Protease-catalyzed peptide synthesis on solid support

Article abstract of DOI:10.1021/ja026912d

The direct enzymatic synthesis of peptides from amino acids is widely used as a useful alternative to chemical synthesis. However, good yields of such enzyme-catalyzed reactions require altered reaction conditions to overcome the bias for hydrolysis in aqueous medium. We argue that the synthesis/hydrolysis equilibrium can be shifted toward synthesis in aqueous medium by immobilizing the amine on solid support. In this report, we show the first examples of solid-phase peptide synthesis catalyzed by a protease in bulk aqueous buffer. Copyright

Full text of DOI:10.1021/ja026912d

Published on Web 08/23/2002
Protease-Catalyzed Peptide Synthesis on Solid Support
†
†
‡
,†
Rein V. Ulijn, Beatriz Baraga n˜ a, Peter J. Halling, and Sabine L. Flitsch*
Department of Chemistry, The UniVersity of Edinburgh, King’s Buildings, West Mains Road,
Edinburgh EH9 3JJ, Scotland, and Department of Chemistry, The UniVersity of Strathclyde, 295 Cathedral Street,
Glasgow G1 1XL, Scotland
Received May 15, 2002
The direct enzymatic synthesis of peptides from amino acids is
widely used as a useful alternative to chemical synthesis. However,
good yields of such enzyme-catalyzed reactions require altered
reaction conditions to overcome the preference for hydrolysis in
Scheme 1. Amide Synthesis/Hydrolysis on Solid Supported
Substrate
Table 1. Thermolysin-Catalyzed Synthesis of a Range of
Dipeptides on PEGA1900
1
aqueous medium. For example, organic (co-) solvents and highly
2
concentrated substrate suspensions have been used to shift the
3
equilibrium toward synthesis. Both of these methods have their
drawbacks and limit the scope of the biotransformation. In this
report, we demonstrate that the equilibrium can also be shifted
toward synthesis in aqueous medium by immobilizing the amine
on solid support (Scheme 1). Thus, we report the first examples of
protease-catalyzed high-yielding peptide synthesis on solid support
in bulk aqueous buffer.
acyl
conversion
donora
productb
X
R
(%)b
1
2a
2b
2c
3a
3b
3c
3d
3e
3c
3f
3g
3h
3i
Fmoc
Fmoc
Fmoc
Cbz
Fmoc
Fmoc
Fmoc
Fmoc
Fmoc
Fmoc
Fmoc
H
99
99
99
72
0
It is well known that the synthesis of amides from free amines
and carboxylic acids is thermodynamically highly unfavorable in
dilute aqueous media. Various “low-water” reaction media have
been developed to shift this reaction equilibrium toward synthesis.
The most intensively studied approach consists of (partly) replacing
CH2-CH-(CH3)2
CH2-C6H5
2d
2e
CH -C H
5
2
6
D-CH2-C6H5
D/L-CH2-C6H5
CH2-CH2-CONH2
CH2-OH
c
2
2
2
2
c,2e
f
g
h
99
84
10
77
70
99
1
water with organic solvents. This approach offers several advan-
tages, and many applications of protease-catalyzed peptide synthesis
in such reaction media have been reported. A drawback is that
enzyme activity is often significantly lower and substrate solubility
can be limiting. Recently, an additional type of “low-water” reaction
medium has been used which gave good results in peptide synthesis
when the reaction product precipitated from aqueous substrate
+
CH2-(imidazole)H
-
2i
CH2-COO
2j
3j
CH2-(CH2)2-CH3
a
Unless specified, L amino acids were used. b Yields determined by
HPLC. c Only the L,L-dipeptide was detected by HPLC and LC-MS. A
mixture of both diastereoisomers prepared chemically was used as the
standard.
2
suspensions. A drawback of these “solid-to-solid” reactions is that
product precipitation (and high yields) is not thermodynamically
favored for all reactions.
8
overall positive charge of the resin. A final effect may result from
2c
the improved solvation of hydrophobic acyl donors in PEGA resin
when compared to aqueous environment. This increase will result
in a local higher concentration of acyl donor near the site of
catalysis. For these reasons, we believed solid-phase peptide
synthesis via the direct reversal of the hydrolytic reaction would
be feasible, and we put this hypothesis to the test.
Thermolysin was chosen as a suitable biocatalyst, because it is
known to be robust, and it is a commonly used enzyme in
There is a considerable current interest in using enzymes on solid
9
4
supported substrates in screening of combinatorial libraries, chemo-
5
6
enzymatic synthesis, and enzyme cleavable linkers. Meldal and
co-workers have developed special supports, such as PEGA (poly-
10
(
ethylene glycol)-acrylamide), that have been shown to be useful
7
for these applications because they are accessible to small enzymes.
For example, proteases have been employed to hydrolyze solid
supported peptides for identification of their substrate specificity
and in screening for inhibitors.7 The feasibility of employing
proteases for solid-phase peptide synthesis has, however, not been
reported before.
11
thermodynamically controlled peptide synthesis. Because of its
broad specificity for the carboxylic acid substrate, thermolysin can
be employed to catalyze a variety of peptide synthesis reactions
with acyl donors with different polarities and functionalities.
Thus, Fmoc-phenylalanine was coupled to PEGA1900 (Polymer
Laboratories, UK) via the Wang linker, and the amino group was
then deprotected with piperidine. Resin bound phenylalanine (1)
was then treated with excess carboxylic acids (2a-2j) in the
presence of thermolysin. Reaction products (3a-3j) were released
from the resin by acid cleavage and quantified by HPLC. Table 1
summarizes the results. With hydrophobic acyl donors (entries 2a-
d, 2j), very good yields were observed. These results demonstrated
that the shift in equilibrium for amide formation on solid phase
can be sufficient to yield complete conversion to the amide.
There are several reasons why a shift in equilibrium toward
synthesis can be expected when the amine substrate is linked to a
solid support. One widely recognized advantage of solid-phase
synthesis is that large excesses of substrates can be used to help
drive reactions to completion. A second contribution is expected
from suppressed ionization of the solid supported amine due to the
*
To whom correspondence should be addressed. E-mail: S.Flitsch@ed.ac.uk.
†
The University of Edinburgh. Fax: +44 (0)131 650 4737. Tel: +44 (0)131
6
50 4737.
‡
The University of Strathclyde. Fax: +44 (0)141 552 4822. Tel: +44 (0)141
5
48 2683.
10988
9
J. AM. CHEM. SOC. 2002, 124, 10988-10989
10.1021/ja026912d CCC: $22.00 © 2002 American Chemical Society

C O MMU N I C A T I O N S
The solubility of the Fmoc amino acids is very low in aqueous
buffer (generally <1 mM), but this did not limit the yields observed.
Indeed, better yields were observed when the acyl donors were more
hydrophobic. For example, with Cbz-Phe (2d), the observed yield
phosphate buffer of pH 7.5. Reactions were briefly mixed and
subsequently incubated overnight at room temperature on a blood
rotator. The next day, the resin was washed extensively using 5
mL volumes in the following sequence: 5× DMF, 5 × 50:50 (v/
v) DMF:MeOH, 5× MeOH, 5 × 50:50 (v/v) acetonitrile/water,
5× MeOH. The products were cleaved from the resin with 2 mL
of TFA:water 95:5 during 2 h. Resin was then washed with 10 mL
of a mixture of 50:50 acetonitrile in water with 0.1% TFA, solvent
was evaporated off, and the residue was redissolved in 1 mL of a
50:50 mixture of acetonitrile and water. The samples were analyzed
by HPLC (Waters 2690 LC system equipped with a Waters 468
UV detector) and by LCMS (Waters 2790 LC system coupled with
a Micromass Platform II mass spectrometer using Electrospray
ionization mode).
(72%) for the phenylalanine dipeptide was still good but consider-
ably lower. This observation may be explained in terms of a lower
hydrophobic contribution to the shift in equilibrium. Indeed, for
polar substrates (2f-i), the yields of 3f-i were also lower.
In conventional peptide synthesis, an excess of 5-10-fold of
amino acid is generally required for complete coupling. It is
interesting to note that excess of acyl donor required in our studies
compares favorably (4-fold excess needed for complete conversion
of 2j to 3j) with chemical methods. However, we expect that the
excess required will depend on the aqueous solubility of the Fmoc-
amino acid and will be higher for more polar amino acids. We are
currently working on a model that aims to quantify relative
contributions from suppressed ionization, substrate solvation, and
these substrate excess effects. With increased understanding, the
yields of some of these reactions may be improved further in the
future.
Acknowledgment. The authors gratefully acknowledge financial
support from the EC and the Wellcome Trust. We would also like
to thank Polymer Laboratories (U.K.) for the supply of PEGA1900
.
References
Entries 3f-i illustrate a major advantage of using enzymes for
peptide synthesis instead of conventional Fmoc chemistry in that
side chains of these amino acids do not need to be protected, even
when the side chains are significantly basic (3h) or acidic (3i).
A further advantage of using enzymes in synthetic applications
is their high enantioselectivity. Entry 3c shows that it was possible
to synthesize the L,L-diastereoisomer with high selectivity when a
DL-mixture of Fmoc-Phe (2c,2e) was presented as the acyl donor.
It is well established that certain proteases can be used to catalyze
peptide formation involving nonnatural amino acids. This is shown
here by using Fmoc-norleucine (2j) successfully as an acyl donor
to generate 3j in excellent yield.
(1) (a) Klibanov, A. M. Nature 2001, 409, 241. (b) Khmelnitsky, Y. L.; Rich,
Y. O. Curr. Opin. Chem. Biol. 1999, 3, 47. (c) Halling, P. J. Curr. Opin.
Chem. Biol. 2000, 4, 74. (d) Carrea, G.; Riva, S. Angew. Chem., Int. Ed.
2000, 39, 2226.
(
2) (a) Erbeldinger, M.; Ni, X.; Halling, P. J. Enzyme Microb. Technol. 1998,
2
3, 141. (b) Straathof, A. J. J.; Litjens, M. J. J.; Heijnen, J. J. In Methods
in Biotechnology; Holland, H. L., Ed.; Humana Press: Totowa, New
Jersey, 2001; p 603. (c) Ulijn, R. V.; Janssen, A. E. M.; Moore, B. D.;
Halling, P. J. Chem.-Eur. J. 2001, 7, 2089.
(
3) In addition, kinetically controlled methods have been developed that make
use of activated acyl substrates. In these reactions, the thermodynamic
equilibrium is not shifted, but temporary high product concentrations can
be obtained. Elegant methods have been developed to suppress the
hydrolytic reactions of product and acyl substrate in favor of synthesis.
See: (a) Sears, P.; Wong, C. H. Biotechnol. Prog. 1996, 12, 423. (b)
Fang, J. M.; Wong, C. H. Synlett 1994, 6, 393.
(4) (a) Smith, H. K.; Bradley, M. J. Comb. Chem. 1999, 1, 326. (b) Leon, S.;
Quarrell, R.; Lowe, G. Bioorg. Med. Chem. Lett. 1998, 8, 2997.
The reactions in Table 1 were all conducted for 14 h to ensure
equilibrium was reached. However, in subsequent studies we have
found that shorter reaction times (e.g., of 2 h for synthesis of 3c)
are sufficient for complete conversion. Preliminary kinetic studies
(
5) (a) Yamada, K.; Nishimura, S. Tetrahedron Lett. 1995, 36, 9493. (b)
Schuster, M.; Wang, P.; Paulson, J. C.; Wong, C. H. J. Am. Chem. Soc.
1994, 116, 1135.
(
6) (a) Grether, U.; Waldmann, H. Chem.-Eur. J. 2001, 7, 959. (b) Grether,
U.; Waldmann, H. Angew. Chem., Int. Ed. 2000, 39, 1629. (c) Bohm, G.;
Dowden, J.; Rice, D. C.; Burgess, I.; Pilard, J.; Guilbert, B.; Haxton, A.;
Hunter, R. C.; Turner, N. J.; Flitsch, S. L. Tetrahedron Lett. 1998, 39,
-
1
-1
suggest an initial rate of reaction at about 0.1 µmol min mg
,
which is around 1 order of magnitude lower than that of thermol-
ysin-catalyzed synthesis in aqueous solution. The difference can
be explained in terms of lower rates of diffusion when the substrates
are linked to solid supports.
In summary, we have reported the first examples of high-yielding
protease-catalyzed peptide synthesis on solid support. It is particu-
larly interesting to note that the reactions can be conducted in bulk
aqueous medium, with no need for organic cosolvent or activated
carboxylic acid. We are currently studying the scope of this reaction
for the synthesis of larger peptides.
3819.
(
7) (a) Buchardt, J.; Shiødt, C. M.; Krog-Jensen, C.; Delaiss e´ , J.; Foged, N.
T.; Meldal, M. J. Comb. Chem. 2000, 2, 624. (b) Rademann, J.; Grøtti,
M.; Meldal, M.; Bock, K. J. Am. Chem. Soc. 1999, 121, 5459.
(
8) It is well known that the favored hydrolysis of amides in dilute aqueous
solutions is largely due to the favorable ionization of amino acids. When
the equilibrium constant for peptide synthesis is expressed in terms of
the un-ionized forms of the reactants only, the value was shown to be
3
.6
-1
10
Janssen, A. E. M.; Halling, P. J. J. Chem. Soc., Perkin Trans. 2 2002, 5,
024-1028. Suppression of ionization leads to a shift of equilibrium
M , largely towards synthesis. See: Ulijn, R. V.; Moore, B. D.;
1
toward synthesis, as observed when organic cosolvents are used. For solid
supported amine substrates, ionization of amino groups is significantly
suppressed by the proximity of positive charges of neighboring amines.
An equilibrium shift towards synthesis is expected. An alternative picture
is that amide synthesis destroys positive amine charges on the resin and
reduces the unfavorable repulsion between them.
Solid-phase substrates were linked to PEGA1900 (Polymer
Laboratories) beads via a Wang-type linker (hydroxymethylphe-
noxyacetamide). PEGA1900-Phe was prepared using standard Fmoc
chemistry in DMF.
For the enzymatic reactions, 10 mg of PEGA1900 (100 mg of
wet weight stored in MeOH, loading 0.2 mmol/g dry polymer) was
washed with 2 × 5 mL of 0.1 M potassium phosphate buffer of
pH 7.5. Five milligrams of thermolysin (protease type X from
Sigma) was added to a suspension of the washed PEGA1900 resin,
(
9) Kitano, H.; Yasushi, M.; Masayo, Y.; Izumida, R. Macromol. Chem. Phys.
1996, 197, 4173-4181.
(
10) Another interpretation is that overall synthesis involves the transfer of
the hydrophobic acyl donor into the resin environment. As a result, there
will be a net loss in unfavorable hydrophobic hydration.
(11) (a) De Martin, L.; Ebert, C.; Gardossi, L.; Linda, P. Tetrahedron Lett.
2001, 42, 3395. (b) Ulijn, R. V.; Erbeldinger, M.; Halling, P. J. Biotechnol.
Bioeng. 2000, 69, 633. (c) Erbeldinger, M.; Mesiano, A. J.; Russell, A. J.
Biotechnol. Prog. 2000, 16, 1131. (d) Liu, P.; Tian, G.; Lee, K.; Wong,
M.; Ye, Y. Tetrahedron Lett. 2002, 43, 2423.
0.2 mmol protected amino acid, and 2 mL of 0.1 M potassium
JA026912D
J. AM. CHEM. SOC.
9
VOL. 124, NO. 37, 2002 10989