carpenter, J. Org. Chem. 33 (1968) 4476.
H. B. Milne, W. Kilday, J. Org. Chem. 30 (1965) 64.
F. Widmer, S. Bayne, G. Houen, B. A. Moss, R. D. Rigby, R. G. Whittaker, J. T. Johansen in U. Ragnarsson: Peptides 1984, Almquist & Wiksell Stockholm 1984, p. 193.
1511 P. Kuhl, G. Doring, K. Neubert, H.-D. Jakubke, Monatsh. Chem. 115
(1984) 423.
[52] P. Kuhl, A. Konnecke, G. Doring, H. Daumer, H.-D. Jakubke, Terrahedron Lett. 1980. 893.
[53] A. Jonczyk, H.-G. Gattner, Hoppe-SeylerS 2. Physiol. Chem. 362 (1981)
[54] K. Morihara, T. Oka in S. Sakakibara: Pepride Chemistry 1982, Protein
Research Foundation, Osaka 1983, p. 231.
1551 K. Breddam, F. Widmer, J. T. Johansen, Carlsberg Res. Commun. 46
(1981) 361.
[56] K. Rose, J. Gladstone, R. E. Offord, Biochem. J. 220 (1984) 189, and ref-
erences cited therein.
[57] R. Obermeier, G. Seipke in W. Voelter, E. Bayer, Y. A. Ovchinnikov, E.
Wiinsch: Chemistry of Peptides and Proteins, Vo1. 2. d e Gruyter, Berlin
1984, p. 3.
[58] K. Morihara, T. Oka, H. Tsuzuki, Nature (London) 280 (1979) 412.
[59] K. Morihara, T. Oka, H. Tsuzuki, Y. Toshiharu, T. Kanaga, Biochem.
Biophys. Res. Commun. 92 (1980) 396.
1601 K. Inouye, K. Watanabe, K. Morihara, Y. Tochino, T. Kanaya, J. Am.
Chem. Soc. 101 (1979) 75 I.
[61] Y . 4 . Zhang, Trends Biochem. Sci. 8 (1983) 16.
I621 A. Konnecke, R. Bullerjahn, H.-D. Jakubke, Monatsh. Chem. 112 (1981)
[63] H.-D. Jakubke, R. Bullerjahn, M. Hansler, A. Konnecke in T. C. J. Grib-
nau, J. Visser, R. J. F. Nivard: Analytical Chemistry Symposia Series. Vol.
9, Elsevier, Amsterdam 1982, p. 529.
1641 A. Konnecke, M. Hansler, V. Schellenberger, H.-D. Jakubke, Monatsh.
Chem. 114 (1983) 433.
[651 K. Oyama, S. Nishimura, Y. Nonaka, K. Kihara, T. Hashimoto, J. Org.
Chem. 46 (1981) 5241.
[661 R. Muneyuki, T. Oka, K. Morihara in T. Shioiri: Peptide Chemistry 1981.
Protein Research Foundation, Osaka 1982, p. 113.
[671 H.-D. Jakubke, A. Konnecke, Method.$ Enzymol.. in press.
(681 Y. V. Mitin, N. P. Zapevalova, E. Y. Gorbunovd, In/.J. Pept. Protein
Res. 23 (1984) 528.
[691 P. Kuhl, J. Walpuski, H:D. Jakubke, Pharmazie 39 (1984) 280.
1701 G. P. Royer, G. M. Anantharamaiah, J. Am. Chem. Soc. 101 (1979)
[711 A. Konnecke, V. Pchalek, H.-D. Jakubke, Monatsh. Chem. 116 (1985) 111.
[721 R. W. Holley, J. A m . Chem. Soc. 77(1955) 2552.
[731 A. Konnecke, S. Dettlaff, H.-D. Jakubke, Monatsh. Chem. 113 (1982)
Без категории
Размер файла
928 Кб
bond, basic, formation, protease, peptide, principles, catalyzed
Пожаловаться на содержимое документа