Communications
towards the carboxy but also the amino component.[6] This
this indicates that ADP tolerates even longer substrate
mimetics. Importantly, besides hydrolyzed peptide esters, no
further side products could be detected even after extended
reaction times of 24 h. Clearly, competing peptide bond
cleavages did not occur (see the Supporting Information).
Finally, we evaluated the function of ADP for the
synthesis of native proteins. As the synthesis target we
selected the peptidyl prolyl cis–trans isomerase (PPIase)
parvulin 10 from E. coli (Par10). The primary catalytic
function of PPIases is to facilitate cis–trans isomerizations
of Xaa-Pro bonds; they are involved in the folding of newly
synthesized proteins, in the control of the cell cycle control,
and in the immune system in higher developed organisms.[11]
According to Scheme 1, the synthetic route to Par10 follows a
general substrate-mimetic effect is, however, affected by the
individual amino acid of the OGp ester. Interestingly, with the
exception of the less bulky Gly and l-Ala moieties, this yield-
improving effect appears to be inversely related to the
enzymeꢀs primary specificity. Thus, the l-Glu-derived OGp
ester, the slowest substrate in the hydrolysis studies, gave the
highest yields in the synthesis reactions. Vice versa, the
readily hydrolyzed l-Phe ester shows the lowest synthesis
efficiency of all OGp esters tested. But even here, the yields
with specific acyl acceptors are in a synthetically useful range.
After proving the acceptance of short amino acid derived
reactants, we evaluated the scope of ADP for coupling
elongated peptides. For this purpose, reactions with selected
OGp esters covering 1, 3, and 10 amino acids and acyl
acceptors with 1, 10, and 16 amino acids were performed. To
exclude an inference by the nature of the amino acids at the
site of ligation, both the C-terminal amino acid of the acyl
donor (Gly) and the acyl acceptorsꢀ N-terminal residue (Leu)
were kept constant. The reactions, including all control
reactions, were performed under conditions similar to those
described for the initial dipeptide syntheses; however, for
synthesis-economy reasons, the concentration of the acyl
acceptors was reduced from 20 to 10 mm. While the control
reactions lacking ADP did not indicate any spontaneous
aminolysis or considerable hydrolysis of the acyl donor esters,
the enzyme-catalyzed syntheses were very rapid, resulting in
complete conversion of all esters to the desired peptide
products within 15 min (Table 2). As for the efficiency of the
reaction using Bz-Gly-OGp und H-l-Leu-NH2, the lower
yield can be explained by the reduced acyl acceptor concen-
tration. Considering this, the elongation of the acceptorꢀs
chain length leads only to a slight reduction in the product
yield. The reactions with the extended substrate mimetics
support this finding, indicating that ADP tolerates longer acyl
acceptor peptides without a significant loss of its synthesis
activity. In contrast, chain length of the acyl donor has a more
apparent effect on the product yields. Elongation of Bz-Gly-
OGp to Bz-Phe-Gly-Gly-OGp results in a decrease in the
yields of about 30% nearly regardless of the bulkiness of the
acceptor component. However, an additional elongation of
the acyl donor to Bz-AYLDAYVKAG-OGp does not lead to
further considerable reductions in the synthesis efficiencies;
Scheme 1. General course of the synthetic route to full-length E. coli
Par10. a,b) Resin loading and solid-phase peptide synthesis; c–
f) linker activation, peptide release, side-chain deprotection, and
purification; g,h) resin loading and solid-phase peptide synthesis; i–
k) simultaneous peptide release/deprotection and purification; l) ADP-
catalyzed ligation; m) N-terminal deprotection. Z-Par10(1–35)-OGp: Z-
AKTAAALHILVKEEKLA LDLLEQIKNGADFGKLAK-OGp; Par10(36–92):
KHSICPSGKRGGDLGEFRQGQMVPAFDKVVFSCPV LEPTGPLHTQFGY-
HIIKVLYRN. Conditions: 308C, 0.1m HEPES buffer pH 8.0, 20 vol%
DMF, 1 mm TCEP, [acyl donor]=0.2 mm, [acyl acceptor]=0.1 mm,
[enzyme]=2.010À5 m. TCEP=tris(2-carboxyethyl)phosphine.
single-step enzymatic coupling of an OGp ester of Na-
protected Par10(1–35) and the remaining 57-mer Par10(36–
92) fragment. The ligation site at Lys35–Lys36 was chosen for
purely chemical reasons (the efficiency of the solid-phase
peptide synthesis of the two fragments). CD-spectroscopic
studies reveal a that the structure of the longer Par10(36–92)
fragment is similar to that of the full-length protein with
corresponding a-helical and b-sheet regions (Figure 1). Thus,
the site of ligation can be assumed to be an a-helical motif,
which is known to be hardly recognized by proteases. Thus, in
the present synthesis Par10 is a difficult sequence for
protease-based ligations. The starting fragments were pre-
pared by standard solid-phase Fmoc chemistry using either 2-
chlorotrityl chloride resin[12] for the preparation of Par10(36–
92) or Kennerꢀs 4-sulfamylbenzoyl aminomethyl (AM)
safety-catch resin[13] for synthesizing Z-Par10(1–35)-OGp
(see the Supporting Information). Peptide release, deprotec-
tion, and purification led to the desired fragments which for
Table 2: Yields of the “alkaline d-peptidase” catalyzed ligation of
elongated all-l-peptide fragments.[a]
Acyl donor
Acyl acceptor
Yield [%]
Bz-Gly-OGp
H-l-Leu-NH2
86.1
77.5
83.2
54.0
45.6
49.0
46.9
40.1
43.1
H-LGSVKASAYK-OH
H-LIVDAVLEPVKAAGAY-OH
H-l-Leu-NH2
H-LGSVKASAYK-OH
H-LIVDAVLEPVKAAGAY-OH
H-l-Leu-NH2
Bz-Phe-Gly-Gly-OGp
Bz-AYLDAYVKAG-OGp
H-LGSVKASAYK-OH
H-LIVDAVLEPVKAAGAY-OH
[a] Conditions: 308C, 0.1m HEPES buffer pH 8.0, 10 vol% DMF, [acyl
donor]=2 mm, [acyl acceptor]=10 mm, [enzyme]=(5.5–8.7)10À6 m.
Reaction time: 15 min. Errors are less than 5% (Æ2.5%).
5458
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 5456 –5460