"Polar patch" proteases as glycopeptiligases

Article abstract of DOI:10.1039/b412030b

The strategy of combined site directed mutagenesis and chemical modification with polar prosthetic groups was used to broaden substrate specificity of proteases resulting in the first successful formation of glycopeptides through the use of glucoamino acid acyl donors in yields of up to 90%.

Full text of DOI:10.1039/b412030b

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
‘‘Polar patch’’ proteases as glycopeptiligases{
Katie J. Doores and Benjamin G. Davis*
Received (in Cambridge, UK) 5th August 2004, Accepted 24th September 2004
First published as an Advance Article on the web 3rd December 2004
DOI: 10.1039/b412030b
We have previously shown that a combination of site-directed
The strategy of combined site directed mutagenesis and
chemical modification with polar prosthetic groups was used
to broaden substrate specificity of proteases resulting in the first
successful formation of glycopeptides through the use of
glucoamino acid acyl donors in yields of up to 90%.
mutagenesis and chemical modification of serine protease subtilisin
Bacillus lentus (SBL) is a powerful technique for the efficient and
rapid creation of new active site environments in enzymes with a
broad substrate acceptance.7 Introduction of a cysteine at position
166 at the base of the primary specificity determining S1 pocket
followed by addition of a ‘‘polar patch’’ (polar modification) was
found to dramatically influence the specificity of SBL allowing
D-amino acid ligation by chemically modified mutant enzymes
(CMMs).
Glycoproteins exist in different glycoforms each with different
carbohydrate structures that can alter function.1 Current methods
for the isolation or creation of single glycoforms are difficult and
often only small quantities of proteins can be obtained.1 Therefore
new methods for glycopeptide synthesis are desirable.
To probe the potential of these ‘‘polar patch’’ and other
modified proteases in the direct ligation of glycoamino acids, we
synthesised novel glycoamino acid substrate 4 containing a
chromophoric moiety, which upon hydrolysis by the enzyme
would be released and detected by UV spectroscopy, thus allow-
ing the ready identification of this novel enzyme activity
through screening. Glucose was chosen as an archetypal sugar.
N-terminal Cbz (Z) protection was chosen on the basis of prior
ligation reactions with simple amino acids8 and this was
introduced to serine using benzyl chloroformate.9 The hydroxyl
group was protected with TBDMS using TBDMSCl and
imidazole in DMF. Reaction with para-nitroaniline (pNA) and
phosphorous oxychloride in pyridine afforded the fully pro-
tected amino acid in a pleasing 80% yield.10 The TBDMS group
was removed using TBAF in THF11 providing the pNA
amino acid 1. Glycosylation of the serine side chain is a known
synthetic challenge.12 Many conditions were explored for the
glycosylation of 1,13 with most resulting in low yields and multiple
by-products. The most efficient method used anomeric trichlor-
oacetimidate 2 as the glycosyl donor14 which was activated with
catalytic TMSOTf in dry DCM resulting in the glucosyl
amino acid 3 in 40% yield15 which, to the best of our knowledge,
is the first pNA glycoamino acid to be synthesised. Acetate
protection of the glycosyl donor allowed absolute stereocontrol
giving anomerically pure b-O-glycoside. The sugar–serine linkage
is sensitive to b-elimination in basic environments16 and this can
cause difficulties in the removal of base-cleavable protecting
Enzymes are now widely accepted as useful catalysts in organic
synthesis often with high stereo- and regiospecificity.2 Their use in
glycopeptide synthesis might therefore advantageously allow the
coupling of carbohydrate and amino acid derivatives with reduced
protecting group manipulation (Fig. 1).1 Enzymatic conditions are
also mild, so reactions can proceed without racemization of the
amino acid a centre or b-elimination of the sugar. They are often
carried out in water, a more desirable solvent for large peptide or
oligosaccharide fragments, which have limited solubility in organic
solvents. However, the use of enzymes may be limited by their
typically stringent structural and stereospecificity: naturally
occurring wild type (WT) enzymes cannot accept all the structures
of synthetic chemical interest.
Enzymatic coupling of glycoamino acids with other amino acids
has not yet been mastered. Wong et al.3 have successfully coupled
N- and O-linked glycoamino acid acyl donor fragments with
nucleophilic amino acids and peptide acyl acceptors in DMF or
aqueous environments using the serine protease subtilisin BPN9.
However, this reaction was restricted to xylose containing
glycoamino acids at the amide bond forming site (direct ligation).
For more complex glycoamino acids, at least one amino acid
spacer residue was required.4 This potentially powerful strategy
culminated in the enzyme catalysed ligation of glycopeptide
fragments to create a variant of RNase bearing sialyl Lewis x.5
The utility of peptide ligation has recently been emphasised further
by its use in a combined native ligation–enzymatic ligation
approach.6
Fig. 1 Enzymatic ligation as one strategy for glycopeptide synthesis.1
{ Electronic supplementary information (ESI) available: experimental
details. See http://www.rsc.org/suppdata/cc/b4/b412030b/
*ben.davis@chem.ox.ac.uk
Scheme 1 (i) TMSOTf, DCM, (3: 40%, 6: 46%), (ii) NH2NH2?H2O,
MeOH, 79%, (iii) NH2NH2?H2O, MeOH, 33%.
168 | Chem. Commun., 2005, 168–170
This journal is ß The Royal Society of Chemistry 2005

View Article Online
Fig. 2 Activities towards glycoamino acid 4 of a range of novel proteases created through the in situ chemical modification of S166C. Novel proteases
and their creation are shown in blue, corresponding modifications of the enzymes to alter activity shown in red and the reaction catalyzed and used as a
screen shown in black.
groups. However, after a survey of a range of methods,
deprotection was pleasingly achieved using hydrazine monohy-
drate in 79% methanol resulting in deprotected p-nitroanilide
glycoamino acid 4.
Table 1 Catalytic activity for the hydrolysis of glycoamino acid 4 by
selected modified proteases
Enzyme Est. kcat/KM/M21 s21 KM /M kcat/s21 kcat/KM/M21 s21
WT
3.62
6.79
7.39
14.6
33.2
0.0023 0.007
0.0036 0.008
0.0020 0.009
0.0014 0.0005
0.0054 0.008
3.04
2.16
4.47
0.39a
1.51a
Armed with this useful chromophoric probe for our desired
enzyme activity we surveyed various protease catalysts. Potential
structural variation through chemical modification in the active
site of SBL is so vast that a combinatorial approach to enzyme
screening and preparation was taken to identify the best CMMs
for this transformation. The method developed by Plettner et al.17
took advantage of the quantitative and rapid reaction of
methanethiosulfonate (MTS) reagents with free protein cysteine
residues (inset in Fig. 2).18 Modifications of cysteine mutant
SBL-S166C (Fig. 2) were carried out on a microscale in a 96-well
plate and were monitored using Ellman’s reagent.19 Estimations
for kcat/KM were then obtained directly from the rate of pNA
released from 4 (e414 5 2502 M21 cm21) using the limiting case of
the Michaelis–Menten equation at low substrate concentrations of
0.1 mM using these modified enzymes. The results are summarised
in Fig. 2.
S166C-g
S166C-e
S166C-a
S166C-c
a
Large difference from est. may be due to ‘low substrate
approxiamation’ breakdown.
Having identified the modifications which gave the most
desirable catalytic activity, peptide ligation reactions were
attempted with the glycoamino acid methyl ester substrate 7 as a
suitable acyl donor. The serine methyl ester was successfully
prepared using thionyl chloride and methanol.21 Glycosylation
using the same conditions as for 3 yielded 6, followed by hydrazine
deprotection to afford 7 (Scheme 1).
With glycinamide as an acyl acceptor, the majority of the
modifications showed an increase in efficiency compared to the
wild type, with S166C-e yielding a pleasing 90% glycopeptide.
Excitingly, using S166C-e, substrate specificity was also broadened
to allow b-alaninamide and c-butyricamide as acyl acceptors,
albeit in lower yields than for glycinamide (Table 2). These are the
first direct ligations of hexosyl amino acids.
Promising catalysts, S166C-a, -c, -e, -g, were selected for more
detailed evaluation on the basis of these results and previous
significant substrate specificity broadening by polar aromatics20
and charged polar patches.7 Larger scale modification reactions
in aqueous buffer were rapid and quantitative, as judged by
titration of free thiols with Ellman’s reagent.19 Mass spectrometry
was used to confirm CMM structure. The precise effects of the
modifications to SBL were assessed by full Michaelis–Menten
kinetics for the hydrolysis of 8 at pH 8.6 over a range of
concentrations (0.02–8.0 mM) and the results are summarised in
Table 1.
Application to potentially therapeutic targets was also investi-
gated. Compound 11 is an analogue of known active glycolipid
galactosyl–ceramide mimics that are able to inhibit HIV uptake
and infection of CD4-negative cells.22 The desired glycolipid
analogue 11 was formed under catalysis by S166C-e in 35% yield
(Scheme 3).
This journal is ß The Royal Society of Chemistry 2005
Chem. Commun., 2005, 168–170 | 169

View Article Online
Katie J. Doores and Benjamin G. Davis*
Department of Chemistry, University of Oxford, Chemistry Research
Laboratory, Mansfield Road, Oxford, UK OX1 3TA.
E-mail: ben.davis@chem.ox.ac.uk; Fax: + 44 (0)1865 285002;
Tel: + 44 (0)1865 275652
Notes and references
Scheme 2 (i) Enzyme (0.1 mol%), Et3N, DMF–H2O (1 : 1). Acyl donor :
1 B. G. Davis, Chem. Rev., 2002, 102, 579.
2 C.-H. Wong and G. M. Whitesides, Enzymes in Synthetic Organic
Chemistry, Pergamon, Oxford, 1994.
acceptor, 1 : 3.
3 C.-H. Wong, M. Schuster, P. Wang and P. Sears, J. Am. Chem. Soc.,
1993, 115, 5893.
4 K. Witte, O. Seitz and C.-H. Wong, J. Am. Chem. Soc., 1998, 120,
1979.
5 K. Witte, P. Sears, R. Martin and C.-H. Wong, J. Am. Chem. Soc.,
1997, 119, 2114.
Table 2 Results from enzyme catalysed ligation reactions according
to Scheme 2 with amino acid acyl acceptors
Yield (%)a
(5 days) S166C
Amino acid
n Product
WT -g
-e -c
6 Z. Machova, R. Eggelkraut-Gottanka, N. Wehofsky, F. Bordusa and
A. Beck-Sickinger, Angew. Chem., Int. Ed., 2003, 42, 4916.
7 K. Matsumato, B. G. Davis and J. Jones, Chem. Eur. J., 2002, 8,
4129.
8 K. Khumtaveeporn, G. DeSantis and J. B. Jones, Tetrahedron:
Asymmetry, 1999, 10, 2563.
Gly–NH2?HCl 1 Glc–Z–Ser–Gly–NH2 8
b-Ala–NH2?HCl 2 Glc–Z–Ser–b-Ala–NH2
60
22
73 90 20
—
—
b
b
b
9
53
36
—
—
b
c-Aba–NH2?HCl 3 Glc–Z–Ser–c-Aba–NH2 10 15
a
b
Yield based on compound 7. Ligation not attempted.
9 A. Scheurer, P. Mosset, W. Bauer and R. W. Saalfrank, Eur. J. Org.
Chem., 2001, 16, 3067.
10 D. Rijkers, H. Adams, H. C. Hemker and G. Tesser, Tetrahedron, 1995,
51, 11235.
11 Y. Bilokin, A. Melman, V. Niddam, B. Benhamu and M. Bachi,
Tetrahedron, 2000, 56, 3425.
12 C. Taylor, Tetrahedron, 1998, 54, 11317.
13 K. P. R. Kartha, L. Ballell, J. Bilke, M. McNeil and R. A. Field,
J. Chem. Soc., Perkin Trans. 1, 2001, 770.
14 R. R. Schmidt, Angew. Chem., Int. Ed. Engl., 1986, 25, 212.
15 G. Grundler and R. R. Schmidt, Liebigs Ann. Chem., 1984, 1826.
16 P. Sjolin and J. Kihlberg, Tetrahedron Lett., 2000, 41, 4435.
17 E. Plettner, K. Khumtaveeporn, X. Shang and J. Jones, Bioorg. Med.
Chem. Lett., 1998, 8, 2291.
Scheme 3 Synthesis of glycolipid analogue via direct enzymatic ligation.
In summary, we have shown that combined site-directed
mutagenesis and chemical modification can be used to broaden
the substrate specificity of SBL and was utilised in the first direct
ligations of hexosyl amino acids. A triamino substituent in the S1
pocket leads to the greatest substrate broadening and was used in
the synthesis of various glycopeptides and analogues of com-
pounds showing anti-HIV activity.
18 T. Bruice and G. Kenyon, J. Protein. Chem., 1982, 1, 47.
19 G. Ellman, K. Courtney, V. Andres and R. Featherstone, Biochem.
Pharmacol., 1961, 7, 88.
20 K. Khumtaveeporn, A. Ullmann, K. Matsumoto, B. G. Davis and
J. Jones, Tetrahedron: Asymmetry, 2001, 12, 249.
21 D.-R. Hwang, P. Helquist and M. Shekhani, J. Org. Chem., 1985, 50,
1265.
22 B. Faroux-Corlay, L. Clary, C. Gadras, D. Hammache, J. Greiner,
C. Santaella, A.-M. Aubertin, P. Vierling and J. Fantini, Carbohydr.
Res., 2000, 327, 223.
We thank the Leverhulme Trust for funding.
170 | Chem. Commun., 2005, 168–170
This journal is ß The Royal Society of Chemistry 2005